Prognosis of Alzheimer's Disease using Machine Learning and The Virtual Brain Simulations

Overview

According to WHO, more than 55 million people are living with dementia worldwide, with Alzheimer’s Disease (AD) representing up to 70 percent of those cases. AD is currently the 6th leading cause of death in the US but no cure exists. Machine Learning (ML) can augment diagnostics of dementia in several ways. The engineering goal is the design of a computational brain network model using The Virtual Brain (TVB) simulation platform augmenting the ML models and multi-modal neuro-imaging to improve early detection of AD. The combination of empirical neuro-imaging features and simulated Local Field Potentials (LFPs) from the TVB should outperform the classification accuracy of empirical features alone by at least 5% and the early prediction accuracy using the combined empirical and simulated features should be at least 70%. Publicly available dataset from the Open Access Series of Imaging Studies (OASIS) consisting of both cross-sectional and longitudinal MRI and PET data was used. The empirical features extracted from the PET and MRI data of subjects in OASIS combined with features from frequency compositions of TVB-simulated LFPs were provided to a Support Vector Machine Classifier. Augmenting with the TVB-simulated features outperformed the classification accuracy of empirical alone by about 4.86% (F1-score of empirical 67.06% vs. combined 71.92%). Results show that TVB’s ability to consider Amyloid β distributions from connectivity-based brain simulation can improve the ML-driven diagnostics classification of Alzheimer’s. The simulation-augmented classification model needs to be tested for clinical usability in a larger cohort with other imaging databases.

Prognosis of Alzheimer's Disease using Machine Learning and The Virtual Brain Simulations

Overview

Source coverage

Awards (1)

Competition history

Resources

Related projects

Predicting the Future: Using Machine Learning to Forecast the Progression of Alzheimer's Disease

Early Diagnosis of Alzheimer's Disease Using Machine Learning

Alzheimer's Disease Prediction

Alzheimer's Recognition Using Artificial Intelligence

Improving Alzheimer's Disease Diagnosis Using AI-Based Artificial Neural Networks

Beyond Scans: Predicting Alzheimer's Through MRI and Handwriting Analysis

Developing a Pre-Risk Assessment Incorporating Machine-Learning and Biomarkers to Diagnose Alzheimer's Disease

Early Classification of Alzheimer's Disease Using Machine Learning

Browse more like this