Multimodal Prediction of Alzheimer's Disease Severity Level Based on Resting-State EEG and Structural MRI

While several biomarkers have been developed for the detection of Alzheimer's disease (AD), not many are available prediction severity, particularly patients in mild stages AD. In this paper, we explore multimodal Mini-Mental State Examination (MMSE) scores using resting-state electroencephalography (EEG) and structural magnetic resonance imaging (MRI) scans. Analyses were carried out on a dataset comprised EEG MRI data collected from 89 diagnosed with minimal-mild Three feature selection algorithms assessed alongside four machine learning algorithms. Results showed that while features alone outperformed features, when both modalities combined, improved results achieved. The top-selected conveyed information about amplitude modulation rate-of-change, whereas top-MRI cortical area white matter volume. Overall, root mean square error between predicted MMSE values true 1.682 was achieved system random forest regression model.