SVM-based Recursive Feature Elimination Research Articles

ADHD classification with cross-dataset feature selection for biomarker consistency detection

Objective. Attention deficit hyperactivity disorder (ADHD) is a prevalent neurodevelopmental disorder in children. While numerous intelligent methods are applied for its subjective diagnosis, they seldom consider the consistency problem of ADHD biomarkers. In practice, these data-driven approaches lead to varying learned features for ADHD classification across diverse ADHD datasets. This phenomenon significantly undermines the reliability of identified biomarkers and hampers the interpretability of these methods. Approach. In this study, we propose a cross-dataset feature selection (FS) module using a grouped SVM-based recursive feature elimination approach (G-SVM-RFE) to enhance biomarker consistency across multiple datasets. Additionally, we employ connectome gradient data for ADHD classification. In details, we introduce the G-SVM-RFE method to effectively concentrate gradient components within a few brain regions, thereby increasing the likelihood of identifying these regions as ADHD biomarkers. The cross-dataset FS module is integrated into an existing binary hypothesis testing (BHT) framework. This module utilizes external datasets to identify global regions that yield stable biomarkers. Meanwhile, given a dataset which waits for implementing the classification task as local dataset, we learn its own specific regions to further improve the performance of accuracy on this dataset. Main results. By employing this module, our experiments achieve an average accuracy of 96.7% on diverse datasets. Importantly, the discriminative gradient components primarily originate from the global regions, providing evidence for the significance of these regions. We further identify regions with the high appearance frequencies as biomarkers, where all the used global regions and one local region are recognized. Significance. These biomarkers align with existing research on impaired brain regions in children with ADHD. Thus, our method demonstrates its validity by providing enhanced biological explanations derived from ADHD mechanisms.

Heterogeneous fusion of biometric and deep physiological features for accurate porcine cough recognition.

Accurate identification of porcine cough plays a vital role in comprehensive respiratory health monitoring and diagnosis of pigs. It serves as a fundamental prerequisite for stress-free animal health management, reducing pig mortality rates, and improving the economic efficiency of the farming industry. Creating a representative multi-source signal signature for porcine cough is a crucial step toward automating its identification. To this end, a feature fusion method that combines the biological features extracted from the acoustic source segment with the deep physiological features derived from thermal source images is proposed in the paper. First, acoustic features from various domains are extracted from the sound source signals. To determine the most effective combination of sound source features, an SVM-based recursive feature elimination cross-validation algorithm (SVM-RFECV) is employed. Second, a shallow convolutional neural network (named ThermographicNet) is constructed to extract deep physiological features from the thermal source images. Finally, the two heterogeneous features are integrated at an early stage and input into a support vector machine (SVM) for porcine cough recognition. Through rigorous experimentation, the performance of the proposed fusion approach is evaluated, achieving an impressive accuracy of 98.79% in recognizing porcine cough. These results further underscore the effectiveness of combining acoustic source features with heterogeneous deep thermal source features, thereby establishing a robust feature representation for porcine cough recognition.

HISET: Hybrid interpretable strategies with ensemble techniques for respiratory sound classification

The human respiratory systems can be affected by several diseases and it is associated with distinctive sounds. For advanced biomedical signal processing, one of the most complex issues is automated respiratory sound classification. In this research, five Hybrid Interpretable Strategies with Ensemble Techniques (HISET) which are quite interesting and robust are proposed for the purpose of respiratory sounds classification. The first approach is termed as an Ensemble GSSR technique which utilizes L2 Granger Analysis and the proposed Supportive Ensemble Empirical Mode Decomposition (SEEMD) technique and then Support Vector Machine based Recursive Feature Elimination (SVM-RFE) is used for feature selection and followed by classification with Machine Learning (ML) classifiers. The second approach proposed is the implementation of a novel Realm Revamping Sparse Representation Classification (RR-SRC) technique and third approach proposed is a Distance Metric dependent Variational Mode Decomposition (DM-VMD) with Extreme Learning Machine (ELM) classification process. The fourth approach proposed is with the usage of Harris Hawks Optimization (HHO) with a Scaling Factor based Pliable Differential Evolution (SFPDE) algorithm termed as HHO-SFPDE and it is classified with ML classifiers. The fifth or the final approach proposed analyzes the application of dimensionality reduction techniques with the proposed Gray Wolf Optimization based Support Vector Classification (GWO-SVC) and another parallel approach utilizes a similar kind of analysis with the Grasshopper Optimization Algorithm (GOA) based Sparse Autoencoder. The results are examined for ICBHI dataset and the best results are shown for the 2-class classification when the analysis is carried out with Manhattan distance-based VMD-ELM reporting an accuracy of 95.39%, and for 3-class classification Euclidean distance-based VMD-ELM reported an accuracy of 90.61% and for 4-class classification, Manhattan distance-based VMD-ELM reported an accuracy of 89.27%.

Distinguishing Patients with MRI-Negative Temporal Lobe Epilepsy from Normal Controls Based on Individual Morphological Brain Network.

Temporal Lobe Epilepsy (TLE) is the most common subtype of focal epilepsy and the most refractory to drug treatment. Roughly 30% of patients do not have easily identifiable structural abnormalities. In other words, MRI-negative TLE has normal MRI scans on visual inspection. Thus, MRI-negative TLE is a diagnostic and therapeutic challenge. In this study, we investigate the cortical morphological brain network to identify MRI-negative TLE. The 210 cortical ROIs based on the Brainnetome atlas were used to define the network nodes. The least absolute shrinkage and selection operator (LASSO) algorithm and Pearson correlation methods were used to calculate the inter-regional morphometric features vector correlation respectively. As a result, two types of networks were constructed. The topological characteristics of networks were calculated by graph theory. Then after, a two-stage feature selection strategy, including a two-sample t-test and support vector machine-based recursive feature elimination (SVM-RFE), was performed in feature selection. Finally, classification with support vector machine (SVM) and leave-one-out cross-validation (LOOCV) was employed for the training and evaluation of the classifiers. The performance of two constructed brain networks was compared in MRI-negative TLE classification. The results indicated that the LASSO algorithm achieved better performance than the Pearson pairwise correlation method. The LASSO algorithm provides a robust method of individual morphological network construction for distinguishing patients with MRI-negative TLE from normal controls.

An efficient model selection for linear discriminant function-based recursive feature elimination

Model selection is an important issue in support vector machine-based recursive feature elimination (SVM-RFE). However, performing model selection on a linear SVM-RFE is difficult because the generalization error of SVM-RFE is hard to estimate. This paper proposes an approximation method to evaluate the generalization error of a linear SVM-RFE, and designs a new criterion to tune the penalty parameter C. As the computational cost of the proposed algorithm is expensive, several alpha seeding approaches are proposed to reduce the computational complexity. We show that the performance of the proposed algorithm exceeds that of the compared algorithms on bioinformatics datasets, and empirically demonstrate the computational time saving achieved by alpha seeding approaches.

Elaboration of a Radiomics Strategy for the Prediction of the Re-positive Cases in the Discharged Patients With COVID-19.

Objective: A considerable part of COVID-19 patients were found to be re-positive in the SARS-CoV-2 RT-PCR test after discharge. Early prediction of re-positive COVID-19 cases is of critical importance in determining the isolation period and developing clinical protocols.Materials and Methods: Ninety-one patients discharged from Wanzhou Three Gorges Central Hospital, Chongqing, China, from February 10, 2020 to March 3, 2020 were administered nasopharyngeal swab SARS-CoV-2 tests within 12–14 days, and 50 eligible patients (32 male and 18 female) with completed data were enrolled. Average age was 48 ± 11.5 years. All patients underwent non-enhanced chest CT on admission. A total of 568 radiomics features were extracted from the CT images, and 17 clinical factors were collected based on the medical record. Student's t-test and support vector machine–based recursive feature elimination (SVM-RFE) method were used to determine an optimal subset of features for the discriminative model development.Results: After Student's t-test, 62 radiomics features showed significant inter-group differences (p < 0.05) between the re-positive and negative cases, and none of the clinical features showed significant differences. These significant features were further selected by SVM-RFE algorithm, and a more compact feature subset containing only two radiomics features was finally determined, achieving the best predictive performance with the accuracy and area under the curve of 72.6% and 0.773 for the identification of the re-positive case.Conclusion: The proposed radiomics method has preliminarily shown potential in identifying the re-positive cases among the recovered COVID-19 patients after discharge. More strategies are to be integrated into the current pipeline to improve its precision, and a larger database with multi-clinical enrollment is required to extensively verify its performance.

An efficient alpha seeding method for optimized extreme learning machine-based feature selection algorithm

Embedded feature selection algorithms, such as support vector machine based recursive feature elimination (SVM-RFE), have proven to be effective for many real applications. However, due to the model selection problem, SVM-RFE naturally suffers from a heavy computational burden as well as high computational complexity. To solve these issues, this paper proposes using an optimized extreme learning machine (OELM) model instead of SVM. This model, referred to as OELM-RFE provides an efficient active set solver for training the OELM algorithm. We also present an effective alpha seeding algorithm to efficiently solve successive quadratic programming (QP) problems inherent in OELM. One of the salient characteristics of OELM-RFE is that it has only one tuning parameter: the penalty constant C. Experimental results from work on benchmark datasets show that OELM-RFE tends to have higher prediction accuracy than SVM-RFE, and requires fewer model selection efforts. In addition, the alpha seeding method works better on more datasets.

Differentiating the learning styles of college students in different disciplines in a college English blended learning setting

Learning styles are critical to educational psychology, especially when investigating various contextual factors that interact with individual learning styles. Drawing upon Biglan's taxonomy of academic tribes, this study systematically analyzed the learning styles of 790 sophomores in a blended learning course with 46 specializations using a novel machine learning algorithm called the support vector machine (SVM). Moreover, an SVM-based recursive feature elimination (SVM-RFE) technique was integrated to identify the differential features among distinct disciplines. The findings of this study shed light on the optimal feature sets that collectively determined students' discipline-specific learning styles in a college blended learning setting.

Diagnosis of Alzheimer's disease using universum support vector machine based recursive feature elimination (USVM-RFE)

Alzheimer's disease is one of the most common causes of death in today's world. Magnetic resonance imaging (MRI) provides an efficient and non-invasive approach for diagnosis of Alzheimer's disease. Efficient feature extraction techniques are needed for accurate classification of MRI images. Motivated by the work on support vector machine based recursive feature elimination (SVM-RFE) [16], we propose a novel feature selection technique to incorporate prior information about data distribution in the recursive feature elimination process. Our method is termed as universum support vector machine based recursive feature elimination (USVM-RFE). The proposed method provides global information about data in the RFE process as compared to the local approach of feature selection in SVM-RFE. We also present the application of feature selection and classification algorithms on both voxel based as well as volume based morphometry analysis of structural MRI images (ADNI database). Feature selection is performed using MRI data of brain tissues such as gray matter, white matter, and cerebrospinal fluid. USVM-RFE provides improvement over SVM-RFE in classification of control normal (CN), mild cognitive impairment (MCI), and Alzheimer's disease (AD) subjects. Moreover, better accuracy is obtained by USVM-RFE with lesser number of features in comparison to SVM-RFE. This leads to identification of prominent brain regions for feature selection and classification of MRI images. The highest accuracies obtained by our method for classification of CN vs AD, CN vs MCI, and MCI vs AD are 100%, 90%, and 73.68%, respectively.

Deep Learning-Based Approach for Civil Aircraft Hazard Identification and Prediction

Safety is an eternal issue in the civil aviation transportation. Once a civil aviation accident occurs, it will cause great casualties and economic losses. In order to ensure the civil aviation safety, the hazard identification and prediction of civil aircraft should be effectively and accurately realized. The civil aircraft uses Aircraft Communications Addressing and Reporting System (ACARS) to interact with the ground during flight. The data generated by ACARS has a simple structure and strong timeliness. In view of the advantages of ACARS data, a hazard identification and prediction method based on support vector machine optimized by particle swarm optimization (PSO-SVM) and long short term memory (LSTM) neural networks which uses ACARS report as analysis data is proposed. First, in order to reduce the identification and prediction time cost, the SVM-based recursive feature elimination method with cross-validation algorithm (SVM-RFECV) is used to select the characteristic parameters. Then, the SVM optimized by PSO is used to identify hazard based on the selected parameters. According to the identification results, the LSTM is used to predict the trend of the selected parameters to realize hazard prediction. An A13 report of APU generated by ACARS is selected as analysis data for hazard identification and predication in this paper. The analysis results show that the proposed identification method based on PSO-SVM and SVM-RFECV has high identification speed and accuracy. The proposed prediction method based on LSTM has the best prediction performance. The proposed method can effectively identify hazards and accurately predict the trend of parameters to improve the safety of aircraft.

Integration of dynamic contrast-enhanced magnetic resonance imaging and T2-weighted imaging radiomic features by a canonical correlation analysis-based feature fusion method to predict histological grade in ductal breast carcinoma

Tumour histological grade has prognostic implications in breast cancer. Tumour features in dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) and T2-weighted (T2W) imaging can provide related and complementary information in the analysis of breast lesions to improve MRI-based histological status prediction in breast cancer. A dataset of 167 patients with invasive ductal carcinoma (IDC) was assembled, consisting of 72 low/intermediate-grade and 95 high-grade cases with preoperative DCE-MRI and T2W images. The data cohort was separated into development (n = 111) and validation (n = 56) cohorts. Each tumour was segmented in the precontrast and the intermediate and last postcontrast DCE-MR images and was mapped to the tumour in the T2W images. Radiomic features, including texture, morphology, and histogram distribution features in the tumour image, were extracted for those image series. Features from the DCE-MR and T2W images were fused by a canonical correlation analysis (CCA)-based method. The support vector machine (SVM) classifiers were trained and tested on the development and validation cohorts, respectively. SVM-based recursive feature elimination (SVM-RFE) was adopted to identify the optimal features for prediction. The areas under the ROC curves (AUCs) for the T2W images and the DCE-MRI series of precontrast, intermediate and last postcontrast images were 0.750 ± 0.047, 0.749 ± 0.047, and 0.788 ± 0.045, respectively, for the development cohort and 0.715 ± 0.068, 0.704 ± 0.073, and 0.744 ± 0.067, respectively, for the validation cohort. After the CCA-based fusion of features from the DCE-MRI series and T2W images, the AUCs increased to 0.751 ± 0.065, 0.803 ± 0.0600 and 794 ± 0.060 in the validation cohort. Moreover, the method of fusing features between DCE-MRI and T2W images using CCA achieved better performance than the concatenation-based feature fusion or classifier fusion methods. Our results demonstrated that anatomical and functional MR images yield complementary information, and feature fusion of radiomic features by matrix transformation to optimize their correlations produced a classifier with improved performance for predicting the histological grade of IDC.

An Exploration of Impact Factors Influencing Students’ Reading Literacy in Singapore with Machine Learning Approaches

This study identified the contextual factors which differentiated 15-year-old students with high- and low-achieving reading literacy in Singapore based on Program for International Student Assessment (PISA) 2015. 4,015 students from Singapore were collected from the public dataset of PISA 2015, with 2,646 high-achieving students and 1,369 low-achieving students in PISA reading literacy test. The impact of the overall 49 contextual factors on reading literacy was analyzed in three levels: student level, family level and school level. Support vector machine (SVM), a machine learning approach, was applied to analyze these contextual features. It indicated that SVM could effectively distinguish these two cohorts of readers with an accuracy score of 0.78. SVM-based recursive feature elimination (SVM-RFE), another machine learning approach, was then applied to rank these selected features. These features were outputted in descending order with regard to the degree of their significance to the differentiation. At last, an optimal set with 15 contextual factors was selected by RFE-CV (cross validation), which collectively affected the differentiation of students with high- and low-level of reading literacy. Based on the analysis, implications to further improving students&amp;rsquo; reading literacy can be achieved.

Detection of premature ventricular contraction (PVC) using linear and nonlinear techniques: an experimental study

Cardiovascular diseases are identified as one of the most dangerous diseases and the major causes of the death around the world. One of the most common cardiac arrhythmias that have always been a concern for cardiologists is premature ventricular contractions. Regarding its abundance among all ages, prediction and diagnosis of this type of arrhythmia has particular importance. One of the most common, most non-invasive; and the least costly method for investigation of heart diseases is to record and analyze the electrocardiogram (ECG) signals. The purpose of this study is to analyze the ECG in order to classify premature ventricular contraction heartbeats. Having proposed a new technique based on evolutionary optimization for R peak detection, several methodologies, such as morphological assessment, polynomial curve fitting, discrete wavelet transform, and nonlinear analysis, are employed to extract features from ECG signal. Support vector machine (SVM) classifier with a linear kernel is used to detect the normal and PVC heart rates. In order to evaluate the proposed method, in addition to the MIT-BIH database, the experimental data is used and the methodology performance is proved for both databases. Finally, using different feature selection criteria such as fisher distinction, minimal-redundancy maximum-relevance, and SVM-based recursive feature elimination with correlation bias reduction, six features are introduced as best ones. The proposed PVC detection algorithm acquires the overall detection accuracy of 99.78%, with the sensitivity of 99.91% and specificity of 99.37%, for MIT-BIH dataset.

Assessment of Optimal Pedagogical Factors for Canadian ESL Learners’ Reading Literacy Through Artificial Intelligence Algorithms

The current study explored the effective pedagogical factors that distinguish high-achieving from low-achieving ESL (English as a second language) primary school learners in reading literacy in Canada. In total, 203 samples (167 high-achieving learners and 36 low-achieving learners from 128 primary schools) in the 4th grade were drawn from the public database of Progress in International Reading Literacy Study (PIRLS) 2016, which is the benchmark for large-scale assessments of reading literacy targeting fourth-grade students. For the first time in the ESL-related research, this study applied an artificial intelligence approach, support vector machine (SVM), to concurrently analyze 41 pedagogical factors associated with reading materials, classroom organization, reading strategies, in-class reading activities and post-reading activities. The overall 41 factors collectively distinguished the high-achieving readers from the low-achieving readers with a high accuracy score (0.793) via SVM. Then, these 41 factors were ranked according to their contribution to the SVM model through SVM-based recursive feature elimination (SVM-RFE). Eventually, an optimal factor set was selected by the SVM-RFE cross validation, which contained 10 effective pedagogical factors centered on reading materials, reading strategies and in-class reading activities for fourth-grade high-achieving ESL learners in reading literacy. Suggestions based on solid data analysis would facilitate infrastructural and pedagogical improvements in ESL reading education.

3D-Deep Learning Based Automatic Diagnosis of Alzheimer's Disease with Joint MMSE Prediction Using Resting-State fMRI.

We performed this research to 1) evaluate a novel deep learning method for the diagnosis of Alzheimer's disease (AD) and 2) jointly predict the Mini Mental State Examination (MMSE) scores of South Korean patients with AD. Using resting-state functional Magnetic Resonance Imaging (rs-fMRI) scans of 331 participants, we obtained functional 3-dimensional (3-D) independent component spatial maps for use as features in classification and regression tasks. A 3-D convolutional neural network (CNN) architecture was developed for the classification task. MMSE scores were predicted using: linear least square regression (LLSR), support vector regression, bagging-based ensemble regression, and tree regression with group independent component analysis (gICA) features. To improve MMSE regression performance, we applied feature optimization methods including least absolute shrinkage and selection operator and support vector machine-based recursive feature elimination (SVM-RFE). The mean balanced test accuracy was 85.27% for the classification of AD versus healthy controls. The medial visual, default mode, dorsal attention, executive, and auditory related networks were mainly associated with AD. The maximum clinical MMSE score prediction accuracy with the LLSR method applied on gICA combined with SVM-RFE features had the lowest root mean square error (3.27 ± 0.58) and the highest R2 value (0.63 ± 0.02). Classification of AD and healthy controls can be successfully achieved using only rs-fMRI and MMSE scores can be accurately predicted using functional independent component features. In the absence of trained clinicians, AD disease status and clinical MMSE scores can be jointly predicted using 3-D deep learning and regression learning approaches with rs-fMRI data.

Radiomics Strategy for Molecular Subtype Stratification of Lower-Grade Glioma: Detecting IDH and TP53 Mutations Based on Multimodal MRI.

Noninvasive detection of isocitrate dehydrogenase (IDH) and TP53 mutations are meaningful for molecular stratification of lower-grade gliomas (LrGG). To explore potential MRI features reflecting IDH and TP53 mutations of LrGG, and propose a radiomics strategy for detecting them. Retrospective, radiomics. A total of 103 LrGG patients were separated into development (n = 73) and validation (n = 30) cohorts. T1 -weighted (before and after contrast-enhanced), T2 -weighted, and fluid-attenuation inversion recovery images from 1.5T (n = 37) or 3T (n = 66) scanners. After data preprocessing, high-throughput features were derived from patients' volumes of interests of different sequences. The support vector machine-based recursive feature elimination (SVM-RFE) was adopted to find the optimal features for IDH and TP53 mutation detection. SVM models were trained and tested on development and validation cohort. The commonly used metric was used for assessing the efficiency. One-way analysis of variance (ANOVA), chi-square, or Fisher's exact test were applied on clinical characteristics to confirm whether significant differences exist between three molecular subtypes decided by IDH and TP53 status. Intraclass correlation coefficients were calculated to assess the robustness of the radiomics features. The constituent ratio of histopathologic subtypes was significantly different among three molecular subtypes (P = 0.017). SVM models for detecting IDH and TP53 mutation were established using 12 and 22 optimal features selected by SVM-RFE. The accuracies and area under the curves for IDH and TP53 mutations on the development cohort were 84.9%, 0.830, and 92.0%, 0.949, while on the validation cohort were 80.0%, 0.792, and 85.0%, 0.869, respectively. Furthermore, the stratified accuracies of three subtypes were 72.8%, 71.9%, and 70%, respectively. Using a radiomics approach integrating SVM model and multimodal MRI features, molecular subtype stratification of LGG patients was implemented through detecting IDH and TP53 mutations. The results suggested that the proposed approach has promising detecting efficiency and T2 -weighted image features are more important than features from other images. 3 Technical Efficacy: Stage 2 J. Magn. Reson. Imaging 2018;48:916-926.

Automated discrimination of dementia spectrum disorders using extreme learning machine and structural T1 MRI features.

The classification of neuroimaging data for the diagnosis of Alzheimer's Disease (AD) is one of the main research goals of the neuroscience and clinical fields. In this study, we performed extreme learning machine (ELM) classifier to discriminate the AD, mild cognitive impairment (MCI) from normal control (NC). We compared the performance of ELM with that of a linear kernel support vector machine (SVM) for 718 structural MRI images from Alzheimer's Disease Neuroimaging Initiative (ADNI) database. The data consisted of normal control, MCI converter (MCI-C), MCI non-converter (MCI-NC), and AD. We employed SVM-based recursive feature elimination (RFE-SVM) algorithm to find the optimal subset of features. In this study, we found that the RFE-SVM feature selection approach in combination with ELM shows the superior classification accuracy to that of linear kernel SVM for structural T1 MRI data.

A machine learning approach for the identification of key markers involved in brain development from single-cell transcriptomic data.

BackgroundThe ability to sequence the transcriptomes of single cells using single-cell RNA-seq sequencing technologies presents a shift in the scientific paradigm where scientists, now, are able to concurrently investigate the complex biology of a heterogeneous population of cells, one at a time. However, till date, there has not been a suitable computational methodology for the analysis of such intricate deluge of data, in particular techniques which will aid the identification of the unique transcriptomic profiles difference between the different cellular subtypes. In this paper, we describe the novel methodology for the analysis of single-cell RNA-seq data, obtained from neocortical cells and neural progenitor cells, using machine learning algorithms (Support Vector machine (SVM) and Random Forest (RF)).ResultsThirty-eight key transcripts were identified, using the SVM-based recursive feature elimination (SVM-RFE) method of feature selection, to best differentiate developing neocortical cells from neural progenitor cells in the SVM and RF classifiers built. Also, these genes possessed a higher discriminative power (enhanced prediction accuracy) as compared commonly used statistical techniques or geneset-based approaches. Further downstream network reconstruction analysis was carried out to unravel hidden general regulatory networks where novel interactions could be further validated in web-lab experimentation and be useful candidates to be targeted for the treatment of neuronal developmental diseases.ConclusionThis novel approach reported for is able to identify transcripts, with reported neuronal involvement, which optimally differentiate neocortical cells and neural progenitor cells. It is believed to be extensible and applicable to other single-cell RNA-seq expression profiles like that of the study of the cancer progression and treatment within a highly heterogeneous tumour.

Eye tracking data guided feature selection for image classification

Feature selection has played a critical role in image classification, since it is able to remove irrelevant and redundant features and to eventually reduce the dimensionality of feature space. Although existing feature selection methods have achieved promising progress, human factors have seldom been taken into account. To tackle such a problem, a novel two-stage feature selection method is proposed for image classification by taking human factors into account and leveraging the value of eye tracking data. In the coarse selection stage, with the help of eye tracking data, Regions of Interests (ROIs) from the human perspective are first identified to represent an image with visual features. Then, with an improved quantum genetic algorithm (IQGA) that incorporates a novel mutation strategy for alleviating the premature convergence, a subset of features is obtained for the subsequent fine selection. In the fine selection stage, a hybrid method is proposed to integrate the efficiency of the minimal-Redundancy-Maximal-Relevance (mRMR) and the effectiveness of the Support Vector Machine based Recursive Feature Elimination (SVM-RFE). In particular, the ranking criterion of the SVM-RFE is improved by incorporating the ranking information obtained from the mRMR. Comprehensive experimental results in two benchmark datasets demonstrate that eye tracking data are of great importance to improve the performance of feature selection for image classification.

Multiclass Classification for the Differential Diagnosis on the ADHD Subtypes Using Recursive Feature Elimination and Hierarchical Extreme Learning Machine: Structural MRI Study.

The classification of neuroimaging data for the diagnosis of certain brain diseases is one of the main research goals of the neuroscience and clinical communities. In this study, we performed multiclass classification using a hierarchical extreme learning machine (H-ELM) classifier. We compared the performance of this classifier with that of a support vector machine (SVM) and basic extreme learning machine (ELM) for cortical MRI data from attention deficit/hyperactivity disorder (ADHD) patients. We used 159 structural MRI images of children from the publicly available ADHD-200 MRI dataset. The data consisted of three types, namely, typically developing (TDC), ADHD-inattentive (ADHD-I), and ADHD-combined (ADHD-C). We carried out feature selection by using standard SVM-based recursive feature elimination (RFE-SVM) that enabled us to achieve good classification accuracy (60.78%). In this study, we found the RFE-SVM feature selection approach in combination with H-ELM to effectively enable the acquisition of high multiclass classification accuracy rates for structural neuroimaging data. In addition, we found that the most important features for classification were the surface area of the superior frontal lobe, and the cortical thickness, volume, and mean surface area of the whole cortex.