Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Decoding Cognitive Health Using Machine Learning: A Comprehensive Evaluation for Diagnosis of Significant Memory Concern

M. Sajid, Rahul Sharma, Iman Beheshti, M. Tanveer

TL;DR

This work tackles early detection of Significant Memory Concern (SMC) by conducting a thorough state-of-the-art review and a large-scale empirical evaluation of two ML families, Randomized Neural Networks (RNNs) and Hyperplane-based Classifiers (HbCs), using ADNI2 baseline sMRI features. Features are drawn from gray matter, white matter, Jacobian determinant, and cortical thickness, totaling 273 GM/WM/JD features and 68 CT features after processing with CAT12 and Brainnetome/DKT atlases. The study identifies dRVFL and edRVFL as the top-performing RNNs, and Pin-GTSVM-K as the leading HbC, with all-feature ensembles and WM/CT-specific insights guiding model choice. SHAP analyses provide interpretable feature contribution maps, highlighting regions such as the cerebellum, primary auditory cortex, and frontal areas, thereby offering clinically relevant biomarkers. The results, supported by rigorous statistical tests, demonstrate the framework’s potential to assist automated, accurate SMC assessment and pave the way for broader, multi-modal cognitive health applications, with code and data publicly accessible.

Abstract

The timely identification of significant memory concern (SMC) is crucial for proactive cognitive health management, especially in an aging population. Detecting SMC early enables timely intervention and personalized care, potentially slowing cognitive disorder progression. This study presents a state-of-the-art review followed by a comprehensive evaluation of machine learning models within the randomized neural networks (RNNs) and hyperplane-based classifiers (HbCs) family to investigate SMC diagnosis thoroughly. Utilizing the Alzheimer's Disease Neuroimaging Initiative 2 (ADNI2) dataset, 111 individuals with SMC and 111 healthy older adults are analyzed based on T1W magnetic resonance imaging (MRI) scans, extracting rich features. This analysis is based on baseline structural MRI (sMRI) scans, extracting rich features from gray matter (GM), white matter (WM), Jacobian determinant (JD), and cortical thickness (CT) measurements. In RNNs, deep random vector functional link (dRVFL) and ensemble dRVFL (edRVFL) emerge as the best classifiers in terms of performance metrics in the identification of SMC. In HbCs, Kernelized pinball general twin support vector machine (Pin-GTSVM-K) excels in CT and WM features, whereas Linear Pin-GTSVM (Pin-GTSVM-L) and Linear intuitionistic fuzzy TSVM (IFTSVM-L) performs well in the JD and GM features sets, respectively. This comprehensive evaluation emphasizes the critical role of feature selection and model choice in attaining an effective classifier for SMC diagnosis. The inclusion of statistical analyses further reinforces the credibility of the results, affirming the rigor of this analysis. The performance measures exhibit the suitability of this framework in aiding researchers with the automated and accurate assessment of SMC. The source codes of the algorithms and datasets used in this study are available at https://github.com/mtanveer1/SMC.

Decoding Cognitive Health Using Machine Learning: A Comprehensive Evaluation for Diagnosis of Significant Memory Concern

TL;DR

This work tackles early detection of Significant Memory Concern (SMC) by conducting a thorough state-of-the-art review and a large-scale empirical evaluation of two ML families, Randomized Neural Networks (RNNs) and Hyperplane-based Classifiers (HbCs), using ADNI2 baseline sMRI features. Features are drawn from gray matter, white matter, Jacobian determinant, and cortical thickness, totaling 273 GM/WM/JD features and 68 CT features after processing with CAT12 and Brainnetome/DKT atlases. The study identifies dRVFL and edRVFL as the top-performing RNNs, and Pin-GTSVM-K as the leading HbC, with all-feature ensembles and WM/CT-specific insights guiding model choice. SHAP analyses provide interpretable feature contribution maps, highlighting regions such as the cerebellum, primary auditory cortex, and frontal areas, thereby offering clinically relevant biomarkers. The results, supported by rigorous statistical tests, demonstrate the framework’s potential to assist automated, accurate SMC assessment and pave the way for broader, multi-modal cognitive health applications, with code and data publicly accessible.

Abstract

The timely identification of significant memory concern (SMC) is crucial for proactive cognitive health management, especially in an aging population. Detecting SMC early enables timely intervention and personalized care, potentially slowing cognitive disorder progression. This study presents a state-of-the-art review followed by a comprehensive evaluation of machine learning models within the randomized neural networks (RNNs) and hyperplane-based classifiers (HbCs) family to investigate SMC diagnosis thoroughly. Utilizing the Alzheimer's Disease Neuroimaging Initiative 2 (ADNI2) dataset, 111 individuals with SMC and 111 healthy older adults are analyzed based on T1W magnetic resonance imaging (MRI) scans, extracting rich features. This analysis is based on baseline structural MRI (sMRI) scans, extracting rich features from gray matter (GM), white matter (WM), Jacobian determinant (JD), and cortical thickness (CT) measurements. In RNNs, deep random vector functional link (dRVFL) and ensemble dRVFL (edRVFL) emerge as the best classifiers in terms of performance metrics in the identification of SMC. In HbCs, Kernelized pinball general twin support vector machine (Pin-GTSVM-K) excels in CT and WM features, whereas Linear Pin-GTSVM (Pin-GTSVM-L) and Linear intuitionistic fuzzy TSVM (IFTSVM-L) performs well in the JD and GM features sets, respectively. This comprehensive evaluation emphasizes the critical role of feature selection and model choice in attaining an effective classifier for SMC diagnosis. The inclusion of statistical analyses further reinforces the credibility of the results, affirming the rigor of this analysis. The performance measures exhibit the suitability of this framework in aiding researchers with the automated and accurate assessment of SMC. The source codes of the algorithms and datasets used in this study are available at https://github.com/mtanveer1/SMC.
Paper Structure (22 sections, 1 equation, 3 figures, 13 tables)

This paper contains 22 sections, 1 equation, 3 figures, 13 tables.

Table of Contents

  1. Introduction
  2. State-of-the-art Review
  3. Literature selection
  4. Literature survey
  5. Data Acquisition and Preprocessing
  6. Data Acquisition
  7. Data Preprocessing
  8. Classification Models
  9. Randomized Neural Networks (RNNs)
  10. Rationale behind behind selecting the RNN classifiers
  11. Hyperplane-based Classifiers (HbCs)
  12. Rationale behind behind selecting the HbCs
  13. Experimental Evaluation and Discussion
  14. Experimental Setup
  15. Implementation Details and Experimental Setup
  16. ...and 7 more sections

Figures (3)

  • Figure 1: Flow diagram of the proposed comprehensive evaluation approach.
  • Figure 2: Confusion matrix for the top performing model of RNN and HbC, i.e., dRVL and Pin-GTSVM-K respectively.
  • Figure 3: Top five feature visualization for (a) GM, (b) WM, (c) JD, and (d) CT modalities identified from Shapley for dRVFL.