Related Experiment Video
Updated: Sep 30, 2025
A Machine Learning Approach to Design an Efficient Selective Screening of Mild Cognitive Impairment
Published on: January 11, 2020
Searching for optimal machine learning model to classify mild cognitive impairment (MCI) subtypes using multimodal
Tatsuya Jitsuishi1, Atsushi Yamaguchi2
1Department of Functional Anatomy, Graduate School of Medicine, Chiba University, 1-8-1 Inohana, Chuo-ku, Chiba, 260-8670, Japan.
Abstract:
The intervention at the stage of mild cognitive impairment (MCI) is promising for preventing Alzheimer's disease (AD). This study aims to search for the optimal machine learning (ML) model to classify early and late MCI (EMCI and LMCI) subtypes using multimodal MRI data. First, the tract-based spatial statistics (TBSS) analyses showed LMCI-related white matter changes in the Corpus Callosum. The ROI-based tractography addressed the connected cortical areas by affected callosal fibers. We then prepared two feature subsets for ML by measuring resting-state functional connectivity (TBSS-RSFC method) and graph theory metrics (TBSS-Graph method) in these cortical areas, respectively. We also prepared feature subsets of diffusion parameters in the regions of LMCI-related white matter alterations detected by TBSS analyses. Using these feature subsets, we trained and tested multiple ML models for EMCI/LMCI classification with cross-validation. Our results showed the ensemble ML model (AdaBoost) with feature subset of diffusion parameters achieved better performance of mean accuracy 70%. The useful brain regions for classification were those, including frontal, parietal lobe, Corpus Callosum, cingulate regions, insula, and thalamus regions. Our findings indicated the optimal ML model using diffusion parameters might be effective to distinguish LMCI from EMCI subjects at the prodromal stage of AD.
Insights
Machine learning models can distinguish early from late mild cognitive impairment (MCI) using MRI data. An AdaBoost model with diffusion parameters showed 70% accuracy, identifying key brain regions for early Alzheimer's disease detection.
Area of Science:
- Neuroimaging
- Machine Learning
- Neurology
Background:
- Early intervention for mild cognitive impairment (MCI) may prevent Alzheimer's disease (AD).
- Distinguishing between early MCI (EMCI) and late MCI (LMCI) is crucial for targeted interventions.
- Multimodal MRI data offers potential for identifying subtle differences between MCI subtypes.
Purpose of the Study:
- To identify the optimal machine learning (ML) model for classifying EMCI and LMCI subtypes.
- To utilize multimodal MRI data, including diffusion parameters and functional connectivity, for classification.
- To pinpoint specific brain regions and features indicative of LMCI.
Main Methods:
- Tract-based spatial statistics (TBSS) identified white matter changes in the Corpus Callosum associated with LMCI.
- ROI-based tractography mapped connected cortical areas affected by callosal fiber alterations.
- Feature subsets were created using resting-state functional connectivity (TBSS-RSFC), graph theory metrics (TBSS-Graph), and diffusion parameters.
- Multiple ML models were trained and tested using cross-validation for EMCI/LMCI classification.
Main Results:
- The ensemble AdaBoost ML model, using diffusion parameters, achieved the highest performance with a mean accuracy of 70%.
- Key brain regions for classification included the frontal lobe, parietal lobe, Corpus Callosum, cingulate regions, insula, and thalamus.
- White matter changes in the Corpus Callosum were significant in LMCI patients.
Conclusions:
- The optimal ML model utilizing diffusion parameters can effectively differentiate LMCI from EMCI subjects.
- This approach holds promise for identifying individuals at the prodromal stage of Alzheimer's disease.
- Identifying specific brain regions and features aids in understanding the progression of MCI.
