Ensemble support vector machine classification of dementia using structural MRI and mini-mental state examination

Flood susceptibility mapping using a novel ensemble weights-of-evidence and support vector machine models in GIS

Support Vector Machines (SVMs) ensembles have been widely used to improve classification accuracy in complicated pattern recognition tasks. In this work we propose to apply an ensemble of SVMs coupled with feature-subset selection methods to aleviate the curse of dimensionality associated with expression-based classification of DNA microarray data. We compare the single SVM classifier to SVM ensembles applying two different feature-subset selection techniques, namely random selection and k-means clustering, the base classifiers are combined using either majority vote or SVM fusion. Two real-world benchmarks datasets are used to evaluate and compare the performance. Experimental results show that the SVM ensemble of SVM base classifiers using k-means clustering for feature-subset selection and employing an SVM combiner achieve the best classification accuracy, and that feature-subset-selection methods can have considerable impact on the classification accuracy.

Cancer recognition with bagged ensembles of support vector machines

This paper describes the experiment of analyzing MRI images of the brain of patients with Alzheimer's disease (AD), mild cognitive impairment (MCI) and normal control (NC). The aim of this research is to use new methods of extracting features from MRI images and selecting the most relevant features. The values of cortical thickness in regions of interest are used as a features for MRI images. The Support Vector Machine — Recursive Feature Elimination, Infinite Feature Selection, Relief-F, Feature Selection Via Concave Minimization and Support Vector Machine, Minimum Redundancy Maximum Relevance, Fisher, Unsupervised Discriminative Feature Selection, L0-norm Based Feature Selection, and Feature Selection via Eigenvector Centrality techniques for feature selection and ranking were used. In the experiment, the MRI images from ADNI database was used, which contains data from 539 subjects. In the result, the following maximum values of area under ROC curves for three datasets are obtained: NC-AD — 0.90, NC-MCI — 0.71, MCI-AD — 0.69. Classification accuracy with linear Support Vector Machine for the corresponding datasets: NC-AD — 80%, NC-MCI — 64%, MCI-AD — 65%. The results of this paper can be used for building a classifier in space of selected features.

Objective To predict different progression of Alzheimer' s disease (AD) by using support vector machine (SVM) model based on brain magnetic resonance imaging (MRI) data. Methods A total of 271 MRI data of 543 subjects from the American Alzheimer' s Disease Neuroimaging Initiative (ANDI) database were included in the study. The Mini Mental State Examination (MMSE) scores, age, gender, and years of education were included as the indicators in the study. The useful characteristics were extracted and combined with SVM model to establish the comprehensive predictive model. The different characteristics among groups were used to predict different progression of AD. Results The accuracy of 54 different characteristics based on SVM comprehensive model in predicting different AD progression was 97.53%. Conclusion The SVM prediction model established in this study and the characteristics closely related to AD can obtain accurate prediction, providing accurate evidence for clinical and basic research, investigation of etiology and changes of pathology, as well as the accurate location of different brain regions, which justifies widespread use in prediction of other disease progression. Key words: Alzheimer' s disease; Magnetic resonance spectroscopy; Mild cognitive impairment; Support vector machine

In order to develop Human-centric Driver Fatigue Monitoring Systems (HDFMS) with aims to increase driving safety, an efficient combined features extraction from Curvelet transform and Gabor wavelet transform for fatigue expressions descriptions of vehicle drivers is proposed, and Random Subspace Ensemble (RSE) of Support Vector Machines (SVMs) with polynomial kernel as the base classifier is then exploited for classification of three predefined fatigue expressions classes, namely, awake expressions, moderate fatigue expressions, and severe fatigue expressions. The results of holdout and cross-validation experiments show that CF by RSE of SVMs with polynomial kernel outperforms other seven classifiers, i.e., Curvelet features by SVMs classifier, Gabor features by SVMs classifier, CF by five individual SVMs classifiers. With CF and RSE of SVMs with polynomial kernel, the classification accuracies of drivers' fatigue expressions are over 90% in both of the holdout and cross-validation experiments, which show the proposed approach of combined features extraction and RSE of SVMs can be used for developing Human-centric Driver Fatigue Monitoring Systems to increase driving safety.

In this paper, the algorithm design of the support vector machines (SVMs) ensemble in a practical multi-stage framework is analyzed which can be implemented efficiently by evolutionary multi-objective optimization algorithm. The designing of SVMs ensemble is considered in three stages: first, the bootstrap method and a strategy of dynamical parameter range adjustment are used to generate more diverse base SVMs, and the NSGA-II algorithm which can efficiently tune the parameters of SVMs is applied to ensure the accuracy of base SVMs; Second, the NSGA-II algorithm is used again to select the member of ensemble based on the accuracy and diversity of the ensemble we have measured; Last, the reliability of different class is computed and combined to decide the outputs of ensemble in terms of the decision values of base SVMs. The proposed algorithm is applied to the UCI datasets, some useful results has been concluded for the future work in this field.

A movie review can be considered as an essential data source for both movie producers and potential consumers. The reviews are usually used as a benchmark for knowing the movie’s quality and are utilized as references to whether the film is worth watching. The study of sentiment analysis has received more attention from researchers. This current study aims to classify the sentiment analysis of movie reviews obtained from the IMDb site. The support vector machine (SVM) method was employed to classify the movie review’s sentiments. Meanwhile, the radial basis function (RBF) kernel and information gain (IG) were used to enhance classification. Feature selection was conducted by removing irrelevant features and selecting features with a strong correlation to classification. The IG algorithm was used for feature selection. The current study revealed that the classification accuracy of movie review sentiment analysis using the SVM algorithm, SVM + RBF kernel, and SVM+RBF Kernel+IG are 81.50%, 82.25%, and 87.25%, respectively.

Magnetic resonance imaging (MRI) is increasingly used in the diagnosis of Alzheimer's disease (AD) in order to identify abnormalities in the brain. Indeed, cortical atrophy, a powerful biomarker for AD, can be detected using structural MRI (sMRI), but it cannot detect impairment in the integrity of the white matter (WM) preceding cortical atrophy. The early detection of these changes is made possible by the novel MRI modality known as diffusion tensor imaging (DTI). In this study, we integrate DTI and sMRI as complementary imaging modalities for the early detection of AD in order to create an effective computer‐assisted diagnosis tool. The fused Bag‐of‐Features (BoF) with Speeded‐Up Robust Features (SURF) and modified AlexNet convolutional neural network (CNN) are utilized to extract local and deep features. This is applied to DTI scalar metrics (fractional anisotropy and diffusivity metric) and segmented gray matter images from T1‐weighted MRI images. Then, the classification of local unimodal and deep multimodal features is first performed using support vector machine (SVM) classifiers. Then, the majority voting technique is adopted to predict the final decision from the ensemble SVMs. The study is directed toward the classification of AD versus mild cognitive impairment (MCI) versus cognitively normal (CN) subjects. Our proposed method achieved an accuracy of 98.42% and demonstrated the robustness of multimodality imaging fusion.

IC-P2-125: Functional and structural MRI measures of cognitive progression in Alzheimer's disease

Background Few previous cohorts have studied the different type of physical activities and the degree of cognitive decline. The objective of this work was to analyze the leisure, domestic and professional activities with mild and moderate cognitive disorders in older people living in community. Methods The study used data from the longitudinal and observational study, FrèLE (FRagility: Longitudinal Study of Expressions). The collected data included: socio-demographic variables, lifestyle, and health status (frailty, comorbidities, cognitive status, depression). Cognitive decline was assessed by using: MMSE (Mini-Mental State Examination) and MoCA (Montreal Cognitive Assessment). MoCA was used with two cut-offs (26 and 17) so as to define mild and moderate cognitive disorders Physical activity was assessed by the PASE (Physical Activity Scale for the Elderly), structured in three sections: leisure, domestic and professional activities. Spline and proportional hazards regression models (Cox) were used to estimate the risk of cognitive disorders. Results At baseline, 1623 participants were included and the prevalence of cognitive disorders was 6.9% (MMSE) and 7.2% (MoCA), mild cognitive disorders was 71.3%. The mean age was 77 years, and 52% of the participants were women. After a 2 years long follow-up, we found 6.9% (MMSE) and 6% (MoCA) cognitive disorders on participants. Analyses showed that domestic activities were associated to cognitive decline (HR = 0.52 [0.28-0.94] for MMSE and HR = 0.48 [0.28-0.80] for MoCA). No association were found with leisure and professional activities, and no spline were significant with mild cognitive disorders. Conclusions Analysis showed a relationship between cognitive disorders and type of physical activity, thanks to the use of specific questionnaire of elderly and two global test of cognition. These findings will contribute to the debate on the beneficial effects of physical activity on cognition. Key messages This work allowed to compare two test of cognition and their link with physical activity. It contributes to the debate on the beneficial effects of physical activity on cognition. The work allowed us to see the effect of the different types of physical activity and the impact of the statistical method on the results.

Breast cancer poses a critical global health challenge and continues to be one of the most prevalent causes of cancer-related deaths among women worldwide. Accurate and early classification of cancer severity is essential for improving treatment outcomes and guiding clinical decision-making, since timely intervention can significantly reduce mortality rates and enhance patient survival. This study evaluates the performance of Support Vector Machine (SVM) models using different kernel functions of Linear, Polynomial, Radial Basis Function (RBF), and Sigmoid for breast cancer severity prediction. The impact of feature selection was also examined, using the Random Forest algorithm to select the top features based on Mean Decrease Accuracy (MDA), which serves to reduce redundancy, improve interpretability, and enhance model efficiency. Experimental results show that the RBF kernel consistently outperformed other kernels, especially in terms of sensitivity, a critical metric in medical diagnostics that emphasizes the ability of the model to identify positive cases correctly. Without feature selection, the RBF kernel achieved an accuracy of 0.9744, a sensitivity of 0.9772, a precision of 0.9722, and an AUC of 0.9968, indicating strong performance across all evaluation metrics. After applying feature selection, the RBF kernel further improved the accuracy to 0.9754, the sensitivity to 0.9770, the precision to 0.9742, and the AUC to 0.9975, which demonstrated enhanced generalization and reduced overfitting, highlighting the benefits of targeted feature reduction. While the Polynomial kernel yielded the highest precision (up to 0.9799), its lower sensitivity (as low as 0.9237) indicates a greater risk of false negatives, which is particularly concerning in cancer detection. These findings underscore the importance of optimizing both kernel function and feature selection. The RBF kernel, when combined with targeted feature selection, offers the most balanced and sensitive model, making it highly suitable for breast cancer classification tasks where diagnostic accuracy is vital

d-glutamate, which is involved in N-methyl-d-aspartate receptor modulation, may be associated with cognitive ageing. This study aimed to use peripheral plasma d-glutamate levels to differentiate patients with mild cognitive impairment (MCI) and Alzheimer's disease (AD) from healthy individuals and to evaluate its prediction ability using machine learning. Overall, 31 healthy controls, 21 patients with MCI and 133 patients with AD were recruited. Serum d-glutamate levels were measured using high-performance liquid chromatography (HPLC). Cognitive deficit severity was assessed using the Clinical Dementia Rating scale and the Mini-Mental Status Examination (MMSE). We employed four machine learning algorithms (support vector machine, logistic regression, random forest and naïve Bayes) to build an optimal predictive model to distinguish patients with MCI or AD from healthy controls. The MCI and AD groups had lower plasma d-glutamate levels (1097.79 ± 283.99 and 785.10 ± 720.06 ng/mL, respectively) compared to healthy controls (1620.08 ± 548.80 ng/mL). The naïve Bayes model and random forest model appeared to be the best models for determining MCI and AD susceptibility, respectively (area under the receiver operating characteristic curve: 0.8207 and 0.7900; sensitivity: 0.8438 and 0.6997; and specificity: 0.8158 and 0.9188, respectively). The total MMSE score was positively correlated with d-glutamate levels (r = 0.368, p < 0.001). Multivariate regression analysis indicated that d-glutamate levels were significantly associated with the total MMSE score (B = 0.003, 95% confidence interval 0.002-0.005, p < 0.001). Peripheral plasma d-glutamate levels were associated with cognitive impairment and may therefore be a suitable peripheral biomarker for detecting MCI and AD. Rapid and cost-effective HPLC for biomarkers and machine learning algorithms may assist physicians in diagnosing MCI and AD in outpatient clinics.

Mild cognitive impairment (MCI) is a prodromal stage of Alzheimer's disease (AD), where neurodegeneration is not as considerable, thereby potentially increasing the effect of treatments. Therefore, highly sensitive and specific classification of subjects with MCI is necessary, where various MRI modalities have displayed promise. Structural, diffusion, and resting-state (RS) functional MRI analyses were performed on the AD (n = 26), MCI (n = 5), and healthy control (HC) (n = 14) group. Structural analysis was performed via voxel-based morphometry (VBM) and volumetric subcortical segmentation analysis. Fractional anisotropy and mean diffusivity were estimated during the diffusion analysis. RS analysis investigated seed-based functional connectivity. Classification via support vector machine was performed to evaluate which MRI modality most accurately differentiated the groups. Multiple linear regression was conducted to evaluate the MRI modalities correlation with clinical assessment scores. Classification of MCI and HC displayed highest accuracy based on diffusion MRI, which besides demonstrated high correlation with clinical scores. Classification was equally accurate in AD, when using VBM or diffusion tensor imaging measures. Yet, more variance was explained by VBM measures in the clinical assessment scores of the AD group. This study highlights the potential of diffusion MRI in differentiating MCI from HC and AD. However, the results need to be interpreted with caution as sample size and artifacts in the MRI data probably influenced the results.

Structural magnetic resonance imaging (sMRI) and functional MRI (fMRI) based features toward identification of mild cognitive impairment (MCI) status has gained popularity due to its non-invasiveness allowing repeated measurements. Despite of this great potential, however an effort to integrate the sMRI- and fMRI-based features to increase MCI detection accuracy has been limited. In this study, we were motivated to investigate whether the detection capability of the MCI status can be improved via the integration of feature sets from sMRI and fMRI data. The characteristic traits of regional volumes and level of neuronal activity of the brain associated with the MCI in comparison to healthy control were exploited using sMRI and fMRI data, respectively, in which these characteristic traits (i.e., biomarkers) were identified from group comparison via two-sample t-test. In the subsequent classification phase, the MCI status were automatically detected using a support vector machine (SVM) algorithm employing the identified sMRI- and fMRI-driven biomarkers as input features vectors. The results indicate that the fMRI-based biomarkers appear to increase the detection accuracy of the MCI status than the sMRI-based biomarkers. Moreover, the integrated feature sets using the sMRI- and fMRI-based biomarkers constantly showed superior performance than the feature sets based on the fMRI-driven biomarkers. This study successfully demonstrated an anecdotal evidence that the integration of the sMRI and fMRI modalities can provide a supplemental information toward diagnosis of the MCI status compared to either the sMRI or fMRI modality.