Related Experiment Video
Updated: Jun 13, 2025
A Machine Learning Approach to Design an Efficient Selective Screening of Mild Cognitive Impairment
Published on: January 11, 2020
Prediction of future dementia among patients with mild cognitive impairment (MCI) by integrating multimodal clinical
Andrew Cirincione1, Kirsten Lynch2, Jamie Bennett1
1Department of Genetics and Genomic Sciences, Icahn School of Medicine at Mount Sinai, 1 Gustave L. Levy Pl, New York, NY, 10029, USA.
Abstract:
Efficiently and objectively analyzing the complex, diverse multimodal data collected from patients at risk for dementia can be difficult in the clinical setting, contributing to high rates of underdiagnosis or misdiagnosis of this serious disorder. Patients with mild cognitive impairment (MCI) are especially at risk of developing dementia in the future. This study evaluated the ability of multi-modal machine learning (ML) methods, especially the Ensemble Integration (EI) framework, to predict future dementia development among patients with MCI. EI is a machine learning framework designed to leverage complementarity and consensus in multimodal data, which may not be adequately captured by methods used by prior dementia-related prediction studies. We tested EI's ability to predict future dementia development among MCI patients using multimodal clinical and imaging data, such as neuroanatomical measurements from structural magnetic resonance imaging (MRI) and positron emission tomography (PET) scans, from The Alzheimer's Disease Prediction of Longitudinal Evolution (TADPOLE) challenge. For predicting future dementia development among MCI patients, on a held out test set, the EI-based model performed better (AUC = 0.81, F-measure = 0.68) than the more commonly used XGBoost (AUC = 0.68, F-measure = 0.57) and deep learning (AUC = 0.79, F-measure = 0.61) approaches. This EI-based model also suggested MRI-derived volumes of regions in the middle temporal gyrus, posterior cingulate gyrus and inferior lateral ventricle brain regions to be predictive of progression to dementia.
Insights
A new Ensemble Integration (EI) machine learning model accurately predicts dementia in mild cognitive impairment (MCI) patients. This AI approach outperforms others, identifying key brain regions linked to dementia progression.
Area of Science:
- Neuroscience
- Artificial Intelligence
- Medical Diagnostics
Background:
- Analyzing complex multimodal patient data for dementia risk is challenging, leading to misdiagnosis.
- Mild cognitive impairment (MCI) is a significant risk factor for future dementia development.
- Existing prediction models may not fully utilize the potential of diverse data types.
Purpose of the Study:
- To evaluate the efficacy of multimodal machine learning (ML), specifically the Ensemble Integration (EI) framework, in predicting dementia development in MCI patients.
- To compare the performance of the EI framework against traditional ML methods like XGBoost and deep learning.
- To identify specific neuroimaging biomarkers associated with dementia progression.
Main Methods:
- Utilized the Ensemble Integration (EI) framework, a novel ML approach designed for multimodal data analysis.
- Employed multimodal clinical and neuroimaging data (MRI, PET) from The Alzheimer's Disease Prediction of Longitudinal Evolution (TADPOLE) challenge.
- Tested EI's predictive performance on a held-out dataset of MCI patients.
Main Results:
- The EI-based model achieved superior predictive performance (AUC = 0.81, F-measure = 0.68) compared to XGBoost (AUC = 0.68, F-measure = 0.57) and deep learning (AUC = 0.79, F-measure = 0.61).
- The EI model identified specific brain regions, including the middle temporal gyrus, posterior cingulate gyrus, and inferior lateral ventricle, as predictive of dementia progression.
- The EI framework demonstrated an enhanced ability to leverage data complementarity and consensus.
Conclusions:
- The Ensemble Integration (EI) framework offers a powerful and accurate method for predicting dementia development in MCI patients using multimodal data.
- EI surpasses conventional ML and deep learning models in this predictive task.
- Neuroimaging features from specific brain regions are crucial indicators for dementia risk assessment in MCI populations.
Related Concept Videos
01:30Dementia
The progression of dementia is generally gradual....
01:26Alzheimer's Disease: Overview
The clinical diagnosis of AD hinges on the presence of memory and other cognitive impairments. Biomarkers, such as changes in Aβ...