analysis research vector

• History & Society
• Science & Tech
• Biographies
• Animals & Nature
• Geography & Travel
• Arts & Culture
• Games & Quizzes
• On This Day
• One Good Fact
• New Articles
• Lifestyles & Social Issues
• Philosophy & Religion
• Politics, Law & Government
• World History
• Health & Medicine
• Browse Biographies
• Birds, Reptiles & Other Vertebrates
• Bugs, Mollusks & Other Invertebrates
• Environment
• Fossils & Geologic Time
• Entertainment & Pop Culture
• Sports & Recreation
• Visual Arts
• Demystified
• Image Galleries
• Infographics
• Top Questions
• Britannica Kids
• Saving Earth
• Space Next 50
• Student Center
• Introduction

Vector algebra

Products of vectors, coordinate systems, calculus of vectors.

• Where was science invented?
• When did science begin?

Explore CUIMC

Diversity, equity, and inclusion.

At CUIMC, we are committed to continuous improvement in providing culturally inclusive medical education and clinical care.

Columbia Vagelos College of Physicians and Surgeons is dedicated to developing the next generation of leaders in medicine

Patient Care

• Find a Doctor

Search for a provider by specialty, expertise, location and insurance. Schedule an appointment online.

Read the latest news stories about CUIMC faculty, research, and events

A Cellular Community in the Brain Drives Alzheimer’s Disease

Share this page.

• Share on Facebook
• Share on X (formerly Twitter)
• Share on LinkedIn
• Share by email

An analysis of more than 1.6 million brain cells from older adults has captured the cellular changes that occur in the early stages of Alzheimer’s disease, potentially revealing new routes for preventing the most common cause of dementia in older individuals.

The study also identified a second community of cells that drives the older brain down a different path that does not lead to Alzheimer’s disease.

“Our study highlights that Alzheimer’s is a disease of many cells and their interactions, not just a single type of dysfunctional cell,” says Columbia neurologist Philip De Jager , who led the study with Vilas Menon , assistant professor of neurological sciences at Columbia University Vagelos College of Physicians and Surgeons, and Naomi Habib of the Hebrew University of Jerusalem.

“We may need to modify cellular communities to preserve cognitive function, and our study reveals points along the sequence of events leading to Alzheimer’s where we may be able to intervene.”

Crunching data from 1.6 million brain cells

The study was a technical marvel, cleverly combining new molecular technologies, machine-learning techniques, and a large collection of brains donated by aging adults.

Though previous studies of brain samples from Alzheimer’s patients have provided insights into molecules involved in the disease, they have not revealed many details about where in the long sequence of events leading to Alzheimer’s those genes play a role and which cells are involved at each step of the process.

“Past studies have analyzed brain samples as a whole and they lose all cellular detail,” De Jager says. “We now have tools to look at the brain in finer resolution, at the level of individual cells. When we couple this with detailed information on the cognitive state of brain donors before death, we can reconstruct trajectories of brain aging from the earliest stages of the disease.”

The new analysis required over 400 brains, which were provided by the Religious Orders Study and the Memory & Aging Project based at Rush University in Chicago.

Within each brain, the researchers collected several thousand cells from a brain region impacted by Alzheimer’s and aging. Every cell was then run through a process—single-cell RNA sequencing—that gave a readout of the cell’s activity and which of its genes were active.

Data from all 1.6 million cells were then analyzed by algorithms and machine-learning techniques developed by Menon and Habib to identify the types of cells present in the sample and their interactions with other cells.

“These methods allowed us to gain new insights into potential sequences of molecular events that result in altered brain function and cognitive impairment,” Menon says. “This was only possible thanks to the large number of brain donors and cells the team was fortunate enough to generate data from.”

Aging vs. Alzheimer’s

Because the brains came from people at different points in the disease process, the researchers were able to solve a major challenge in Alzheimer’s research: identifying the sequence of changes in cells involved in Alzheimer’s and distinguishing these changes from those associated with normal brain aging.

“We propose that two different types of microglial cells—the immune cells of the brain—begin the process of amyloid and tau accumulation that define Alzheimer’s disease,” De Jager says.

Then after the pathology has accumulated, different cells called astrocytes play a key role in altering electrical connectivity in the brain that leads to cognitive impairment. The cells communicate with each other and bring in additional cell types that lead to a profound disruption in the way the human brain functions.

“These are exciting new insights that can guide innovative therapeutic development for Alzheimer’s and brain aging,” De Jager says.

“By understanding how individual cells contribute to the different stages of the disease, we will know the best approach with which to reduce the activity of the pathogenic cellular communities in each individual, returning brain cells to their healthy state,” De Jager says.

Additional information

Top image of astrocytes and neurons in a dish by Kevin Richetin, University of Lausanne, CC BY-NC-ND 2.0 .

Philip De Jager, MD, PhD, is the Weil-Granat Professor of Neurology at Columbia University Vagelos College of Physicians and Surgeons, where he is also director of the Center for Translational and Computational Neuroimmunology, director of the Multiple Sclerosis Center, and deputy director of the Taub Institute for Research on Alzheimer's Disease and the Aging Brain.

Vilas Menon, PhD, is assistant professor in the Department of Neurology and Taub Institute for Research on Alzheimer Disease and the Aging Brain at Columbia University Vagelos College of Physicians and Surgeons.

The paper, “ Cellular communities reveal trajectories of brain aging and Alzheimer’s disease ,” was published Aug. 28 in Nature.

All authors: Gilad Sahar Green (Hebrew University of Jerusalem), Masashi Fujita (Columbia), Hyun-Sik Yang (Harvard), Mariko Taga (Columbia), Anael Cain (Hebrew University of Jerusalem), Cristin McCabe (Broad Institute), Natacha Comandante-Lou (Columbia), Charles C. White (Broad Institute), Anna K. Schmidtner (Hebrew University of Jerusalem), Lu Zeng (Columbia), Alina Sigalov (Columbia), Yangling Wang (Rush University Medical Center), Aviv Regev (Genentech), Hans-Ulrich Klein (Columbia), Vilas Menon (Columbia), David A. Bennett (Rush University Medical Center), Naomi Habib, (Hebrew University of Jerusalem), and Philip L. De Jager (Columbia).

The work was supported by the NIH (grants RF1AG057473, U01AG061356, U01AG046152, R01AG070438, R01AG015819, U01AG072572, R01AG066831, K23AG062750, K23AG062750, and R01AG080667), the Chan Zuckerberg Initiative (CS-02018-191971), the Israel Science Foundation (1709/19), the European Research Council (853409), Chinese Ministry of Science and Technology (3-15687), the Myers Foundation, Alzheimer’s Association (ADSF-21-816675), and a Minerva Fellowship of the Minerva Stiftung Gesellschaft fuer die Forschung mbH.

The Columbia authors declare no competing interests.

Thank you for visiting nature.com. You are using a browser version with limited support for CSS. To obtain the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles and JavaScript.

• View all journals
• Explore content
• About the journal
• Publish with us
• Sign up for alerts
• Open access
• Published: 28 August 2024

Predicting ground vibration during rock blasting using relevance vector machine improved with dual kernels and metaheuristic algorithms

• Yewuhalashet Fissha 1 , 2 ,
• Jitendra Khatti 3 ,
• Hajime Ikeda 4 ,
• Kamaldeep Singh Grover 3 ,
• Narihiro Owada 1 ,
• Hisatoshi Toriya 1 ,
• Tsuyoshi Adachi 1 &
• Youhei Kawamura 5  

Scientific Reports volume  14 , Article number:  20026 ( 2024 ) Cite this article

Metrics details

• Engineering
• Mathematics and computing

The ground vibration caused by rock blasting is an extremely hazardous outcome of the blasting operation. Blasting activity has detrimental effects on both the ecology and the human population living in proximity to the area. Evaluating the magnitude of blasting vibrations requires careful evaluation of the peak particle velocity (PPV) as a fundamental and essential parameter for quantifying vibration velocity. Therefore, this study employs models using the relevance vector machine (RVM) approach for predicting the PPV resulting from quarry blasting. This investigation utilized the conventional and optimized RVM models for the first time in ground vibration prediction. This work compares thirty-three RVM models to choose the most efficient performance model. The following conclusions have been mapped from the outcomes of the several analyses. The performance evaluation of each RVM model demonstrates each model achieved a performance of more than 0.85 during the testing phase, there was a strong correlation observed between the actual ground vibrations and the predicted ones. The analysis of performance metrics (RMSE = 21.2999 mm/s, 16.2272 mm/s, R = 0.9175, PI = 1.59, IOA = 0.8239, IOS = 0.2541), score analysis (= 93), REC curve (= 6.85E−03, close to the actual, i.e., 0), curve fitting (= 1.05 close to best fit, i.e., 1), AD test (= 11.607 close to the actual, i.e., 9.790), Wilcoxon test (= 95%), Uncertainty analysis (WCB = 0.0134), and computational cost (= 0.0180) demonstrate that PSO_DRVM model MD29 outperformed better than other RVM models in the testing phase. This study will help mining and civil engineers and blasting experts to select the best kernel function and its hyperparameters in estimating ground vibration during rock blasting project. In the context of the mining and civil industry, the application of this study offers significant potential for enhancing safety protocols and optimizing operational efficiency.

Similar content being viewed by others

A Novel Hybrid Model for Predicting Blast-Induced Ground Vibration Based on k-Nearest Neighbors and Particle Swarm Optimization

Assessment of the ground vibration during blasting in mining projects using different computational approaches

A combination of fuzzy Delphi method and hybrid ANN-based systems to forecast ground vibration resulting from blasting

Introduction.

Rock blasting is a frequently used and economically efficient method in the context of mining and civil engineering activities. Blasting is a widely employed technique in mining for obtaining metal and non-metal resources, such as in hard rock mining excavation and quarrying. During quarry operations, the process of blasting entails the systematic drilling of many rows of blast holes, each following to certain requirements for stemming, spacing, burden, sub-drill, face angle, bench height, and hole diameter 1 , 2 .

Different researchers like Fissha et al. 3 , underlines approximately 20–30% of the energy generated by the explosion is efficiently used to fragment the rock during blasting. The remaining energy is wasted in various ways, including back break, blasting vibration, fly rock, and air overpressure. The events mentioned have diverse environmental impacts and provide challenges for those living near the area where blasting occurs (Fig.  1 ). Among these impacts, ground vibration is particularly significant 4 , 5 , 6 . This phenomenon has the potential to inflict harm on physical constructions and eventually impact the lives and belongings of individuals, especially in cases when the buildings and structures were not specifically engineered to withstand the immense devastation caused by the explosion 7 , 8 . Explosive-induced ground vibrations have an impact on the development of plants and may result in the loss of forested areas. It also causes ground and slope instabilities, posing a risk to the safety of persons involved in loading, drilling, and subsequent blasting activities. In addition, individuals residing or working in close vicinity to the explosion site may experience physical discomfort as well as stress because of ground vibrations.

The favourable and unfavourable outcomes of rock blasting (Source: 9 ).

Based on Hosseini et al. 10 , ground vibration is caused by the energy released during an explosion. The magnitude of the vibration is influenced by several variables, such as the quantity of explosives used, the composition of the rock being exploded, and the distance from the blast source. Lawal et al. 11 , state that the strength of the ground vibration caused by explosions is linked to both the factors that can be controlled and those that cannot be controlled during the blasting process. The adjustable parameters for blasting, such as burden, spacing, blast hole depth, hole diameter, stemming type and height, maximum charge weight per delay (W), specific charge, explosive type, detonation velocity (VoD), and powder factor, can be easily adjusted and are meticulously planned according to the current conditions. Hence, it is incumbent upon the blasting engineer to calibrate and formulate these components while devising the blasting methodology.

The uncontrolled variables of the rock include the mechanical and physical properties of the rock, as well as the geological features of the surrounding environment; most uncontrolled factors are contingent upon the rock's inherent creation 12 , 13 , 14 .

Mostly the ground vibration movement has a wave-like pattern that propagates in a radial direction away from the blasting source 11 , 15 . Figure  2 shows the movement of waves during blasting vibration. This wave motion has similarity to the circular ripples that propagate outward when an object hits into a body of water and comes into contact. The PPV is the main parameter utilised to assess the magnitude of ground vibration resulting from blasting activities. PPV represents the velocity of the primary particles in terms of their transverse (T), vertical (V), and longitudinal (L) velocity.

Propagation of ground vibration wave during rock blasting (Source: 16 ).

Several scientists have established empirical formulae to anticipate the amplitude of blast-induced ground vibration coming from blasting operations. One of the earliest and most important equations to estimate PPV was developed by the USBM, Duvall, and Fogleson 17 , 18 , 19 . After many years, numerous academics produced adjustments to the USBM formula, considering the scaled distance and MC. The mathematical formulation of these techniques is found in Table 1 . However, the precision of the predictive model, the assessment of the complex nature of the rock mass conditions and input data parameters, and other factors contribute to the increased difficulty and time required for predicting and estimating blasting vibration. Hence, empirical models are inadequate due to the constraints imposed by their empirical formulations.

With the advent of digitalization, numerous researchers have begun employing artificial intelligence techniques such as machine learning, and deep learning methodologies to forecast blast-induced ground vibrations. The study by Hosseini et al. 10 , aims to analyse advanced deep learning, machine learning, and hybrid learning methods to provide an appropriate computational model for quantifying ground vibrations in mining blasting activities. The ANN, LSTM, ET, LSSVM, DT, SVM, GPR, and MLR models are utilised with a dataset consisting of 162 data points. The new black hole-optimized LSTM model has been utilised for the initial time to predict ground vibrations resulting from blasting. 15 performance metrics have been devised to precisely evaluate the predicting capabilities of different ML models. Table 2 presents a brief overview of the most recent research on utilising machine learning to predict ground vibrations.

Novelty of the research

The present investigation has the following novelty.

The present study analyses and employs the capabilities of RVM models in assessing and predicting ground vibration for the first time.

This study compares the conventional RVM models configured by Laplacian, linear, Gaussian, exponential, sigmoid, and polynomial kernels. In addition, these models have been enhanced by the application of both genetic and particle swarm optimisation algorithms. Also, the comparison of conventional, genetic, and particle swarm-optimized single kernel-based RVM models reveals the impact of optimization on the accuracy and performance of RVM models.

This investigation employs dual kernel (k1 + k2) RVM models (k1 kernel is selected from the comparison of non-optimized single kernel RVM models). Thus, this research employs five dual kernels based RVM models and optimizes them using each genetic and particle swarm optimization algorithm.

Fifteen metrics, WI, LMI, RSR, MBE, NMBE, NS, BF, PI, VAF, MAPE, R, MAE, WMAPE, RMSE, IOS, IOA, and a20, evaluate the performance and precision of each model and showcase the dependability of the models.

Research significance

This study addresses the significant issue of blast-induced ground vibrations in sectors like mining and construction. Conventional models see (Table 1 ) often fall short due to oversimplified approaches. By employing relevance vector machines (RVM), this research offers a novel method to precisely analyse vibrations, considering complex relationships and uncertainties. This not only advances academic understanding but also provides practical insights for risk assessment and mitigation. The flexibility of the proposed model makes it valuable for various blasting scenarios, benefiting engineers, policymakers, and researchers in addressing environmental and structural impacts.

The rest of the paper is organized as follows: the research methodology of this study is provided in “ Research methodology ”, and details regarding the datasets and preprocessing, data analysis, and developing soft computing approaches are included in “ Data analysis and soft computing approaches ”. Section “ Results and discussion ” outlines the results and discussions of the RVM model predictions and the results of the different statistical analyses. The key conclusions and remarks from the research and their implications are presented in “ Conclusion and summary ”.

Research methodology

This study presents an optimised performance model for accurately predicting ground vibration caused by blasting using several RVM models. Two hundred datasets have been gathered from published work by Hammed et al. 41 to test, train, and validate the RVM models. The data is collected from ten quarry sites in Nigeria. The dataset consists of different variables such as charge weight ( W ) in Kg, monitoring distance ( D ) in meters, scaled distance ( SD ) in m/kg½, and peak particle velocity ( PPV ) in mm/s. In this study, the PPV is the target output of the investigation hence we consider one vector sum PPV only because the data set from the literature consists of three different PPVs based on their vectorial direction such as longitudinal, vertical, and transverse. PPV by default it is a vector quantity so, it follows vector sum. The data set of this study is summarised in Table 3 , as follows.

Using the data proportionality method, the testing, validation, and training databases have been created by selecting 20, 20, and 160 data points arbitrarily. In this study, we developed thirty-three RVM models. The conventional single kernel based SRVM models were developed using linear, polynomial, Gaussian, sigmoid, exponential, and Laplacian kernel functions. Therefore, these SRVM models were optimized by each GA and PSO algorithm. Conversely, to determine the kernel 1 (k1), the effectiveness of conventional SRVM models was evaluated. After finding k1, the different combinations of kernels were prepared. In this research, the Gaussian kernel based conventional SRVM model achieved higher performance. Hence, the following combinations were prepared to develop the dual kernel based DRVM model: (1) Gaussian + linear, (2) Gaussian + exponential, (3) Gaussian + sigmoid, (4) Gaussian + Laplacian, and (5) Gaussian + polynomial. Thus, five conventional DRVM models were developed, similarly the five DRVM models were optimized each by PSO and GA algorithms are compared and analysed to introduce an optimal performance model for predicting the ground vibration during blasting. To compute the performance of each model, the , LMI, WI, NMBE, RSR, BF, PI, NS, MBE, WMAPE, VAF, MAPE, RMSE, R, MAE, IOS, IOA, and a20 statistical metrics were implemented. These statistical metrics have been compared for each case, i.e., SRVM, SRVM_GA, SRVM_PSO, DRVM, DRVM_GA, and DRVM_PSO , and determined six better-performing models. In addition, the score analysis has been performed and compared for each model. Furthermore, the predictive abilities of the six top-performing models have been examined by visualising the regression error characteristics (REC) curve. Uncertainty analysis, Wilcoxon test, and Anderson–Darling (AD) test were conducted on the better-performing models to determine the ideal performance model for predicting ground vibration in blasting. Additionally, the optimum performance model has undergone calculations for curve fitting, generalizability, and internal validation. After conducting several analysis and experiments, this study presents an optimum RVM model for precisely predicting ground vibration in rock blasting. Figure  3 illustrates the comprehensive flow of the research approach used in this study.

Illustration of flow chart of the main research methodology.

Software support

MATLAB R2020a: for developing soft computing models, evaluation, analysis, and prediction. Origin Lab 2022b: for graphical presentations and data analysis, and Minitab Statistical Software: for statistical analysis.

Data analysis and soft computing approaches

Data analysis.

The authors investigated a dataset collected from Hammed et al. 41 , to develop accurate predictive models for the PPV. The dataset consisted of 200 test results for the PPV induced from the quarry blasting in Nigeria. The data analysis section is done using Origin Pro 2024 software. Table 4 reveals the summary of the database's descriptive statistics. It also contained many possible input factors that were believed to have an impact on the PPV. Figure  4 provides further clarification on the distribution of each variable using a frequency histogram plot in navy blue (left) and ridgeline plots in orange (right). This study utilizes both frequency histograms and ridgeline plots to verify the normal distribution of the dataset for each variable. Based on this analysis, each variable shows a normal distribution. The visual representation of the data in this figure is advantageous for identifying any possible outliers or trends that might impact the accuracy of the models. Histograms provide a visual representation of the distribution of data points within a variable. By examining the shape, spread, and skewness of the histogram, we can identify unusual observations that deviate significantly from the overall pattern of the data. Specifically, outliers can often be observed as isolated bars or gaps in the histogram that are distant from the main cluster of data.

Graphic representation depicting the distribution of variables based on frequency histogram plot in blue navy colour (left) and ridgeline plots in orange colour (right).

Figure  5 demonstrates the relationship between variables in terms of the scatter matrix plot. It illustrates that all variables very strongly correlate with each other. It shows a strong correlation between D and SD with (r = 0.94).

Graphic representation of scatter matrix plot of all the variables to elaborate the relationship between the variables.

The violin plots of the variables are depicted in Fig.  6 . It is useful for visualizing the distribution of the blasting dataset of this study, especially it describes in terms of quartiles. Quartile 1 (Q1) represents the 25th percentile, and Quartile 3 (Q3) represents the 75th percentile. The box-in-a-box plot spans from Q1 to Q3. Based on this analysis the data set in the distance shows distributed equally. Hence this is due to the monitoring station distance taking place at 50-m intervals starting from the first point up to the last 20 monitoring points with 1250 m. For SD it shows the data points are highly scattered in Q 1 and Q 3, at a range of 7 m/kg 1/2 –35 m/kg 1/2 . Similarly, the CPD data set is highly distributed in a range of 750–2250 kg.

Depiction of violin plots of all variables.

Pearson and Spearman correlation coefficients are the most popular correlation coefficients in statistics, both ranging from − 1 to 1, and are frequently used to evaluate the relationship between two variables A correlation coefficient of 0 indicates no connection between variables. A positive correlation coefficient suggests a positive relationship, while a negative coefficient implies a negative relationship. The strength of the correlation is gauged by the absolute value of the coefficient; a higher absolute number indicates a stronger correlation.

In this current study, we introduce Spearman correlation technique (Fig.  7 ) to capture both linear and nonlinear correlation relationships among the variables. Spearman quantifies how much information about one variable can be gained by observing another. Both methods handle categorical and continuous variables. Figure  7 confirms a strong positive correlation between SD and distance with a correlation value of 0.95, and a correlation value of 0.047 between CPD and PPV, and 0.065 between distance and CPD.

Illustration of spearman correlation plot.

During the data analysis stage, researchers use the probability (PP) plot to visualize dataset dispersion. The PP plot compares real data quantiles to predictions of a theoretical distribution, revealing the data's distribution. This graph helps determine whether data follows a probability distribution like the normal distribution. Variations from a linear trajectory in the PP plot may indicate variations from the anticipated distribution, highlighting dataset issues. For this current study, the PP plot investigates each variable (Fig.  8 ). Based on the PP plot result variables such as CPD , and SD show almost good distribution in terms of the normal pattern of the data, however, the PP plot for PPV and distance shows there is no fully perfect normal distribution among the data points, some of the data points are distributed outside of the normal line. Specifically, the normal probability plot for the target variable of this study PPV indicates a mean ( μ ) of 64.17 and a standard deviation ( σ ) of 48.75 it suggests that the data is being compared or assessed in the context of a normal distribution with these specified parameters. Essentially, the specified μ and σ in the normal probability plot provide a reference for assessing the normality of the dataset under consideration.

Illustration of normal probability (PP) plot of all variables.

In addition to the above-advanced data analysis techniques, this study also integrates a 2D mean line graph to assess the visual depiction of data points on a coordinate system. Every point on the plot corresponds to a set of values for two variables in the blasting dataset. Examining a two-dimensional scatter plot may provide several significant observations, including the identification of patterns, detection of anomalies, and assessment of data distribution . Based on the 2D mean line graph in Fig.  9 , shows that there is almost similar data distribution among CPD and PPV. However, there is no harmonize data distribution between PPV, SD, and distance.

Depiction of 2D mean line graph plot of each input variable (y-axis) with the target variable (x-axis) to illustrate the interaction between each variable.

Soft computing approaches

This current study introduces 33 RVM models. The conventional single kernel based SRVM models were developed using polynomial, linear, sigmoid, Gaussian, exponential, and Laplacian kernel functions. Therefore, these SRVM models were optimized by each GA and PSO algorithms. Conversely, the performance of conventional SRVM models was compared to find the kernel 1 (k1). After finding k1, the different combinations of kernels were prepared. In this research, the Gaussian kernel based conventional SRVM model achieved higher performance. Hence, the following combinations were prepared to develop the dual kernel based DRVM model: (1) Gaussian + exponential, (2) Gaussian + linear, (3) Gaussian + Laplacian, (4) Gaussian + sigmoid, and (5) Gaussian + polynomial. Thus, five conventional DRVM models were developed. These five DRVM models were optimized by each PSO and GA algorithm.

Relevance vector machine (RVM)

RVM belongs to the category of kernel methods in machine learning, tailored for both regression and classification tasks. Kernel SVM, a variation of Support Vector Machine (SVM), was introduced to handle uncertainties in training data by integrating a probabilistic approach. RVM aims to address certain drawbacks present in traditional SVMs, such as difficulties in handling vast datasets and the necessity of selecting appropriate kernel functions and regularization parameters 42 .

Khatti et al. 43 , in their research, describe the key features of RVM, based on their analysis the main features are probabilistic framework, sparsity, kernel trick, sparse Bayesian learning, and model training. It means it is very similar to a Gaussian model with kernel (Eq.  1 )

where \(\varphi \) is the Gaussian kernel, \({\propto }_{m}\) are the variances of the prior on the weight vector \(\omega \) N (0, \({\alpha }^{-1}I\) ), and \({j}_{1}\) ,… \({j}_{N}\) are the input variables of the training datasets 43 , 44 . The Gaussian

kernels have developed the RVM models in the present study (Eq.  2 ). The \({\text{x}}_{\text{i}}\) , \({\text{x}}_{\text{j}}\) are the basic input and out variables, c is constant, m is the gradient/slope of a line, d is the degree, D is the scale factor, 1 is the length scale hyper-parameter, and \(\upsigma \) is the standard deviation.

Therefore, for the first time this study introduced RVM models for predicting the ground vibration (PPV), a total of six RVM models were implemented for predicting the PPV. These are 6 single kernel-based RVM conventional models (SRVM), 5 dual kernel-based RVM conventional models (DRVM), 6 GA-optimized single kernel-based RVM models (GA_SRVM), 6 PSO-optimized single kernel-based RVM models (PSO_SRVM), 5 GA-optimized Dual kernel-based RVM models (GA_DRVM), and 5 PSO-optimized Dual kernel-based RVM models (PSO_DRVM).

Two types of kernel function based RVM models have been developed for PPV prediction: single (SRVM) and dual (DRVM) models. Table 5 presents the RVM model configurations that were developed. This research has created 33 RVM models by using the hyperparameters specified in Table 5 .

Sensitivity analysis (SA)

The study employs sensitivity analysis (SA) to pinpoint the most influential independent variables in regression or numerical prediction models. Eliminating insignificant input parameters is crucial for improving model performance. In this research, the cosine amplitude technique (CAM) is utilized for SA. Figure  10 shows that charge per delay has the highest influence score of 0.772, followed by distance at 0.579 and scaled distance at 0.577. In Eq.  3 , the vector x (of length m) represents the predictors x i in the data array is specified as follows.

Depiction of cosine amplitude sensitivity analysis (SA) for input variables.

In Eq. ( 4 ), the significance between predictors ( x i ) and targets ( x j ) is estimated 45 .

In this study correlation plots and sensitivity analysis serve different purposes. Correlation plots visualize the relationship between variables, showing the strength and direction of linear or non-linear relationships through correlation coefficients. They are useful for identifying patterns within a dataset. For this target this study introduces different correlation plots such as Pearson and spearman correlation plots (see Figs.  5 and 7 ). Sensitivity analysis, on the other hand, assesses how variations in input variables affect the output of a model. It quantifies the impact of each variable, aiding in model validation and decision-making by identifying critical variables. In our current study we introduced cosine amplitude sensitivity analysis based on this CPD shows higher influence on the output with 0.77.

Performance evaluation

The current study has used the following eleven basic metrics to assess the performance of models: (1) Root means square error ( RMSE ), (2) Mean absolute percentage error ( MAPE ), (3) Mean absolute error ( MAE ), (4) Weighted Mean Absolute Percentage Error ( WMAPE ), (5) Correlation coefficient ( R ), (6) Variance account for ( VAF ), (7) Nash–Sutcliffe Efficiency (NS), (8) Ratio of Standard Deviation (RSR), and (9) the Bias factor (BF), (10) Performance index (PI), (11) Index of scatter (IOS), (12) Index of agreement (IOA), (13) Normalized mean bias error (NMBE), (14) Legate and McCabe’s index (LMI), (15) a20-index (a20), and (16) Mean bias error (NMBE) 42 . All the performance evaluation equations are included starting from Eqs. ( 5 ) to ( 20 ).

The variables α and ω represent the actual and predicted i -th values respectively. The variable n represents the total number of data points. β represents the mean of the actual values, while \(\overline{{\varvec{\upomega}} }\) represents the mean of the predicted values. The variable k represents the number of independent variables. The variable m20 represents the ratio of experimental to predicted values, which can vary between 0.8 and 1.2. H represents the total number of data samples.

The main benefit of the a20-index is its ability to predict the values within a deviation range of ± 20% in comparison to the actual measured values. Conversely, the index of agreement is limited to a range of − 1.0 to 1.0n 46 . A perfect predictive model always has performance indicators value equal to the ideal value, as given in Table 6 .

Results and discussion

In this study total six models are selected from the thirty-three total models i.e. MD1, MD10, MD16, MD21, MD26, and MD29 based on their different kernel configurations. The conventional single kernel based SRVM models were developed using linear, polynomial, Gaussian, sigmoid, exponential, and Laplacian kernel functions. Therefore, these SRVM models were optimized by each GA and PSO algorithm. Conversely, the performance of conventional SRVM models was compared to find the kernel 1 (k1). After finding k1, the different combinations of kernels were prepared. In this research, the Gaussian kernel based conventional SRVM model achieved higher performance. Hence, the following combinations were prepared to develop the dual kernel based DRVM model: (1) Gaussian + exponential, (2) Gaussian + linear, (3) Gaussian + Laplacian, (4) Gaussian + sigmoid, and (5) Gaussian + polynomial. Thus, five conventional DRVM models were developed. These five DRVM models were optimized by each GA and PSO algorithm. 6 single kernel-based RVM conventional models (SRVM), 5 dual kernel-based RVM conventional models (DRVM), 6 GA-optimized single kernel-based RVM models (GA_SRVM), 6 PSO-optimized single kernel-based RVM models (PSO_SRVM), 5 GA-optimized Dual kernel-based RVM models (GA_DRVM), and 5 PSO-optimized Dual kernel-based RVM models (PSO_DRVM) approaches have been employed, performed, and analysed. The seventeen-evaluation metrics have measured the training (TR) and testing (TS) performances of all models, as summarized in Table 7 . Table 7 demonstrates that the PSO-optimized Dual kernel-based RVM models (PSO_DRVM) model MD29 has higher performance (TS = 0.9175, TR = 0.9114) than other RVM models. It has been measured that model MD29 predicts the ground vibration with RMSE = 21.299 mm/s, MAE = 16.2272 mm/s, MAPE = 0.2094 mm/s, WMAPE = 0.1936 mm/s, and NMBE = 5.4120, in the TS phase. The comparison shows that model MD29 predicts ground vibration with the least residuals in the TR and TS phases. In similar style the results obtained from the soft computing models in the training, validation and testing phase include the different models and seventeen metrics are shown in Table 8 .

The overall comparison reveals that MD29 (DRVM_PSO with Gaussian + Exponential kernel) is the optimum performance model for assessing the ground vibration during blasting. A statistical relationship is drawn between actual and predicted PPV using models MD1, MD10, MD16, MD21, MD26, and MD29, as depicted in Fig.  11 .

Illustration of the relationship between actual and predicted PPV using RVM models ( a ) MD1, ( b ) MD10, ( c ) MD16, ( d ) MD21, ( e ) MD26, and ( f ) MD29.

Score analysis

The evaluation of computational models' effectiveness is conducted through score analysis, employing statistical methods. Each model is assigned a score, denoted as 'n', indicating its ability to accurately determine optimal values for performance indicators. This study utilized a sample size of 6 and focused on soft computing models. The highest and lowest values of performance indicators in the score analysis represent the best and worst training and testing instances for the models. The overall model score is determined by averaging the scores of performance indicators across both training and testing phases.

The testing and training results are used to get the overall score of a model. The score analysis findings for the training and testing performances of the computational models MD1, MD10, MD16, MD21, MD26, and MD29 are summarized and shown in Fig.  12 .

Total score of better performing models.

Figure  12 illustrate that MD29 exhibits superior performance in both training and testing phases. Consequently, MD29 emerges as the most effective model for assessing ground vibration PPV resulting from blasting activities. In contrast, MD21 is identified as the poor performing model in this study, given its lowest score.

Regression error characteristics (REC) curve

The REC curve is a visual tool that effectively illustrates the distribution of prediction errors, offering valuable insights into the performance of regression models. The REC curve differs from conventional metrics like MSE or MAE by emphasizing the cumulative distribution function of the absolute errors. The function graphs the ratio of occurrences with errors that are less or equal to a specified threshold, versus that threshold. REC curves extend the concept of ROC curves to regression problems. The y-axis of REC curves displays the proportion of projected points that are inside the specified error tolerance, while the x-axis represents the error tolerance. The resultant curve offers an approximation of the cumulative distribution function of the error. The current study included the creation of REC curves for testing, training, and validation of several RVM models.

These curves are shown in Fig.  13 , along with the corresponding AOC values listed in Table 9 .

Depiction of AOC results for ( a ) training and ( b ) testing, ( c ) validation phase.

Table 8 indicates that model MD29 achieved the lowest AOC values. Specifically, it obtained an AOC of 6.84E−03 during training, 6.85E−03 during testing, and 5.20E−03 during validation. These values are like the AOC of the real ground vibration. Therefore, model MD29 is a model that delivers optimal performance.

Curve fitting

In this study, six RVM models, i.e. MD1, MD10, MD16, MD21, MD26, and MD29, have been developed. MD29 is recognized as the better-performing model in predicting the PPV. Models MD1, MD10, MD16, MD21, MD26, and MD29 have been trained by 70%, 80%, 70%, 50%, 70%, 80%, 70%, 60%, 50%, and 100% training databases, respectively.

Based on Fig.  14 , the overfitting comparison shows that model MD29 performs well with the lowest overfitting of 1.05 in the testing phase and 0.92 in the validation phase.

Illustration of curve fitting of the six RVM models, MD1, MD10, MD16, MD21, MD26, and MD29.

Illustrations of Taylor plots for (a) testing, (b) training, and (c) validation phase have been found in Fig.  15 . Taylor plots offer invaluable advantages for predicting blasting vibration. They provide a clear visual representation of the relationship between predicted and observed values, facilitating easy interpretation and assessment of model performance. Through quantitative measures like correlation coefficient and standard deviation ratio, Taylor plots enable analysts to gauge the accuracy and precision of predictive models. Moreover, they pinpoint systematic bias and variability, aiding in model refinement and improvement. Taylor plots also allow for the comparison of multiple models on the same graph, facilitating informed decisions about model selection. Ultimately, they serve as a diagnostic tool for enhancing predictive accuracy and reliability, making them indispensable in the field of blasting vibration prediction.

Illustration of Taylor plots for ( a ) training, ( b ) testing, and ( c ) validation phase.

Anderson–Darling (AD) test

The Anderson–Darling (AD) test is a statistical tool used to assess whether a given dataset conforms to a particular probability distribution, particularly focusing on extreme values. It is commonly employed to evaluate normality, comparing the observed data to a theoretical distribution, often the normal (Gaussian) distribution. The research hypothesis (HR) suggests that the data deviate from the specified distribution, while the null hypothesis (H0) contends that the data are drawn from that distribution. A lower AD test statistic indicates a stronger alignment with the specified distribution. In our study, we applied the AD test to evaluate the fit of actual data and predictions from several models (MD1, MD10, MD16, MD21, MD26, and MD29), as depicted in Fig.  16 . Notably, model MD29 exhibited an AD value of 11.607 for predicting the ground vibration PPV closely matching the actual PPV dataset. The AD test results support the hypothesis of a normal distribution and suggest that MD29 performs best among the models examined.

Depiction of AD test results for models MD1, MD10, MD16, MD21, MD26, and MD29.

Wilcoxon test

The Wilcoxon test includes two variations: the rank sum test and the signed-rank test. These tests are used to assess and compare two groups that have been effectively matched. The Wilcoxon test is used to evaluate the presence of a statistically significant difference between two or more sets of paired data. The Wilcoxon test was used on the RVM models to predict the PPV of the blasting in both the testing, and training phases. Table 10 presents the results of the Wilcoxon test. The table indicates that throughout the training phase, model MD29 made accurate predictions of the PPV, with a different confidence interval (CI). This confidence range is comparable to the confidence interval of the true PPV of the blasting in the testing database. In the training phase, model MD29 assessed the PPV with a confidence interval of 0.2240 (upper bound) and 0.1596 (lower bound). The confidence interval (CI) of the actual test database is quite like this CI, with an upper level of 0.4140 and a lower level of 0.1333. The findings indicate that model MD29 has superior performance compared to the MD1, MD10, MD16, MD21, and MD26.

Uncertainty analysis (UA)

Assessing the reliability of soft computing models is crucial for accurately predicting results, particularly in scenarios like projecting the PPV resulting from quarry blasting. This study employs UA to gauge the prediction error of the utilized models across both training and testing phases. A comparison between predicted outputs and actual data points is imperative to evaluate model reliability, for which UA proves highly suitable. This analysis involves computing various statistical measures such as absolute error, margin of error (MOE), standard deviation (StDev), standard error (SE), margin of error at a 95% confidence level (ME), white blood cell count (WBC), upper bound (UB), and lower bound (LB). The findings are subsequently documented in Table 11 . A successful model regularly demonstrates a reduced WCB value 47 .

Table 11 indicates that models MD1, MD10, MD16, MD21, MD26, and MD29 both for training, testing, and validation in the uncertainty analysis, with MD29 getting the highest position. The width of the confidence bound (WCB) of various models has been examined to establish the most accurate architectural model for predicting the positive predictive value PPV of blasting.

In comparing the computational costs of different algorithms for predicting PPV resulting from blasting vibrations. While simpler algorithms like linear regression offer relatively fast computation times, they often struggle to capture the intricacies of the geological and blasting parameters that influence PPV accurately. Conversely, more sophisticated methods such as RVM or ensemble techniques like random forests may demand higher computational resources due to their complexity, especially during model training and optimization. Moreover, employing advanced techniques like convolutional neural networks (CNNs) for predicting the PPV could substantially increase computational expenses, especially due to the extensive preprocessing of data and fine-tuning of the model. Balancing computational efficiency with predictive accuracy is crucial in selecting the optimal algorithm for our PPV prediction task, ensuring timely insights into potential blast-induced ground vibrations while maintaining computational feasibility. The comparison of computational cost of this study both for training, testing and validation have summarised in Table 12 . Based on Table 12 , MD29 shows the highest value of comparison of computational cost such as 0.7243 in training, 0.0180 in testing, and 0.0191 in validation.

Figure  17 shows radar charts that show two mathematical measures used to rate how well training, testing, and validation sets work when using various optimization models 48 . To make it easier to compare the models, a thorough scoring method was employed. Using this method, many success indicators are put together and given a rank number based on how well they work. In our study, we calculate the final score for each model by summing up their individual ranks. The model with the highest total score is considered the best performing one.

Illustration of radar plots of RMSE & MAE ( a1 , b1 , c1 ), R & IOA ( a2 , b2 , c2 ), and PI ( a3 , b3 , c3 ) in the prediction of ground vibration during ( a ) training, ( b ) testing, ( c ) validation.

Conclusion and summary

A precise and accurate assessment of ground vibration in the mining project is essential to mining engineering. Many researchers developed, trained, tested, and analysed machines, advanced machine, deep, and hybrid learning models in assessing ground vibration. This investigation utilized the conventional and optimized RVM models for the first time in ground vibration prediction. Thus, this work compares 33 RVM models (6 SRVM + 6 SRVM_GA + 6 SRVM_PSO, 5 DRVM, 5 DRVM_GA, 5 DRVM_PSO) to find the optimal performance model. The following conclusions have been mapped from the outcomes of the several analyses.

Capabilities of RVM models: The performance evaluation of each RVM model demonstrates each model achieved a performance of more than 0.85 in the testing phase, presenting a good agreement between actual and predicted ground vibrations.

Impact of Dual Kernel: The comparison of conventional SRVM and DRVM models reveals that implementing the secondary kernel function enhances the accuracy and performance of the single kernel based RVM models. Also, dual-kernel RVM models achieve higher computational costs than single-kernel SRVM models.

Effect of Optimization Algorithm: The genetic and particle swarm algorithms optimized each SRVM and DRVM model in this work. The analysis of performance reveals that the genetic and particle swarm algorithms did not improve the performance of SRVM models. Conversely, a significant performance improvement has been observed for DRVM models. The comparison of SRVM_GA, SRVM_PSO, DRVM_GA, and DRVM_PSO revealed that GA and PSO-optimized DRVM models achieved higher performance than SRVM_GA and SRVM_PSO models.

Optimal Performance Model : The analysis of performance metrics (RMSE = 21.2999 mm/s, 16.2272 mm/s, R = 0.9175, PI = 1.59, IOA = 0.8239, IOS = 0.2541), score analysis (= 93), REC curve (= 6.85E-03, close to the actual, i.e., 0), curve fitting (= 1.05 close to best fit, i.e., 1), AD test (= 11.607 close to the actual, i.e., 9.790), Wilcoxon test (= 95%), Uncertainty analysis (WCB = 0.0134), and computational cost (= 0.0180) demonstrate that PSO_DRVM model MD29 outperformed the MD1 (conventional SRVM), MD10 (SRVM_GA), MD16 (SRVM_PSO), MD21 (Conventional DRVM), and MD26 (DRVM_GA) models in the testing phase.

To conclude, the present investigation introduces a particle swarm-optimized Gaussian + exponential kernel-based DRVM model as an optimal performance model for assessing ground vibration in rock blasting. The performance and accuracy of model MD29 demonstrates high capabilities. Therefore, the MD29 model may be implemented to estimate the blasting vibration in mining projects. The current investigation uses 200 data points, which is a research limitation. The overfitting of the soft computing models may be examined using more field databases. In addition, this investigation may be extended by implementing metaheuristic algorithms, i.e., evolutionary, physical, nature, swarm-based, and biological algorithms. As per the authors' knowledge, the present work implements and compares the single kernel-based SRVM, dual kernel-based DRVM, SRVM_GA, SRVM_PSO, DRVM_GA, and DRVM_PSO models in assessing the ground vibrations for the first time. This research will help mining engineers and designers select the best kernel function and its hyperparameters in estimating ground vibration. Future research directions of this study include applying the model to different geological conditions, rock mass condition and explosive property and comparing its performance against other state-of-the-art predictive models. Additionally, investigating the impact of varying blasting parameters and environmental conditions on ground vibration predictions will provide more comprehensive insights.

Data availability

All data generated or analysed during this study are included in this article.

Abbreviations

Adaptive neuro-fuzzy inference system

Bayesian regularization-based artificial neural network

Area over the curve

Burden-to-diameter ratio

Burden-to-spacing

Bias factor

Blastability Index

Cosine amplitude method

Classification and regression tree

Chi-square automatic interaction detection

Hole diameter

Degree of freedom

Distance from the blast face

Delay per row

Young's modulus

Extreme learning machine

Rock hardness

Firefly algorithm

Fuzzy inference system

Gene expression programming

Gaussian kernel function-based LSSVM

Stiffness ratio

Hole depth-to-burden ratio

K-means clustering

Hole length

Imperialist competitive algorithm

Index of agreement

Index of scatter

Interquartile range

Joint spacing

Linear kernel function-based LSSVM

Legate and McCabe's Index

Least square support vector machine

Long short-term memory

Mean absolute error

Mean absolute percentage error

Maximum charge per delay

Means square

Mean of the squared successive differences

Number of rows

Number of drilling holes

Number of datapoints

Number of missing datapoints

Normalized mean bias error

Nash-Sutcliffe efficiency

Powder factor

Performance index

Polynomial kernel function-based LSSVM

Peak particle velocity

• Particle swarm optimization

Explosive weight

First quartile

Third quartile

Correlation coefficient

Coefficient of determination

Rock density

Regression error characteristics curve

Random forest

Rock mass rating

Root mean square error

Receiver operating characteristic curve

Rock quality designation

Root mean square error to observation’s standard deviation ratio

• Relevance vector machine

Specific charge

Sine Cosine Algorithm

Sub-drilling

Scaling distance

Square error

Sum of squares

Sparrow search optimization

Standard deviation

Support vector machine

Total charge

Tensile strength

Uniaxial compressive strength

Variance accounted for

Variance inflation factor

P-wave velocity

Weighted mean absolute percentage error

Whale optimization algorithm

Extreme gradient boosting

B. O. Taiwo et al. Assessment of charge initiation techniques effect on blast fragmentation and environmental safety: An application of WipFrag software. 1–17 (2023).

Taiwo, B. O. et al. Artificial neural network modeling as an approach to limestone blast production rate prediction: A comparison of PI-BANN and MVR models. J. Min. Environ. 14 (2), 375–388. https://doi.org/10.22044/jme.2023.12489.2266 (2023).

Article   MathSciNet   Google Scholar  

Y. Fissha, H. Ikeda, H. Toriya, N. Owada, T. Adachi, & Y. Kawamura. Evaluation and prediction of blast-induced ground vibrations: A Gaussian Process Regression (GPR) Approach. 659–682 (2023).

Zhou, J., Li, C., Koopialipoor, M., Armaghani, D. J. & Pham, B. T. Development of a new methodology for estimating the amount of PPV in surface mines based on prediction and probabilistic models (GEP). Int. J. Mining Reclam. Environ. 35 (1), 48–68. https://doi.org/10.1080/17480930.2020.1734151 (2021).

Article   CAS   Google Scholar  

H. Zhang, J. Zhou, D. J. Armaghani, M. M. Tahir, & B. T. Pham. Applied sciences A combination of feature selection and random forest techniques to solve a problem related to. Appl. Sci. (2020).

Choudhary, B. S. & Agrawal, A. Minimization of blast-induced hazards and efficient utilization of blast energy by implementing a novel stemming plug system for eco-friendly blasting in open pit mines. Nat. Resour. Res. 31 (6), 3393–3410. https://doi.org/10.1007/s11053-022-10126-8 (2022).

Olamide Taiwo, B. Improvement of small-scale dolomite blasting productivity: Comparison of existing empirical models with image analysis software and artificial neural network models. J. Min. Environ. 13 (3), 627–641. https://doi.org/10.22044/jme.2022.11771.2169 (2022).

Article   Google Scholar  

Fissha, Y., Ikeda, H., Toriya, H., Adachi, T. & Kawamura, Y. Application of Bayesian Neural Network (BNN) for the prediction of blast-induced ground vibration. Appl. Sci. https://doi.org/10.3390/app13053128 (2023).

Zhou, J., Zhang, Y. & Qiu, Y. State-of-the-Art Review of Machine Learning and Optimization Algorithms Applications in Environmental Effects of Blasting (Springer Netherlands., 2024). https://doi.org/10.1007/s10462-023-10636-8 .

Book   Google Scholar  

Hosseini, S. et al. Assessment of the ground vibration during blasting in mining projects using different computational approaches Cosine amplitude method. Sci. Rep. https://doi.org/10.1038/s41598-023-46064-5 (2023).

Article   PubMed   PubMed Central   Google Scholar  

Lawal, A. I., Kwon, S., Hammed, O. S. & Idris, M. A. Blast-induced ground vibration prediction in granite quarries: An application of gene expression programming, ANFIS, and sine cosine algorithm optimized International Journal of Mining Science and Technology Blast-induced ground vibration prediction in. Int. J. Min. Sci. Technol. https://doi.org/10.1016/j.ijmst.2021.01.007 (2021).

Chen, W., Hasanipanah, M., Nikafshan Rad, H., Jahed Armaghani, D. & Tahir, M. M. A new design of evolutionary hybrid optimization of SVR model in predicting the blast-induced ground vibration. Eng. Comput. 37 (2), 1455–1471. https://doi.org/10.1007/s00366-019-00895-x (2021).

S. Alzabeebee, M. Jamei, M. Hasanipanah, & H. B. Amnieh. Development of a new explicit soft computing model to predict the blast-induced ground vibration. (2022), https://doi.org/10.12989/gae.2022.30.6.551 .

Erten, O., Konak, G., Kizil, M. S., Onur, A. H. & Karakus, D. Analysis of quarry-blast-induced ground vibrations to mitigate their adverse effects on nearby structures. Int. J. Min. Miner. Eng. 1 (4), 313–326. https://doi.org/10.1504/IJMME.2009.029317 (2009).

Khandelwal, M. & Singh, T. N. Prediction of blast-induced ground vibration using artificial neural network. Int. J. Rock Mech. Min. Sci. 46 (7), 1214–1222. https://doi.org/10.1016/j.ijrmms.2009.03.004 (2009).

Zhang, Y., He, H., Khandelwal, M., Du, K. & Zhou, J. Knowledge mapping of research progress in blast-induced ground vibration from 1990 to 2022 using CiteSpace-based scientometric analysis. Environ. Sci. Pollut. Res. 30 (47), 103534–103555. https://doi.org/10.1007/s11356-023-29712-1 (2023).

Ragam, P. & Nimaje, D. S. Assessment of blast-induced ground vibration using different predictor approaches—A comparison. Chem. Eng. Trans. 66 , 487–492. https://doi.org/10.3303/CET1866082 (2018).

Alipour, A., Mokhtarian, M. & Sharif, J. A. Artificial neural network or empirical criteria? A comparative approach in evaluating maximum charge per delay in surface mining—Sungun copper mine. J. Geol. Soc. India 79 (6), 652–658. https://doi.org/10.1007/s12594-012-0102-3 (2012).

Taheri, K., Hasanipanah, M., Golzar, S. B. & Majid, M. Z. A. A hybrid artificial bee colony algorithm-artificial neural network for forecasting the blast-produced ground vibration. Eng. Comput. 33 (3), 689–700. https://doi.org/10.1007/s00366-016-0497-3 (2017).

Nguyen, H., Bui, X.-N. & Topal, E. Reliability and availability artificial intelligence models for predicting blast-induced ground vibration intensity in open-pit mines to ensure the safety of the surroundings. Reliab. Eng. Syst. Saf. 231 , 109032 (2023).

Zhang, H. et al. A combination of feature selection and random forest techniques to solve a problem related to blast-induced ground vibration. Appl. Sci. https://doi.org/10.3390/app10030869 (2020).

Article   PubMed   Google Scholar  

Zhou, J., Asteris, P. G., Armaghani, D. J. & Pham, B. T. Prediction of ground vibration induced by blasting operations through the use of the Bayesian Network and random forest models. Soil Dyn. Earthq. Eng. https://doi.org/10.1016/j.soildyn.2020.106390 (2020).

Huang, J., Koopialipoor, M. & Armaghani, D. J. A combination of fuzzy Delphi method and hybrid ANN-based systems to forecast ground vibration resulting from blasting. Sci. Rep. 10 (1), 1–21 (2020).

J. Zhou, C. Li, M. Koopialipoor, D. J. Armaghani, B. T. Pham. (2020) Development of a new methodology for estimating the amount of PPV in surface mines based on prediction and probabilistic models (GEP-MC). Int. J. Mining, Reclam. Environ . https://doi.org/10.1080/17480930.2020.1734151 .

Nguyen, H., Bui, X.-N., Tran, Q.-H. & Mai, N.-L. A new soft computing model for estimating and controlling blast-produced ground vibration based on hierarchical K-means clustering and cubist algorithms. Appl. Soft Comput. 77 , 376–386 (2019).

Nguyen, H., Choi, Y., Bui, X. N. & Nguyen-Thoi, T. Predicting blast-induced ground vibration in open-pit mines using vibration sensors and support vector regression-based optimization algorithms. Sensors (Switzerland). https://doi.org/10.3390/s20010132 (2020).

Article   PubMed Central   Google Scholar  

Armaghani, D. J., Hasanipanah, M., Amnieh, H. B. & Mohamad, E. T. Feasibility of ICA in approximating ground vibration resulting from mine blasting. Neural Comput. Appl. 29 (9), 457–465 (2018).

Hasanipanah, M., Faradonbeh, R. S., Amnieh, H. B., Armaghani, D. J. & Monjezi, M. Forecasting blast-induced ground vibration developing a CART model. Eng. Comput. https://doi.org/10.1007/s00366-016-0475-9 (2017).

Ghoraba, S., Monjezi, M., Talebi, N., Armaghani, D. J. & Moghaddam, M. R. Estimation of ground vibration produced by blasting operations through intelligent and empirical models. Environ. Earth Sci. https://doi.org/10.1007/s12665-016-5961-2 (2016).

Shirani Faradonbeh, R. et al. Prediction of ground vibration due to quarry blasting based on gene expression programming: A new model for peak particle velocity prediction. Int. J. Environ. Sci. Technol. https://doi.org/10.1007/s13762-016-0979-2 (2016).

Hajihassani, M., Armaghani, D. J., Marto, A. & Mohamad, E. T. Ground vibration prediction in quarry blasting through an artificial neural network optimized by imperialist competitive algorithm. Bull. Eng. Geol. Environ. 74 (3), 873–886 (2015).

M. Hajihassani, D. Jahed, & A. Masoud. Blast-induced air and ground vibration prediction: A particle swarm optimization-based artificial neural network approach. (2015). https://doi.org/10.1007/s12665-015-4274-1 .

Hasanipanah, M., Monjezi, M., Shahnazar, A., Jahed Armaghani, D. & Farazmand, A. Feasibility of indirect determination of blast induced ground vibration based on support vector machine. Meas. J. Int. Meas. Confed. https://doi.org/10.1016/j.measurement.2015.07.019 (2015).

Armaghani, D. J., Momeni, E., Abad, S. V. A. N. K. & Khandelwal, M. Feasibility of ANFIS model for prediction of ground vibrations resulting from quarry blasting. Environ. Earth Sci. https://doi.org/10.1007/s12665-015-4305-y (2015).

Armaghani, D. J., Hajihassani, M., Mohamad, E. T., Marto, A. & Noorani, S. A. Blasting-induced flyrock and ground vibration prediction through an expert artificial neural network based on particle swarm optimization. Arab. J. Geosci. 7 (12), 5383–5396 (2014).

Mohamadnejad, M., Gholami, R. & Ataei, M. Comparison of intelligence science techniques and empirical methods for prediction of blasting vibrations. Tunn. Undergr. Sp. Technol. 28 , 238–244 (2012).

Monjezi, M., Hasanipanah, M. & Khandelwal, M. Evaluation and prediction of blast-induced ground vibration at Shur River Dam, Iran, by artificial neural network. Neural Comput. Appl. 22 (7–8), 1637–1643. https://doi.org/10.1007/s00521-012-0856-y (2013).

Mohamed, M. T. Performance of fuzzy logic and artificial neural network in prediction of ground and air vibrations. JES. J. Eng. Sci. 39 (2), 425–440 (2011).

Google Scholar  

A. Fi & C. Kuzu. Prediction of environmental impacts of quarry blasting operation using fuzzy logic. 461–470 (2011). https://doi.org/10.1007/s10661-010-1470-z .

Iphar, M., Yavuz, M. & Ak, H. Prediction of ground vibrations resulting from the blasting operations in an open-pit mine by adaptive neuro-fuzzy inference system. Environ. Geol. https://doi.org/10.1007/s00254-007-1143-6 (2008).

Hammed, O. S. et al. Peak particle velocity data acquisition for monitoring blast induced earthquakes in quarry sites. Data Br. 19 , 398–408. https://doi.org/10.1016/j.dib.2018.04.103 (2022).

Bahmed, I. T., Khatti, J. & Grover, K. S. Hybrid soft computing models for predicting unconfined compressive strength of lime stabilized soil using strength property of virgin cohesive soil. Bull. Eng. Geol. Environ. https://doi.org/10.1007/s10064-023-03537-1 (2024).

Khatti, J., Grover, K. S., Kim, H. J., Mawuntu, K. B. A. & Park, T. W. Prediction of ultimate bearing capacity of shallow foundations on cohesionless soil using hybrid LSTM and RVM approaches: An extended investigation of multicollinearity. Comput. Geotech. 165 , 105912. https://doi.org/10.1016/j.compgeo.2023.105912 (2024).

J. Qui. Learning with uncertainty—Gaussian processes and relevance vector machines. (2004).

Ghorbani, B., Arulrajah, A., Narsilio, G., Horpibulsuk, S. & Win, M. ScienceDirect Development of genetic-based models for predicting the resilient modulus of cohesive pavement subgrade soils. Soils Found. 60 (2), 398–412. https://doi.org/10.1016/j.sandf.2020.02.010 (2020).

J. Khatti, H. Samadi, & K. S. Grover. in Estimation of Settlement of Pile Group in Clay Using Soft Computing Techniques , no. 0123456789. (Springer International Publishing, 2023). https://doi.org/10.1007/s10706-023-02643-x .

Bardhan, A., Samui, P., Ghosh, K., Gandomi, A. H. & Bhattacharyya, S. ELM-based adaptive neuro swarm intelligence techniques for predicting the California bearing ratio of soils in soaked conditions. Appl. Soft Comput. 110 , 107595. https://doi.org/10.1016/j.asoc.2021.107595 (2021).

Xi, B., Li, E., Fissha, Y., Zhou, J. & Segarra, P. LGBM-based modeling scenarios to compressive strength of recycled aggregate concrete with SHAP analysis. Mech. Adv. Mater. Struct. https://doi.org/10.1080/15376494.2023.2224782 (2023).

Download references

Yewuhalashet Fissha want to acknowledge the funding supported by Akita University Fellowship Program.

Author information

Authors and affiliations.

Department of Geosciences, Geotechnology, and Materials Engineering for Resources, Graduate School of International Resource Sciences, Akita University, Akita, 010-8502, Japan

Yewuhalashet Fissha, Narihiro Owada, Hisatoshi Toriya & Tsuyoshi Adachi

Department of Mining Engineering, Aksum University, 7080, Aksum, Tigray, Ethiopia

Yewuhalashet Fissha

Department of Civil Engineering, Rajasthan Technical University, Kota, Rajasthan, India

Jitendra Khatti & Kamaldeep Singh Grover

Department of Systems, Control and Information Engineering, National Institute of Technology, Asahikawa College, 2-2-1-6 Syunkodai, Asahikawa City, Hokkaido, 071-8142, Japan

Hajime Ikeda

Faculty of Engineering, Hokkaido University, Kita 8, Nishi 5, Kita-Ku, Sapporo, 0608628, Japan

Youhei Kawamura

You can also search for this author in PubMed   Google Scholar

Contributions

Y.F, and J.K: Conceptualization, Methodology, Investigation, Software, Writing – original draft, Application of AI models, Relevance vector machine model development, Statistical analysis, Detailing, and Overall analysis. H.I, K.S.G, and H.T.: Conceptualization, Resources, Supervision, Testing. Y.F and J.K.: Conceptualization, Writing – original draft, Methodology, Software, Investigation, Resources, Validation, Formal analysis. Y.F, N.O. and J.K.: Investigation, Formal analysis, Validation, Software, Writing – review & editing. H.I, K.S.G, T.A, and H.T.: Visualization, Validation, Project administration, acquired the funding for this research.: Investigation, Formal analysis, Validation, Software, Writing – review & editing. H.T. T.A., and Y. K.: Visualization, Project administration, Software, Formal analysis, Validation, Writing – review & editing.

Corresponding authors

Correspondence to Yewuhalashet Fissha or Jitendra Khatti .

Ethics declarations

Competing interests.

The authors declare no competing interests.

Additional information

Publisher's note.

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License, which permits any non-commercial use, sharing, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if you modified the licensed material. You do not have permission under this licence to share adapted material derived from this article or parts of it. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by-nc-nd/4.0/ .

Reprints and permissions

About this article

Cite this article.

Fissha, Y., Khatti, J., Ikeda, H. et al. Predicting ground vibration during rock blasting using relevance vector machine improved with dual kernels and metaheuristic algorithms. Sci Rep 14 , 20026 (2024). https://doi.org/10.1038/s41598-024-70939-w

Download citation

Received : 11 June 2024

Accepted : 22 August 2024

Published : 28 August 2024

DOI : https://doi.org/10.1038/s41598-024-70939-w

Share this article

Anyone you share the following link with will be able to read this content:

Sorry, a shareable link is not currently available for this article.

Provided by the Springer Nature SharedIt content-sharing initiative

• Genetic algorithm

By submitting a comment you agree to abide by our Terms and Community Guidelines . If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.

Quick links

• Explore articles by subject
• Guide to authors
• Editorial policies

Sign up for the Nature Briefing: AI and Robotics newsletter — what matters in AI and robotics research, free to your inbox weekly.

SOAR: New algorithms for even faster vector search with ScaNN

April 10, 2024

Philip Sun and Ruiqi Guo, Software Engineers, Google Research

SOAR is an algorithmic improvement to vector search that introduces effective and low-overhead redundancy to ScaNN, Google’s vector search library, making ScaNN even more efficient.

Quick links

• Vertex AI Vector Search
• ScaNN for AlloyDB index
• Copy link ×

Efficient vector similarity search is critical for many machine learning (ML) applications at Google. It’s commonly used to search over embeddings , which are vector representations of real-world entities, such as images, websites, or other media. These vector representations allow computers to mathematically compare and find similarities between objects, for example grouping together portraits of the same subject. Once the dataset of embeddings becomes too large for the brute-force approach of comparing the query to every embedding in the dataset, more efficient vector similarity search methods become necessary for further scaling.

The desire to highlight Google’s innovations in vector search algorithms are what led to the open-sourcing of the ScaNN vector search library in 2020 (more in this prior Research Blog post ). Since then, larger datasets and the popularization of vector search in novel use cases (such as retrieval-augmented generation ) have both driven demand for more scalable vector search algorithms.

ScaNN has been actively maintained and improved since its release, but today we are particularly excited to announce a major algorithmic advancement to ScaNN: Spilling with Orthogonality-Amplified Residuals (SOAR). In “ SOAR: Improved Indexing for Approximate Nearest Neighbor Search ,” presented at NeurIPS 2023 , we describe how introducing some redundancy to ScaNN’s vector index leads to improved vector search efficiency, all while minimally impacting index size and other key vector index metrics. SOAR helps ScaNN step up to meet the ever-growing scaling and performance demands placed upon vector search libraries.

Originally featured in the 2020 ScaNN open-source release blog post , the above illustration shows how vector search over embeddings can help answer natural-language queries over a hypothetical literary database. Today we are showcasing research that allows ScaNN to perform vector search even faster.

SOAR and redundancy

The key intuition to SOAR is to introduce some mathematically crafted and implementation-optimized redundancy to ScaNN’s approximate vector search routine. Redundancy is the concept of using multiple replicas, which can each function as backup when another replica fails, to decrease the chance of total failure of a given system. Redundancy is used in many engineering disciplines, from having multiple systems that can adjust wing flaps to ensure an airplane’s stability, to storing copies of data across multiple physical drives to protect against data loss. On the other hand, redundancy may provide a false sense of security when its replicas are vulnerable to common, correlated failures — for example, storing multiple copies of data in the same datacenter still exposes them to the common threat of natural disaster in the area, and is therefore less robust than geographically distributed replicas of that data.

One common example of redundancy is replicating data across multiple datacenters around the world to decrease the chance of data loss. SOAR uses redundancy to decrease the risk of failing to find a nearest-neighbor vector.

In the context of approximate vector search, “failure” means failing to find the true nearest neighbors for a query, and instead retrieving less similar vectors. In accordance with the above themes, SOAR’s redundancy makes it less likely for ScaNN to miss the nearest neighbors, and SOAR’s specific mathematical formulation, discussed below, minimizes correlated failures that would otherwise harm search efficiency. As a result, SOAR improves the search accuracy achievable at a fixed search cost, or equivalently decreases search cost needed to achieve the same search accuracy.

SOAR: Mathematical details

We first describe ScaNN’s approximate search paradigm without SOAR in order to highlight SOAR’s differences. We focus on how ScaNN solves the maximum inner product search (MIPS) variant of vector similarity search, which defines vector similarity as the query’s inner product with a database vector. This is both because SOAR targets MIPS and because MIPS is adaptable to a variety of other similarities, including cosine and Euclidean distance.

Before ScaNN can answer vector search queries, it first goes through an indexing phase, where the input dataset is pre-processed to construct data structures that facilitate efficient vector search. A crucial part of this indexing step is clustering via k -means ; before SOAR, the clustering was performed such that each vector in the dataset was assigned to exactly one k -means cluster. Each cluster has a cluster center, which acts as a coarse approximation for the vectors assigned to the cluster; the vectors in a cluster are only evaluated if their cluster center is among the N closest centers to the query vector (with N a parameter; higher N gives greater search accuracy at the expense of greater computational cost).

An animation demonstrating how clustering was performed prior to SOAR to perform efficient vector search in ScaNN.

To illustrate when such an algorithm has difficulty finding the nearest neighbors, consider a vector x assigned to a cluster with center c . Denote the difference between x and c as the residual, or r . For a query q , the difference between the query-vector similarity ⟨ q , x ⟩ and the estimated similarity ⟨ q , c ⟩ is ⟨ q , x ⟩ - ⟨ q , c ⟩ = ⟨ q , x - c ⟩ = ⟨ q , r ⟩, which is maximized when r is parallel to q , as illustrated below:

When the query q is parallel to r, the estimated inner product ⟨q, c⟩ greatly underestimates the true inner product ⟨q, x⟩. SOAR helps find the nearest neighbors in these cases, which otherwise tend to diminish vector search accuracy.

In such situations, if x is a nearest neighbor to q , x is typically hard to find, because even though ⟨ q , x ⟩ is high, ⟨ q , r ⟩ is also high, resulting in the query-center similarity ⟨ q , c ⟩ being low, so this particular cluster is likely to be pruned and not searched further. There are a number of ways to mitigate this problem. For example, computing a higher-quality clustering tends to decrease the magnitude of r, which lowers the average estimation error, while anisotropic vector quantization (AVQ) “shapes” the error such that it tends to be largest when the query is dissimilar to x , and therefore less likely to impact results.

SOAR addresses this issue by taking a completely different approach: allowing vectors to be assigned to more than one cluster. Intuitively, this is effective by the principle of redundancy: secondary assignments may act as “backup clusters” that facilitate efficient, accurate vector search when the primary assignment performs poorly (when q is highly parallel with r of the primary assignment).

This redundancy arises from the fact that the second assignment provides a new vector-center difference r ’. As long as this r ’ isn’t near-parallel with q when r is near-parallel with q , this secondary center should help ScaNN locate the nearest neighbors to q . However, SOAR goes a step further than this naïve redundancy, and modifies the assignment loss function for secondary assignments to explicitly optimize for independent , effective redundancy: it aims to find secondary clusters whose r ’ are perpendicular to r , so that when q is near-parallel to r and the primary center has high error, q will be near-orthogonal to r ’ and the secondary center will have low error. The effect of SOAR’s modified loss is visualized below:

SOAR not only introduces redundancy via secondary assignments, but does so via a modified loss to maximize the effectiveness of redundancy. Unmodified loss (shown first) often leads to ineffective secondary assignments like c’ with high error. SOAR’s modified loss (later) selects c’’, a much better choice.

There are additional details to SOAR’s implementation and theoretical analysis, but the core idea behind SOAR is as simple as described above: assign vectors to multiple clusters with a modified loss. This technique is named SOAR because prior research has described multiple-assignment as “spilling,” and the modified loss encourages orthogonal (perpendicular) residuals, leading to the name Spilling with Orthogonality-Amplified Residuals.

Experimental results

SOAR enables ScaNN to maintain its existing advantages, including low memory consumption, fast indexing speed, and hardware-friendly memory access patterns, while endowing ScaNN with an additional algorithmic edge. As a result, ScaNN makes the best tradeoff among the three major metrics for vector search performance, as highlighted below for the ann-benchmarks glove-100 dataset . The only libraries to come near ScaNN’s querying speed require over 10× the memory and 50× the indexing time. Meanwhile, ScaNN achieves querying throughputs several times higher than libraries of comparable indexing time.

ScaNN with SOAR makes by far the best query speed / indexing speed trade-off among all libraries benchmarked, all while having the smallest memory footprint.

ScaNN also achieves state-of-the-art performance in the two tracks of Big-ANN 2023 benchmarks to which it’s applicable. These benchmarks involve larger datasets, which, in fact, further increase SOAR’s efficacy. The effect of dataset size on SOAR, as well as benchmarks on up to billion-vector datasets, are all discussed further in the paper .

For both the out-of-distribution and streaming tracks of Big-ANN 2023, SOAR allows ScaNN to achieve the highest result in each track’s respective ranking metric. The website’s leaderboard is pending update here .

SOAR provides ScaNN a robust “backup” route to identify nearest neighbors when ScaNN’s traditional clustering-based approach has the most difficulty. This allows ScaNN to perform even faster vector search, all while maintaining low index size and indexing time, leading to the all-around best set of tradeoffs among vector search algorithms.

We invite the community to use ScaNN to solve its vector search challenges. ScaNN is open-sourced on GitHub and can be easily installed via Pip . In addition, ScaNN vector search technology is available in Google Cloud products: Vertex AI Vector Search leverages ScaNN to offer a fully managed, high-scale, low-latency, vector similarity matching service, and AlloyDB recently launched ScaNN for AlloyDB index — a vector database on top of the popular PostgreSQL-compatible database. We are excited to see how more efficient vector search enables the next generation of machine learning applications.

Acknowledgements

This post reflects the work of the entire ScaNN team: David Simcha, Felix Chern, Philip Sun, Ruiqi Guo, Sanjiv Kumar and Zonglin Li. We’d also like to thank Apurv Suman, Dave Dopson, John Guilyard, Rasto Lenhardt, Susie Flynn, and Yannis Papakonstantinou.

• Algorithms & Theory
• Conferences & Events
• Data Mining & Modeling

Other posts of interest

August 16, 2024

• Data Mining & Modeling ·
• Machine Intelligence ·

June 4, 2024

May 30, 2024

• Conferences & Events ·
• Natural Language Processing

Trending Searches

• Text Effects
• Business Cards

Vector Pages

Photo pages, video pages.

• Search by Image

Explore Other Popular Vector Searches

Recent searches, research analysis vector images,  - 98,173 royalty free vector graphics and clipart matching research analysis.

• market research
• infographic
• flat design
• business research
• scientific research
• analysis icon
• business analysis
• data analysis
• illustration
• information

• system analysis
• cost analysis
• clinical research
• medical research
• competitor analysis

• market analysis
• research question
• statistical analysis
• financial analysis
• research paper
• research and development

Popular Searches

Related Searches

1500 Logos Ultimate Bundle

• Vector Collections

Analysis Research SVG Vector

Free Download Analysis Research SVG vector file in monocolor and multicolor type for Sketch and Figma from Analysis Research Vectors svg vector collection. Analysis Research Vectors SVG vector illustration graphic art design format.

• COLLECTION: SVG Vector
• LICENSE: CC0 License
• UPLOADER : SVG Repo
• analysis research
• SVG Optimized
• PNG 256x256
• PNG 512x512
• PNG 1024x1024

No Ads here 🤗

Instead, you can give us a Share on Twitter.

SVG Vector SVG Vectors

Analysis Research SVG Vector is a part of SVG Vector vector collection. Following vectors are from the same pack as this vector also checkout all SVG Vector icons and vectors.

Analysis Research SVG Vectors

Checkout other Analysis Research Vectors with different styles in SVG vector and icon library.

Website Content

Search, explore, edit and share open-licensed SVG vectors

Free Vectors

• Monocolor Vectors
• Multicolor Vectors
• Filled Vectors
• Outlined Vectors
• Icon Vectors

Other Collections

• Glyphs Vectors
• Circular Vectors
• Duotone Vectors
• Flat Vectors
• Submit Assets

Developers / Designers

• vectormaker

• Company Overview
• Staff & Associates
• Technology Partners
• Quote / Buy
• License Maintenance Agreement
• License Agreement

Software for use with MATLAB®

• PLS_Toolbox
• MIA_Toolbox
• DataSet Object

Stand-Alone Software

Prediction engines.

• Solo_Predictor
• Model_Exporter

Other Products

• Floating License Server

PLS_Toolbox Unleashed!

Solo  empowers scientists and engineers with a host of point-and-click data-discovery tools including PLS, PCA and many other multivariate and machine learning methods. It includes the main  PLS_Toolbox  graphical user interfaces, but MATLAB is not required! Import data from a variety of different file types and quickly assemble it into convenient DataSet objects to easily manage labels, axis scales, and classes. Include/exclude data from the analysis with just a click. Drag-and-drop data to our comprehensive Analysis window for modeling and analyzing results. Our integrated modeling guide and access to a large number of analysis and preprocessing techniques makes this graphical user interface environment appropriate for novice and expert users alike.

Solo provides the  Graphical Interfaces  to quickly manage and analyze data, author and apply models and interpret results.

Key Methods Included:

• Data Exploration and Pattern Recognition (Principal Components Analysis (PCA), Parallel Factor Analysis (PARAFAC), t-distributed Stochastic Neighbor Embedding (t-SNE), Uniform Manifold Approximation and Projection (UMAP), Multiway PCA…)
• Classification (SIMCA, k-nearest neighbors, Partial Least Squares Discriminant Analysis (PLS-DA), Support Vector Machine Classification (SVM-DA), Artificial Neural Network Classification (ANN-DA), Deep Learning ANN Classification (ANNDL-DA)Boosted Regression and Classification Trees (XGBoost), Clustering (HCA)…)
• Linear and Non-Linear Regression (Partial Least Squares (PLS), Principal Components Regression (PCR), Multiple Linear Regression (MLR), Classical Least Squares (CLS), Support Vector Machine Regression (SVM), Artificial Neural Networks (ANN), Deep Learning ANN (ANNDL), Boosted Regression and Classification Trees (XGBoost), N-way PLS, Locally Weighted Regression…)
• Self-modeling Curve Resolution, Pure Variable Methods (Multivariate Curve Resolution (MCR), Purity (compare to SIMPLSMA), CODA_DW, CompareLCMS…)
• Curve fitting and Distribution fitting and analysis tools
• Instrument Standardization (Piece-wise Direct, Windowed Picewise, OSC, Generalized Least Squares Preprocessing, Spectral Subspace Transformation (SST)…)
• Advanced Graphical Data Set Editing and Visualization Tools
• Advanced Customizable Order-Specific Preprocessing (Centering, Scaling, Smoothing, Derivatizing, Transformations, Baselining, Generalized Least Squares Weighting (GLSW) and many, many more)
• Missing Data Support (SVD and NIPALS)
• Variable Selection (Genetic algorithms, iPLS, Selectivity Ratio, VIP…)

System Requirements

In general, Eigenvector products should work on most modern computers. See our installation instructions ( https://wiki.eigenvector.com/index.php?title=Installation ) for detailed information.

Product Support

Eigenvector Research offers user support for Solo by e-mail at  [email protected] . Questions are almost always answered within 24 hours (and usually much less). Updates and bug fixes can be downloaded from our web site. For information on other support options, see our  technical support page .

Get More Information:

• Download a fully functional 45-day demo
• Learn about our latest release
• View the Software User Guide on-line
• View the License Maintenance Agreement
• Users with current License Maintenance Agreements: download the newest version now . IT’S FREE!
• New users:  order Solo on-line.

For information on multi-client servers, site-licenses, and OEM options, contact us by phone (509.662.9213) or e-mail ( [email protected]) .  Our product pricelist information page includes pricing and other order information for all of our products.

We're transitioning to a new login system. You will be required to reset your password the first time you log into the new system. Please contact us at [email protected] if you have any trouble accessing your account.

We Trust in Human Precision

20,000+ Professional Language Experts Ready to Help. Expertise in a variety of Niches.

API Solutions

• API Pricing
• Cost estimate
• Customer loyalty program
• Educational Discount
• Non-Profit Discount
• Green Initiative Discount1

Value-Driven Pricing

Unmatched expertise at affordable rates tailored for your needs. Our services empower you to boost your productivity.

• Special Discounts
• Enterprise transcription solutions
• Enterprise translation solutions
• Transcription/Caption API
• AI Transcription Proofreading API

Trusted by Global Leaders

GoTranscript is the chosen service for top media organizations, universities, and Fortune 50 companies.

GoTranscript

One of the Largest Online Transcription and Translation Agencies in the World. Founded in 2005.

Speaker 1: In this video, we're going to look at the ever popular qualitative analysis method, thematic analysis. We'll unpack what thematic analysis is, explore its strengths and weaknesses, and explain when and when not to use it. By the end of the video, you'll have a clearer understanding of thematic analysis so that you can approach your research project with confidence. By the way, if you're currently working on a dissertation or thesis or research project, be sure to grab our free dissertation templates to help fast-track your write-up. These tried and tested templates provide a detailed roadmap to guide you through each chapter, section by section. If that sounds helpful, you can find the link in the description down below. So, first things first, what is thematic analysis? Well, as the name suggests, thematic analysis, or TA for short, is a qualitative analysis method focused on identifying patterns, themes, and meanings within a data set. Breaking that down a little, TA involves interpreting language-based data to uncover categories or themes that relate to the research aims and research questions of the study. This data could be taken from interview transcripts, open-ended survey responses, or even social media posts. In other words, thematic analysis can be used on both primary and secondary data. Let's look at an example to make things a little more tangible. Assume you're researching customer sentiment toward a newly launched product line. Using thematic analysis, you could review open-ended survey responses from a sample of consumers looking for similarities, patterns, and categories in the data. These patterns would form a foundation for the development of an initial set of themes. You'd then reduce and synthesize these themes by filtering them through the lens of your specific research aims until you have a small number of key themes that help answer your research questions. By the way, if you're not familiar with the concept of research aims and research questions, be sure to check out our primer video covering that. Link in the description. Now that we've defined what thematic analysis is, let's unpack the different forms that TA can take, specifically inductive and deductive. Your choice of approach will make a big difference to the analysis process, so it's important to understand the difference. Let's take a look at each of them. First up is inductive thematic analysis. This type of TA is completely bottom-up, inductive in terms of approach. In other words, the codes and themes will emerge exclusively from your analysis of the data as you work through it rather than being determined beforehand. This makes it a relatively flexible approach as you can adjust, remove, or add codes and themes as you become more familiar with your data. For example, you could use inductive TA to conduct research on staff experiences of a new office space. In this case, you'd conduct interviews and begin developing codes based on the initial patterns you observe. You could then adjust or change these codes on an iterative basis as you become more familiar with the full data set, following which you develop your themes. By the way, if you're not familiar with the process of qualitative coding, we've got a dedicated video covering that. As always, the link is in the description. Next up, we've got deductive thematic analysis. Contrasted to the inductive option, deductive TA uses predetermined, tightly defined codes. These codes, often referred to as a priori codes, are typically drawn from the study's theoretical framework, as well as empirical research and the researcher's knowledge of the situation. Typically, these codes would be compiled into a codebook where each code would be clearly defined and scoped. As an example, your research might aim to assess constituent opinions regarding local government policy. Applying deductive thematic analysis here would involve developing a list of tightly defined codes in advance based on existing theory and knowledge. Those codes would then be compiled into a codebook and applied to interview data collected from constituents. Importantly, throughout the coding and analysis process, those codes and their descriptions would remain fixed. It's worth mentioning that deductive thematic analysis can be undertaken both individually or by multiple researchers. The latter is referred to as coding reliability TA. As the name suggests, this approach aims to achieve a high level of reliability with regard to the application of codes. By having multiple researchers apply the same set of codes to the same data set, inconsistencies in interpretation can be ironed out and a higher level of reliability can be reached. By the way, qualitative coding is something that we regularly help students with here at Grad Coach, so if you'd like a helping hand with your research project, be sure to check out the link that's down in the description. All right, we've covered quite a lot here. To recap, thematic analysis can be conducted using either an inductive approach where your codes naturally emerge from the data or a deductive approach where your codes are independently or collaboratively developed before analyzing the data. So now that we've unpacked the different types of thematic analysis, it's important to understand the broader strengths and weaknesses of this method so that you know when and when not to use it. One of the main strengths of thematic analysis is the relative simplicity with which you can derive codes and themes and, by extension, conclusions. Whether you take an inductive or a deductive approach, identifying codes and themes can be an easier process with thematic analysis than with some other methods. Discourse analysis, for example, requires both an in-depth analysis of the data and a strong understanding of the context in which that data was collected, demanding a significant time investment. Flexibility is another major strength of thematic analysis. The relatively generic focus on identifying patterns and themes allows TA to be used on a broad range of research topics and data types. Whether you're undertaking a small sociological study with a handful of participants or a large market research project with hundreds of participants, thematic analysis can be equally effective. Given these attributes, thematic analysis is best used in projects where the research aims involve identifying similarities and patterns across a wide range of data. This makes it particularly useful for research topics centered on understanding patterns of meaning expressed in thoughts, beliefs, and opinions. For example, research focused on identifying the thoughts and feelings of an audience in response to a new ad campaign might utilize TA to find patterns in participant responses. All that said, just like any analysis method, thematic analysis has its shortcomings and isn't suitable for every project. First, the inherent flexibility of TA also means that results can at times be kind of vague and imprecise. In other words, the broad applicability of this method means that the patterns and themes you draw from your data can potentially lack the sensitivity to incorporate text and contradiction. Second is the problem of inconsistency and lack of rigor. Put another way, the simplicity of thematic analysis can sometimes mean it's a little too crude for more delicate research aims. Specifically, the focus on identifying patterns and themes can lead to results that lack nuance. For example, even an inductive thematic analysis applied to a sample of just 10 participants might overlook some of the subtle nuances of participant responses in favor of identifying generalized themes. It could also miss fine details in language and expression that might reveal counterintuitive but more accurate implications. All that said, thematic analysis is still a useful method in many cases, but it's important to assess whether it fits your needs. So think carefully about what you're trying to achieve with your research project. In other words, your research aims and research questions. And be sure to explore all the options before choosing an analysis method. If you need some inspiration, we've got a video that unpacks the most popular qualitative analysis methods. Link is in the description. If you're enjoying this video so far, please help us out by hitting that like button. You can also subscribe for loads of plain language actionable advice. If you're new to research, check out our free dissertation writing course, which covers everything you need to get started on your research project. As always, links in the description. Okay, that was a lot. So let's do a quick recap. Thematic analysis is a qualitative analysis method focused on identifying patterns of meaning as themes within data, whether primary or secondary. As we've discussed, there are two overarching types of thematic analysis. Inductive TA, in which the codes emerge from an initial review of the data itself and are revised as you become increasingly familiar with the data. And deductive TA, in which the codes are determined beforehand based on a combination of the theoretical and or conceptual framework, empirical studies, and prior knowledge. As with all things, thematic analysis has its strengths and weaknesses and based on those is generally most appropriate for research focused on identifying patterns in data and drawing conclusions in relation to those. If you liked the video, please hit that like button to help more students find this content. For more videos like this one, check out the Grad Coach channel and make sure you subscribe for plain language, actionable research tips and advice every week. Also, if you're looking for one-on-one support with your dissertation, thesis, or research project, be sure to check out our private coaching service where we hold your hand throughout the research process step by step. You can learn more about that and book a free initial consultation at gradcoach.com.

• Data analysis
• Business analytics
• Data strategy
• Research data
• Business statistics
• Business data
• Data analytics
• Business intelligence
• Data insights

Analysis Research Vectors

• Alzheimer's disease & dementia
• Arthritis & Rheumatism
• Attention deficit disorders
• Autism spectrum disorders
• Biomedical technology
• Diseases, Conditions, Syndromes
• Endocrinology & Metabolism
• Gastroenterology
• Gerontology & Geriatrics
• Health informatics
• Inflammatory disorders
• Medical economics
• Medical research
• Medications
• Neuroscience
• Obstetrics & gynaecology
• Oncology & Cancer
• Ophthalmology
• Overweight & Obesity
• Parkinson's & Movement disorders
• Psychology & Psychiatry
• Radiology & Imaging
• Sleep disorders
• Sports medicine & Kinesiology
• Vaccination
• Breast cancer
• Cardiovascular disease
• Chronic obstructive pulmonary disease
• Colon cancer
• Coronary artery disease
• Heart attack
• Heart disease
• High blood pressure
• Kidney disease
• Lung cancer
• Multiple sclerosis
• Myocardial infarction
• Ovarian cancer
• Post traumatic stress disorder
• Rheumatoid arthritis
• Schizophrenia
• Skin cancer
• Type 2 diabetes
• Full List »

share this!

August 29, 2024

This article has been reviewed according to Science X's editorial process and policies . Editors have highlighted the following attributes while ensuring the content's credibility:

fact-checked

peer-reviewed publication

trusted source

Analysis finds number of people with food allergies doubled in the UK in last decade but many still without treatment

by Ryan O'Hare, Imperial College London

The number of people with food allergies in the U.K. has more than doubled since 2008, with the largest increase seen in young children.

In a new analysis, researchers from Imperial College London estimate that the number of people living with food allergies in England has increased significantly between 2008 and 2018. The analysis found that up to 4% of preschool-aged children have a food allergy , while rates in adults were around 1%. These rates are likely representative of the U.K.

The analysis , published in The Lancet Public Health , also found that many people with a previous severe reaction from a food allergy (anaphylaxis) were not prescribed adrenaline autoinjector "pens" which can be lifesaving.

The researchers highlight that newly diagnosed cases of food allergy may be leveling-off in some age groups, but say it's unclear whether this might be due to changes in infant feeding guidelines.

However, most patients with a food allergy, including those with previous anaphylaxis, are only seen in the primary care setting: only around 10% were seen at least once by a hospital specialist during the 10-year study period.

The authors say there is an urgent need to ensure GPs and other primary care staff are adequately trained and supported to care for people at risk of severe food allergy.

Dr. Paul Turner, Professor of Pediatric Allergy at the National Heart & Lung Institute at Imperial College London, who led the research, said, "This new analysis paints an important, if mixed, picture of food allergy in the U.K.

"The good news is that while the prevalence of food allergy has increased, the numbers of new cases occurring each year look to have plateaued. However, more than one-third of patients at risk of severe reactions do not carry potentially lifesaving rescue adrenaline autoinjectors, like EpiPens.

"We urgently need to address this, and better support GPs and primary care staff who end up looking after the vast majority of food-allergic patients in the U.K."

Food allergy occurs when the immune system overreacts to certain foods, such as cows' milk, peanuts, eggs or shellfish. In severe cases, it can lead to anaphylaxis, which can cause swelling of the airways, breathing difficulties, and cardiac arrest if not treated urgently.

In the latest study, researchers estimated trends in the clinical incidence and prevalence of food allergy in the U.K. population. Using anonymized data from GP practices covering 13 million patients, the team searched for instances of food allergy between 2008 and 2018 and linked the data to relevant GP and hospital encounters in England.

The analysis found that the total estimated prevalence of food allergy in the U.K. increased from 0.4% to 1.1% from 2008 and 2018. In 2018, the highest prevalence was seen in children under 5 years old (4%), with a lower prevalence in school-aged children (2.4% for children aged 5- to 9-years old and 1.7% for 15 to 19-year-olds), and the lowest prevalence in adults (0.7%).

The estimated number of new cases of probable food allergy in the U.K. doubled between 2008 and 2018, from 76 cases per 100,000 people in 2008 to 160 cases per 100,000 people in 2018.

More than 97% of clinical visits for food allergy were to the GP clinic, rather than the hospital setting, and around 90% of patients only had visits recorded to primary care during the entire 10-year study period.

Prescriptions for adrenaline autoinjectors in those with previous anaphylaxis were estimated to be at 64% for children and young people and just 55% for adults. Prescriptions were also less common for people living in more deprived areas of the country.

According to the researchers, their study is the first to describe how the landscape of food allergy has changed in the U.K. using a large, population-based health dataset.

They suggest that while more doctors may have diagnosed food allergy over the past decade, the incidence of new cases may have slowed, possibly due to changes in infant feeding guidelines which no longer recommend delaying the introduction of foods like peanut into the infant diet.

In children at higher risk of food allergy (e.g.. with infant eczema), earlier introduction of egg and peanut are now standard recommendations.

Professor Adnan Custovic, from the National Heart & Lung Institute at Imperial College London, said, "One concerning finding from our study is the suggestion that patients in more deprived areas of the country may not be getting access to vital rescue medications they need, and may be more likely to attend hospital emergency departments. This urgently requires further investigation."

Professor Jenny Quint, from the School of Public Health at Imperial College London, said, "Our work shows that using large health datasets can help us to cut through the noise and see changing trends in allergy in the U.K., and ultimately whether interventions are having an impact."

Earlier this year, Professors Turner and Custovic launched the Frankland-Kay Center for Allergic Diseases at Imperial College London. The Center is dedicated to advancing allergy research and reducing the impact of allergies on people and their families.

Dr. Turner added, "Food allergy can have a huge impact on people's lives, and in some tragic cases, can cut lives short. By better understanding the prevalence of food allergy in the U.K. and working with affected people, their families, scientists and clinicians, we are working together to reduce that impact."

Explore further

Feedback to editors

Managing early stages of abortion care at home after 12 weeks is safe and reduces time spent in hospital, study finds

Billions worldwide consume inadequate levels of micronutrients critical to human health, new study finds

Blocking the longevity gene S6K1 extends lifespan in mice by reducing inflammation

3 hours ago

Machine learning predicts which patients will continue taking opioids after hand surgery

Study describes a new molecular pathway involved in the control of reproduction

4 hours ago

Machine learning helps identify rheumatoid arthritis subtypes

Immune protection against tuberculosis reinfection driven by cells that dampen lung inflammation, shows study

Girls with mental health conditions have lower HPV vaccination coverage

5 hours ago

Mankai plant found to reduce post-meal sugar levels in diabetics

Unique chicken line advances research on autoimmune disease that affects humans

Related stories.

Research finds increase in allergic reactions among children in Finland, decrease in Sweden

Aug 28, 2024

Australia launches landmark peanut allergy treatment for babies

Jul 31, 2024

CDC reports on prevalence of seasonal and food allergies, eczema

Jan 26, 2023

Study finds gender affects food allergy's impact on quality of life

Feb 7, 2024

Trial data underpins FDA approval of omalizumab for food allergy

Feb 16, 2024

Serious allergic reactions to food among children stabilize since guideline changes

Feb 24, 2022

Recommended for you

Smart mask monitors breath for signs of health

6 hours ago

Mechanical stress in the borderzone: A new source of cardiac inflammation

9 hours ago

VR headsets could be life changing for people with intellectual disability

10 hours ago

Discovery of how blood clots harm brain and body in COVID-19 points to new therapy

New findings on tuberculosis could change how we treat inflammatory disorders

Let us know if there is a problem with our content.

Use this form if you have come across a typo, inaccuracy or would like to send an edit request for the content on this page. For general inquiries, please use our contact form . For general feedback, use the public comments section below (please adhere to guidelines ).

Please select the most appropriate category to facilitate processing of your request

Thank you for taking time to provide your feedback to the editors.

Your feedback is important to us. However, we do not guarantee individual replies due to the high volume of messages.

E-mail the story

Your email address is used only to let the recipient know who sent the email. Neither your address nor the recipient's address will be used for any other purpose. The information you enter will appear in your e-mail message and is not retained by Medical Xpress in any form.

Newsletter sign up

Get weekly and/or daily updates delivered to your inbox. You can unsubscribe at any time and we'll never share your details to third parties.

More information Privacy policy

Donate and enjoy an ad-free experience

We keep our content available to everyone. Consider supporting Science X's mission by getting a premium account.

E-mail newsletter

Information

• Author Services

Initiatives

You are accessing a machine-readable page. In order to be human-readable, please install an RSS reader.

All articles published by MDPI are made immediately available worldwide under an open access license. No special permission is required to reuse all or part of the article published by MDPI, including figures and tables. For articles published under an open access Creative Common CC BY license, any part of the article may be reused without permission provided that the original article is clearly cited. For more information, please refer to https://www.mdpi.com/openaccess .

Feature papers represent the most advanced research with significant potential for high impact in the field. A Feature Paper should be a substantial original Article that involves several techniques or approaches, provides an outlook for future research directions and describes possible research applications.

Feature papers are submitted upon individual invitation or recommendation by the scientific editors and must receive positive feedback from the reviewers.

Editor’s Choice articles are based on recommendations by the scientific editors of MDPI journals from around the world. Editors select a small number of articles recently published in the journal that they believe will be particularly interesting to readers, or important in the respective research area. The aim is to provide a snapshot of some of the most exciting work published in the various research areas of the journal.

Original Submission Date Received: .

• Active Journals
• Find a Journal
• Proceedings Series
• For Authors
• For Reviewers
• For Editors
• For Librarians
• For Publishers
• For Societies
• For Conference Organizers
• Open Access Policy
• Institutional Open Access Program
• Special Issues Guidelines
• Editorial Process
• Research and Publication Ethics
• Article Processing Charges
• Testimonials
• Preprints.org
• SciProfiles
• Encyclopedia

Article Menu

• Subscribe SciFeed
• Recommended Articles
• Google Scholar
• on Google Scholar
• Table of Contents

Find support for a specific problem in the support section of our website.

Please let us know what you think of our products and services.

Visit our dedicated information section to learn more about MDPI.

JSmol Viewer

Enhancing depression detection: a stacked ensemble model with feature selection and rf feature importance analysis using nhanes data.

1. Introduction

• Identifying the primary socio-demographic and psychological determinants contributing to the development of depression.
• This study aims to investigate several machine learning algorithms and feature selection techniques to detect the presence of depression effectively.
• This study aims to facilitate the interpretation of high-risk groups of depressive disorders in primary care settings for primary physicians. Primary care clinicians will achieve this by applying the Depression Ensemble Stacking Generalization Model (DESGM) to identify and analyze the essential determinants associated with these high-risk groups.

2. Related Work

2.1. risk factors, 2.2. ensemble learning techniques, 3. proposed method—depression ensemble stacking generalization model (desgm), 3.1. data source representation, 3.2. data pre-processing.

• Let A = ∅ .
• If f may indicate a symptom (e.g., trouble sleeping or sleeping too much), then add f to A.
• Let S o denote the projection of S F o onto A.

3.3. Feature Selection: Iterative with Floating Elimination Algorithm (IFEA)

3.4. Stacked Ensemble Model—DESGM

Base learners.

• Initialize w = 0.
• For each training sample X i , the output value y ^ ( i ) is computed using the following activation function: y ^ i : = Φ w T x i (2)
• Update the weights.
• The weight vector w is then updated based on the computed output and the actual class label y ^ ( i ) of the ith training sample using the update rule:

3.5. Artificial Neural Network (ANN)

3.6. k-nearest neighbor (knn), 3.7. linear support vector machine (lsvm).

• Compute the gram matrix. G i j = y i y j x i x j (7)
• Let α * value: max ∝ ⁡ ∑ i = 0 N α i − 1 2 ∑ i = 1 N ∑ j = 1 N α i α j y i y j x i x j   S u b j e c t   t o 0 ≤ α i ≤ C , ∀ i , ∑ i = 1 N α i y ( i ) = 0 (8) The value of parameter C controls the margin’s width.
• Determine the weight. w * = ∑ i = 1 N α * i y ( i ) x ( i ) (9)
• Calculate the intercept. w 0 * = 1 − min i : y i = 1 ⁡ w * T x i (10)
• Determine the support vector’s x ( i ) . k i = x · x ( i ) (11)
• Calculate f(x). f ( x ) = w 0 * + ∑ i α * i y i k i ( : w 0 * + w * T x ) (12)
• Examine sign (f(x)).

3.8. Linear Discriminant Analysis (LDA)

• Let X k be the collection of d-dimensional samples for class k, represented as { x ( 1 ) , x ( 2 ) , … , x ( N k ) } .
• Let X ∈ R d × N be the data matrix, stacking all the samples from all classes, with each column representing a sample, where N = ∑ k N k .
• Let us specify how samples x are transformed onto a line in [(c 1)-space, for c = 2]. z = w T x = w · x ,   s u b j e c t   t o   w ∈ R d is a projected vector.

Meta-Learner—Logistic Regression (LR)

3.9. depression ensemble stacking generalization model (desgm).

4. Results and Discussions

4.1. performance metrics, 4.2. results of feature selection method, 4.3. results of ensemble model, 4.4. results of desgm, 4.5. role of feature importance—random forest (rf), 4.6. discussion, 5. conclusions and future work, author contributions, institutional review board statement, informed consent statement, data availability statement, conflicts of interest.

• Institute of Health Metrics and Evaluation. Global Health Data Exchange (GHDx). Available online: https://vizhub.healthdata.org/gbd-results/ (accessed on 4 March 2023).
• Chakrabarty, T.; Sarfati, D.; Lam, R.W. Chapter 24—Neurobiological Aspects of Functional Recovery in Major Depressive Disorder. In Neurobiology of Depression ; Academic Press: Cambridge, MA, USA, 2019; pp. 277–284. [ Google Scholar ] [ CrossRef ]
• Murthy, R.S. National Mental Health Survey of India 2015–2016. Indian J. Psychiatry 2017 , 59 , 21–26. [ Google Scholar ] [ CrossRef ] [ PubMed ] [ PubMed Central ]
• Kessler, R.C.; Aguilar-Gaxiola, S.; Alonso, J.; Chatterji, S.; Lee, S.; Ormel, J.; Üstün, T.B.; Wang, P.S. The global burden of mental disorders: An update from the WHO World Mental Health (WMH) surveys. Epidemiol. Psichiatr. Soc. 2009 , 18 , 23–33. [ Google Scholar ] [ CrossRef ] [ PubMed ] [ PubMed Central ]
• Sagar, R.S.; Mohan, D.; Kumar, V.; Khandelwal, S.K.; Nair, P.G. Physical illnesses among elderly psychiatric out-patients with depression. Indian J. Psychiatry 1992 , 34 , 41–45. [ Google Scholar ] [ PubMed ] [ PubMed Central ]
• Nguyen, D.-K.; Chan, C.-L.; Li, A.-H.A.; Phan, D.-V. Deep Stacked Generalization Ensemble Learning models in early diagnosis of Depression illness from wearable devices data. In Proceedings of the 5th International Conference on Medical and Health Informatics (ICMHI ‘21), Association for Computing Machinery, New York, NY, USA, 7–12 May 2021. [ Google Scholar ] [ CrossRef ]
• Bohr, A.; Memarzadeh, K. The rise of artificial intelligence in healthcare applications. In Artificial Intelligence in Healthcare ; Academic Press: Cambridge, MA, USA, 2020; pp. 25–60. [ Google Scholar ] [ CrossRef ] [ PubMed Central ]
• Aleem, S.; Huda, N.U.; Amin, R.; Khalid, S.; Alshamrani, S.S.; Alshehri, A. Machine Learning Algorithms for Depression: Diagnosis, Insights, and Research Directions. Electronics 2022 , 11 , 1111. [ Google Scholar ] [ CrossRef ]
• Tao, X.; Chi, O.; Delaney, P.J.; Li, L.; Huang, J. Detecting depression using an ensemble classifier based on Quality of Life scales. Brain Inf. 2021 , 8 , 2198–4026. [ Google Scholar ] [ CrossRef ]
• Merikangas, A.K.; Mendola, P.; Pastor, P.N.; Reuben, C.A.; Cleary, S.D. The association between major depressive disorder and obesity in US adolescents: Results from the 2001–2004 National Health and Nutrition Examination Survey. J. Behav. Med. 2011 , 35 , 149–154. [ Google Scholar ] [ CrossRef ] [ PubMed ] [ PubMed Central ]
• Kroenke, K.; Spitzer, R.L.; Williams, J.B.; Löwe, B. The Patient Health Questionnaire Somatic, Anxiety, and Depressive Symptom Scales: A systematic review. Gen. Hosp. Psychiatry 2010 , 32 , 345–359. [ Google Scholar ] [ CrossRef ] [ PubMed ]
• Gerych, W.; Agu, E.; Rundensteiner, E. Classifying depression in imbalanced datasets using an autoencoder-based anomaly detection approach. In Proceedings of the 2019 IEEE 13th International Conference on Semantic Computing (ICSC), Newport Beach, CA, USA, 30 January–1 February 2019; pp. 124–127. [ Google Scholar ]
• Oh, J.; Yun, K.; Maoz, U.; Kim, T.S.; Chae, J.H. Identifying depression in the National Health and Nutrition Examination Survey data using a deep learning algorithm. J. Affect. Disord. 2019 , 257 , 623–631. [ Google Scholar ] [ CrossRef ] [ PubMed ]
• Dipnall, J.F.; Pasco, J.A.; Berk, M.; Williams, L.J.; Dodd, S.; Jacka, F.N.; Meyer, D. Why so GLUMM? Detecting depression clusters through graphing lifestyle-environs using machine-learning methods (GLUMM). Eur. Psychiatry 2017 , 39 , 40–50. [ Google Scholar ] [ CrossRef ] [ PubMed ]
• Sharma, A.; Verbeke, W.J.M.I. Improving Diagnosis of Depression with XGBOOST Machine Learning Model and a Large Biomarkers Dutch Dataset (n = 11,081). Front. Big Data 2020 , 3 , 15. [ Google Scholar ] [ CrossRef ] [ PubMed ] [ PubMed Central ]
• Wang, L.; Chignell, M.; Jiang, H.; Lokuge, S.; Mason, G.; Fotinos, K.; Katzman, M. Prioritization of Multi-level Risk Factors, and Predicting Changes in Depression Ratings after Treatment Using Multi-Task Learning. In Proceedings of the 2021 IEEE International Conference on Bioinformatics and Biomedicine (BIBM), Houston, TX, USA, 9–12 December 2021; pp. 3239–3244. [ Google Scholar ] [ CrossRef ]
• Wang, L.; Zhu, D.; Towner, E.; Dong, M. Obesity risk factors ranking using multi-task learning. In Proceedings of the 2018 IEEE EMBS International Conference on Biomedical & Health Informatics (BHI), Las Vegas, NV, USA, 4–7 March 2018; pp. 385–388. [ Google Scholar ]
• Jayawickreme, N.; Atefi, E.; Jayawickreme, E.; Qin, J.; Gandomi, A.H. Association Rule Learning Is an Easy and Efficient Method for Identifying Profiles of Traumas and Stressors that Predict Psychopathology in Disaster Survivors: The Example of Sri Lanka. Int. J. Environ. Res. Public. Health 2020 , 17 , 2850. [ Google Scholar ] [ CrossRef ] [ PubMed ] [ PubMed Central ]
• Jia, Z.; Li, X.; Yuan, X.; Zhang, B.; Liu, Y.; Zhao, J.; Li, S. Depression is associated with diabetes status of family members: NHANES (1999–2016). J. Affect. Disord. 2019 , 249 , 121–126. [ Google Scholar ] [ CrossRef ] [ PubMed ]
• Byeon, H. Developing a Predictive Model for Depressive Disorders Using Stacking Ensemble and Naive Bayesian Nomogram: Using Samples Representing South Korea. Front. Psychiatry 2022 , 12 , 773290. [ Google Scholar ] [ CrossRef ] [ PubMed ]
• Habib, M.; Wang, Z.; Qiu, S.; Zhao, H.; Murthy, A.S. Machine Learning Based Healthcare System for Investigating the Association Between Depression and Quality of Life. IEEE J. Biomed. Health Inform. 2022 , 26 , 2008–2019. [ Google Scholar ] [ CrossRef ] [ PubMed ]
• Thanathamathee, P. Boosting with feature selection technique for screening and predicting adolescents depression. In Proceedings of the 2014 Fourth International Conference on Digital Information and Communication Technology and its Applications (DICTAP), Bangkok, Thailand, 6–8 May 2014; pp. 23–27. [ Google Scholar ] [ CrossRef ]
• Lee, S.-J.; Moon, H.-J.; Kim, D.-J.; Yoon, Y. Genetic algorithm-based feature selection for depression scale prediction. In Proceedings of the GECCO ‘19: Genetic and Evolutionary Computation Conference, Prague, Czech Republic, 13–17 July 2019; pp. 65–66. [ Google Scholar ] [ CrossRef ]
• Santana, R. Genetic Algorithms for Feature Selection in the Children and Adolescents Depression Context. In Proceedings of the 2019 18th IEEE International Conference on Machine Learning and Applications (ICMLA), Boca Raton, FL, USA, 16–19 December 2019; pp. 1470–1475. [ Google Scholar ] [ CrossRef ]
• Yu, B.; Zhang, X.; Wang, C.; Sun, M.; Jin, L.; Liu, X. Trends in depression among Adults in the United States, NHANES 2005–2016. J. Affect. Disord. 2019 , 263 , 609–620, ISSN 0165-0327. [ Google Scholar ] [ CrossRef ]
• Thieme, A.; Belgrave, D.; Doherty, G. Machine learning in mental health: A systematic review of the hci literature to support the development of effective and implementable ML systems. ACM Trans. Comp. Hum. Interact (TOCHI) 2020 , 27 , 1–53. [ Google Scholar ] [ CrossRef ]
• Wang, Y.; Lopez, J.M.S.; Bolge, S.C.; Zhu, V.J.; Stang, P.E. Depression among people with type 2 diabetes mellitus, US National Health and Nutrition Examination Survey (NHANES), 2005–2012. BMC Psychiatry 2016 , 16 , 88. [ Google Scholar ] [ CrossRef ] [ PubMed ]
• Burns, L.; Teesson, M. Alcohol use disorders comorbid with anxiety, depression, and drug use disorders. Findings from the Australian National Survey of Mental Health and Well Being. Drug Alcohol. Depend. 2002 , 68 , 299–307. [ Google Scholar ] [ CrossRef ] [ PubMed ]
• Bromet, E.; Andrade, L.H.; Hwang, I.; Sampson, N.A.; Alonso, J.; de Girolamo, G.; de Graaf, R.; Demyttenaere, K.; Hu, C.; Iwata, N.; et al. Cross-national epidemiology of DSM-IV major depressive episode. BMC Med. 2011 , 9 , 90. [ Google Scholar ] [ CrossRef ] [ PubMed ] [ PubMed Central ]
• Ferrari, A.J.; Charlson, F.J.; Norman, R.E.; Patten, S.B.; Freedman, G.; Murray, C.J.; Vos, T.; Whiteford, H.A. Burden of depressive disorders by country, sex, age, and year: Findings from the global burden of disease study 2010. PLoS Med. 2013 , 10 , e1001547. [ Google Scholar ] [ CrossRef ] [ PubMed ] [ PubMed Central ]
• Vincent, P.M.D.R.; Mahendran, N.; Nebhen, J.; Deepa, N.; Srinivasan, K.; Hu, Y.-C. Performance Assessment of Certain Machine Learning Models for Predicting the Major Depressive Disorder among IT Professionals during Pandemic times. Comput. Intell. Neurosci. 2021 , 2021 , 9950332. [ Google Scholar ] [ CrossRef ]
• Dipnall, J.F.; Pasco, J.A.; Berk, M.; Williams, L.J.; Dodd, S.; Jacka, F.N.; Meyer, D. Fusing Data Mining, Machine Learning and Traditional Statistics to Detect Biomarkers Associated with Depression. PLoS ONE 2016 , 11 , e0148195. [ Google Scholar ] [ CrossRef ] [ PubMed ] [ PubMed Central ]
• Hassan, A.U.; Hussain, J.; Hussain, M.; Sadiq, M.; Lee, S. Sentiment analysis of social networking sites (SNS) data using machine learning approach for the measurement of depression. In Proceedings of the 2017 International Conference on Information and Communication Technology Convergence (ICTC), Jeju, Republic of Korea, 18–20 October 2017; pp. 138–140. [ Google Scholar ] [ CrossRef ]
• Yang, H.; Bath, P.A. Automatic Prediction of Depression in Older Age. In Proceedings of the third International Conference on Medical and Health Informatics 2019, ICMHI 2019, Xiamen, China, 17–19 May 2019; pp. 36–44. [ Google Scholar ] [ CrossRef ]
• Tanha, J.; Abdi, Y.; Samadi, N.; Razzaghi, N.; Asadpour, M. Boosting methods for multi-class imbalanced data classification: An experimental review. J. Big Data 2020 , 7 , 70. [ Google Scholar ] [ CrossRef ]
• Hsieh, W.-H.; Shih, D.-H.; Shih, P.-Y.; Lin, S.-B. An Ensemble Classifier with Case-Based Reasoning System for Identifying Internet Addiction. Int. J. Environ. Res. Public Health 2019 , 16 , 1233. [ Google Scholar ] [ CrossRef ]
• Kumar, U.K.; Nikhil, M.B.S.; Sumangali, K. Prediction of breast cancer using voting classifier technique. In Proceedings of the 2017 IEEE International Conference on Smart Technologies and Management for Computing, Communication, Controls, Energy and Materials (ICSTM), Chennai, India, 2–4 August 2017; pp. 108–114. [ Google Scholar ] [ CrossRef ]
• Centers for Disease Prevention and Control National Health and Nutrition Examination Survey Overview. National Center for Health Statistics. 2017. Available online: https://wwwn.cdc.gov/nchs/nhanes/search/datapage.aspx?Component=Questionnaire&CycleBeginYear=2015 (accessed on 15 March 2023).
• Byeon, H. Exploring Factors for Predicting Anxiety Disorders of the Elderly Living Alone in South Korea Using Interpretable Machine Learning: A Population-Based Study. Int. J. Environ. Res. Public Health 2021 , 18 , 7625. [ Google Scholar ] [ CrossRef ] [ PubMed ]
• Yusta, S.C. Different metaheuristic strategies to solve the feature selection problem. Pattern Recognit. Lett. 2009 , 30 , 525–534. [ Google Scholar ] [ CrossRef ]
• Zian, S.; Kareem, S.A.; Varathan, K.D. An Empirical Evaluation of Stacked Ensembles with Different Meta-Learners in Imbalanced Classification. IEEE Access 2021 , 9 , 87434–87452. [ Google Scholar ] [ CrossRef ]
• Pasco, J.A.; Williams, L.J.; Jacka, F.N.; Ng, F.; Henry, M.J.; Nicholson, G.C.; Kotowicz, M.A.; Berk, M. Tobacco Smoking as a Risk Factor for Major Depressive Disorder: Population-based Study. Br. J. Psychiatry 2008 , 193 , 322–326. [ Google Scholar ] [ CrossRef ]
• Riemann, D.; Krone, L.B.; Wulff, K.; Nissen, C. Sleep, insomnia, and depression. Neuropsychopharmacology 2020 , 45 , 74–89. [ Google Scholar ] [ CrossRef ]
• Yan, C.; Shum, D.; Deng, C.P. Responses to academic stress mediate the association between sleep difficulties and depressive/anxiety symptoms in Chinese adolescents. J. Affect. Disord. 2020 , 263 , 89–98. [ Google Scholar ] [ CrossRef ]
• Cahuas, A.; He, Z.; Zhang, Z.; Chen, W. Relationship of physical activity and sleep with depression in college students. J. Am. Coll. Health 2020 , 68 , 557–564. [ Google Scholar ] [ CrossRef ] [ PubMed ]
• Francis, H.M.; Stevenson, R.J.; Chambers, J.R.; Gupta, D.; Newey, B.; Lim, C.K. A brief diet intervention can reduce symptoms of depression in young adults—A randomised controlled trial. PLoS ONE 2019 , 14 , e0222768. [ Google Scholar ] [ CrossRef ] [ PubMed ]
• Lai, J.S.; Hiles, S.; Bisquera, A.; Hure, A.J.; McEvoy, M.; Attia, J. A systematic review and meta-analysis of dietary patterns and depression in community-dwelling adults. Am. J. Clin. Nutr. 2014 , 99 , 181–197. [ Google Scholar ] [ CrossRef ] [ PubMed ]

Click here to enlarge figure

Share and Cite

Selvaraj, A.; Mohandoss, L. Enhancing Depression Detection: A Stacked Ensemble Model with Feature Selection and RF Feature Importance Analysis Using NHANES Data. Appl. Sci. 2024 , 14 , 7366. https://doi.org/10.3390/app14167366

Selvaraj A, Mohandoss L. Enhancing Depression Detection: A Stacked Ensemble Model with Feature Selection and RF Feature Importance Analysis Using NHANES Data. Applied Sciences . 2024; 14(16):7366. https://doi.org/10.3390/app14167366

Selvaraj, Annapoorani, and Lakshmi Mohandoss. 2024. "Enhancing Depression Detection: A Stacked Ensemble Model with Feature Selection and RF Feature Importance Analysis Using NHANES Data" Applied Sciences 14, no. 16: 7366. https://doi.org/10.3390/app14167366

Article Metrics

Article access statistics, further information, mdpi initiatives, follow mdpi.

Subscribe to receive issue release notifications and newsletters from MDPI journals

Identification and molecular characterization of a novel Babesia orientalis rhoptry neck protein 4 (BoRON4)

• Published: 29 August 2024
• Volume 123 , article number  310 , ( 2024 )

Cite this article

• Fangjie Li 1 , 2   na1 ,
• Jiaying Guo 4   na1 ,
• Sen Wang 1 , 2 ,
• Zhen Han 1 , 2 ,
• Zheng Nie 1 , 2 ,
• Long Yu 1 , 2 ,
• Xiang Shu 1 , 2 ,
• Yingjun Xia 1 , 2 ,
• Lan He 1 , 2 , 3 &
• Junlong Zhao 1 , 2 , 3  

B abesia orientalis , a protozoan parasite transmitted by the tick Rhipicephalus haemaphysaloides , holds significant economic importance along the Yangtze River. Key factors in the host invasion process include rhoptry neck proteins (RON2, RON4, and RON5) and apical membrane antigen 1 (AMA1). However, the intricacies of the interaction between AMA1 and RONs remain incompletely elucidated in B. orientalis . To better understand these crucial invasion components, the RON4 gene of B. orientalis (BoRON4) was cloned and sequenced. RON4 is 3468 base pairs long, encodes 1155 amino acids, and has a predicted molecular weight of 130 kDa. Bioinformatics analysis revealed a unique region (amino acid residues 109–452) in BoRON4, which demonstrates higher sensitivity to epitope activity. The BoRON4 gene was strategically truncated, amplified, and cloned into the pGEX-6p-1 vector for fusion expression. We successfully used the mouse polyclonal antibody to identify native BoRON4 in B. orientalis lysates. Furthermore, the corresponding BoRON4 protein band was detected in the water buffalo serum infected with B. orientalis , while no such band was observed in the control. Additionally, I-TASSER and Discovery Studio software were used to predict the tertiary structures of BoRON4 and its ligands, CH-PKA and CH-complex. These ligands can serve as lead compounds for the development of anti-babesiosis drugs. In conclusion, BoRON4 emerges as a promising candidate antigen for distinguishing water buffalo infected with B. orientalis from their normal counterparts. This study positions BoRON4 as a potential diagnostic antigen for babesiosis in water buffalo, contributing valuable insights to the field of parasitology.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Subscribe and save.

• Get 10 units per month
• Download Article/Chapter or eBook
• 1 Unit = 1 Article or 1 Chapter
• Cancel anytime

Price includes VAT (Russian Federation)

Instant access to the full article PDF.

Rent this article via DeepDyve

Institutional subscriptions

Data availability

No datasets were generated or analysed during the current study.

Aston A et al (2014) Washing red cells after leucodepletion does not decrease human leukocyte antigen sensitization risk in patients with chronic kidney disease. Pediatr Nephrol 29:2005–2011. https://doi.org/10.1007/s00467-014-2823-6

Article   PubMed   Google Scholar  

Baba M, Nozaki M, Tachibana M, Tsuboi T, Torii M, Ishino T (2023) Rhoptry neck protein 4 plays important roles during Plasmodium sporozoite infection of the mammalian liver. mSphere 8:e0058722. https://doi.org/10.1128/msphere.00587-22

Article   CAS   PubMed   Google Scholar  

Besteiro S, Michelin A, Poncet J, Dubremetz J-F, Lebrun M (2009) Export of a Toxoplasma gondii rhoptry neck protein complex at the host cell membrane to form the moving junction during invasion. PLoS Pathog 5:e1000309

Article   PubMed   PubMed Central   Google Scholar  

Cuy-Chaparro L, Barney-Borrero D, Arévalo-Pinzón G, Reyes C, Moreno-Pérez DA, Patarroyo MA (2024) Babesia bovis RON2 binds to bovine erythrocytes through a highly conserved epitope. Vet Parasitol 326:110081. https://doi.org/10.1016/j.vetpar.2023.110081

Fernandes P et al (2022) The AMA1-RON complex drives Plasmodium sporozoite invasion in the mosquito and mammalian hosts. PLoS Pathog 18:e1010643. https://doi.org/10.1371/journal.ppat.1010643

Article   CAS   PubMed   PubMed Central   Google Scholar  

Ferreira MU, da Silva Nunes M, Wunderlich G (2004) Antigenic diversity and immune evasion by malaria parasites. Clin Diagn Lab Immunol 11:987–995. https://doi.org/10.1128/cdli.11.6.987-995.2004

Giovannini D et al (2011) Independent roles of apical membrane antigen 1 and rhoptry neck proteins during host cell invasion by apicomplexan. Cell Host & Microbe 10:591–602

Article   CAS   Google Scholar  

Gohil S, Herrmann S, Günther S, Cooke BM (2013) Bovine babesiosis in the 21st century: advances in biology and functional genomics. Int J Parasitol 43:125–132. https://doi.org/10.1016/j.ijpara.2012.09.008

Guedes CES et al (2018) In vitro evaluation of the anti-leishmanial activity and toxicity of PK11195. Mem Inst Oswaldo Cruz 113:e170345. https://doi.org/10.1590/0074-02760170345

Guérin A, El Hajj H, Penarete-Vargas D, Besteiro S, Lebrun M (2017) RON4(L1) is a new member of the moving junction complex in Toxoplasma gondii. Sci Rep 7:17907. https://doi.org/10.1038/s41598-017-18010-9

Guo J et al (2018) A novel Babesia orientalis 135-kilodalton spherical body protein like: identification of its secretion into cytoplasm of infected erythrocytes. Parasites & Vectors 11:205. https://doi.org/10.1186/s13071-018-2795-7

He L, Liu Q, Quan M, Zhou DN, Zhou YQ, Zhao JL (2009) Molecular cloning and phylogenetic analysis of Babesia orientalis heat shock protein 70. Vet Parasitol 162:183–191. https://doi.org/10.1016/j.vetpar.2009.03.039

He L et al (2014) Mitochondrial genome of Babesia orientalis, apicomplexan parasite of water buffalo (Bubalus babalis, Linnaeus, 1758) endemic in China. Parasites & Vectors 7:82. https://doi.org/10.1186/1756-3305-7-82

Article   Google Scholar  

He L, Liu Q, Yao B, Zhou Y, Hu M, Fang R, Zhao J (2017) A historical overview of research on Babesia orientalis, a protozoan parasite infecting water buffalo. Front Microbiol 8:1323. https://doi.org/10.3389/fmicb.2017.01323

He L, Bastos RG, Sun Y, Hua G, Guan G, Zhao J, Suarez CE (2021) Babesiosis as a potential threat for bovine production in China. Parasites & Vectors 14:460. https://doi.org/10.1186/s13071-021-04948-3

Hisaeda H, Yasutomo K, Himeno K (2005) Malaria: immune evasion by parasites. Int J Biochem Cell Biol 37:700–706. https://doi.org/10.1016/j.biocel.2004.10.009

Kanoi BN et al (2017) Antibody profiles to wheat germ cell-free system synthesized Plasmodium falciparum proteins correlate with protection from symptomatic malaria in Uganda. Vaccine 35:873–881. https://doi.org/10.1016/j.vaccine.2017.01.001

Koh CY et al (2015) A binding hotspot in Trypanosoma cruzi histidyl-tRNA synthetase revealed by fragment-based crystallographic cocktail screens. Acta Crystallogr D Biol Crystallogr 71:1684–1698. https://doi.org/10.1107/s1399004715007683

Koichiro T, Glen S, Sudhir KJMB, Evolution (2021) MEGA11: molecular evolutionary genetics analysis version 11. Mol Biol Evol 38(7):3022–3027

Liu Z, Zhao J, Ma L, Yao B (1997) Babesia orientalis sp. nov. parasitized in buffalo Bubalus bubalis in China (Piroplasmida: Babesiidae). Acta Veterinaria et Zootechnica Sinica 28:84–89

Google Scholar  

Liu Q, Zhao JL, Zhou YQ, Liu EY, Yao BA, Fu Y (2005) Study on some molecular characterization of Babesia orientalis. Vet Parasitol 130:191–198. https://doi.org/10.1016/j.vetpar.2005.03.021

Liu M et al (2021) Tafenoquine is a promising drug candidate for the treatment of babesiosis. Antimicrob Agents Chemother 65:e0020421. https://doi.org/10.1128/aac.00204-21

Ma L, Liu Z, Zhao J (1989) An investigation of water buffalo babesiosis in Hubei province V. adult Rhipicephalus haemaphysaloides transmits the parasites transovarially. Chin J Anim Vet Sci 1:67–70

Najm R et al (2023) Invasion of Toxoplasma gondii bradyzoites: molecular dissection of the moving junction proteins and effective vaccination targets. Proc Natl Acad Sci U S A 120:e2219533120. https://doi.org/10.1073/pnas.2219533120

Nozaki M, Baba M, Tachibana M, Tokunaga N, Torii M, Ishino T (2020) Detection of the rhoptry neck protein complex in Plasmodium sporozoites and its contribution to sporozoite invasion of salivary glands. mSphere 5(4):e00325. https://doi.org/10.1128/mSphere.00325-20

Pulido-Quevedo FA, Arévalo-Pinzón G, Castañeda-Ramírez JJ, Barreto-Santamaría A, Patarroyo ME, Patarroyo MA (2023) Plasmodium falciparum rhoptry neck protein 4 has conserved regions mediating interactions with receptors on human erythrocytes and hepatocyte membrane. Int J Med Microbiol 313:151579. https://doi.org/10.1016/j.ijmm.2023.151579

Ramasamy R (1998) Molecular basis for evasion of host immunity and pathogenesis in malaria. Biochem Biophys Acta 1406:10–27. https://doi.org/10.1016/s0925-4439(97)00078-1

Richard D et al (2010) Interaction between Plasmodium falciparum apical membrane antigen 1 and the rhoptry neck protein complex defines a key step in the erythrocyte invasion process of malaria parasites. J Biol Chem 285:14815–14822. https://doi.org/10.1074/jbc.M109.080770

Richards JS et al (2013) Identification and prioritization of merozoite antigens as targets of protective human immunity to Plasmodium falciparum malaria for vaccine and biomarker development. J Immunol 191:795–809. https://doi.org/10.4049/jimmunol.1300778

Schnittger L, Rodriguez AE, Florin-Christensen M, Morrison DA (2012) Babesia: a world emerging Infection, genetics and evolution. J Mol Epidemiol Evol Genet Infects Dis 12:1788–1809. https://doi.org/10.1016/j.meegid.2012.07.004

Srinivasan P et al (2013) Disrupting malaria parasite AMA1-RON2 interaction with a small molecule prevents erythrocyte invasion. Nat Commun 4:2261. https://doi.org/10.1038/ncomms3261

Tian Y et al (2020) Identification and characterizations of a rhoptries neck protein 5 (BoRON5) in Babesia orientalis. Parasitol Int 77:102106. https://doi.org/10.1016/j.parint.2020.102106

Tokunaga N et al (2019) Expression and localization profiles of rhoptry proteins in Plasmodium berghei sporozoites. Front Cell Infect Microbiol 9:316. https://doi.org/10.3389/fcimb.2019.00316

Verra F, Hughes AL (1999) Biased amino acid composition in repeat regions of Plasmodium antigens. Mol Biol Evol 16:627–633. https://doi.org/10.1093/oxfordjournals.molbev.a026145

Vetrivel U, Nagarajan H (2018) Deciphering ophthalmic adaptive inhibitors targeting RON4 of Toxoplasma gondii: an integrative in silico approach. Life Sci 213:82–93. https://doi.org/10.1016/j.lfs.2018.10.022

Wu Y, Kang Z, Tian Z, Wu M, Wang J (2017) Biosynthesis and characterization of recombinant silk-like polypeptides derived from the heavy chain of silk fibrion. Polymers 9:669

Yang J, Zhang Y (2015) I-TASSER server: new development for protein structure and function predictions. Nucleic Acids Res 43:W174-181. https://doi.org/10.1093/nar/gkv342

Zhang Y (2009) I-TASSER: fully automated protein structure prediction in CASP8. Proteins 77(Suppl 9):100–113. https://doi.org/10.1002/prot.22588

Zhao J, Yao B, Ma L, Liu Z (1994) Continuous in vitro cultivation of Babesia orientalis collected from infected water buffalo II screening and use of frozen serum Chin. J Vet Sci 2:118–120

Zhao J, Liu Z, Yao B (1997) The effects of medium volume, depth and replacing medium interval on the growth of Babesia bovis in vitro. J-huazhong Agric Univ 16:63–66

Zhao JL, Liu ZL, Yao BA, Ma LH (2002) Culture-derived Babesia orientalis exoantigens used as a vaccine against buffalo babesiosis. Parasitol Res 88:S38-40. https://doi.org/10.1007/s00436-001-0569-0

Download references

This study was supported by grant 31930108 from the National Natural Science Foundation of China, grant 2022YFD1800200 from the National Key Research and Development Program of China, and the Fundamental Research Funds for the Central Universities in China (Project 2662020DKPY016 and 2262022DKYJ001).

Author information

Fangjie Li and Jiaying Guo did the equal work. Fangjie Li was responsible for the whole process of experimental design and experimental operation, and Jiaying Guo was responsible for the experimental scheme design.

Authors and Affiliations

State Key Laboratory of Agricultural Microbiology, College of Veterinary Medicine, Huazhong Agricultural University, Wuhan, 430070, Hubei, China

Fangjie Li, Sen Wang, Zhen Han, Zheng Nie, Long Yu, Xiang Shu, Yingjun Xia, Lan He & Junlong Zhao

Key Laboratory of Preventive Veterinary Medicine in Hubei Province, The Cooperative Innovation Center for Sustainable Pig Production, Wuhan, 430070, Hubei, China

Key Laboratory of Development of Veterinary Diagnostic Products, Ministry of Agriculture of the People’s Republic of China, Wuhan, 430070, Hubei, China

Lan He & Junlong Zhao

Heilongjiang Provincial Key Laboratory of Zoonosis, College of Veterinary Medicine, Northeast Agricultural University, Harbin, Heilongjiang, China

Jiaying Guo

You can also search for this author in PubMed   Google Scholar

Contributions

FL, LH, and JZ designed the study and drafted the manuscript. FL and JG performed the experiments. ZH and SW participated in the data analysis. All authors read and approved the final manuscript.

Corresponding author

Correspondence to Junlong Zhao .

Ethics declarations

Ethics approval.

Animal experiments were conducted in strict compliance with the Guidelines for the Care and Use of Laboratory Animals of the Ministry of Science and Technology of China. The above experiments were approved by the Hubei Laboratory Animal Research Center and the Ethics Committee of Huazhong Agricultural University (license number: HZAUMO-2019–016).

Consent to participate

Not applicable.

Consent for publication

All authors gave their consent before submitting this work.

Competing interests

The authors declare no competing interests.

Additional information

Section Editor: Aysegul Taylan Ozkan.

Publisher's Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Supplementary Information

Below is the link to the electronic supplementary material.

Supplementary file1 (DOCX 126 KB)

Rights and permissions.

Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.

Reprints and permissions

About this article

Li, F., Guo, J., Wang, S. et al. Identification and molecular characterization of a novel Babesia orientalis rhoptry neck protein 4 (BoRON4). Parasitol Res 123 , 310 (2024). https://doi.org/10.1007/s00436-024-08328-5

Download citation

Received : 24 February 2024

Accepted : 14 August 2024

Published : 29 August 2024

DOI : https://doi.org/10.1007/s00436-024-08328-5

Share this article

Anyone you share the following link with will be able to read this content:

Sorry, a shareable link is not currently available for this article.

Provided by the Springer Nature SharedIt content-sharing initiative

• Babesia orientalis
• Rhoptry neck protein 4
• Immunogenicity
• Find a journal
• Publish with us
• Track your research