Classification of Alzheimer’s Disease Using Maximal Information Coefficient-Based Functional Connectivity with an Extreme Learning Machine

Alzheimer’s disease (AD) is a progressive chronic illness that leads to cognitive decline and dementia. Neuroimaging technologies, such as functional magnetic resonance imaging (fMRI), deep learning approaches offer promising avenues for AD classification. In this study, we investigate the use of fMRI-based connectivity (FC) measures, including Pearson correlation coefficient (PCC), maximal information (MIC), extended (eMIC), combined with extreme machines (ELM) Our findings demonstrate employing non-linear techniques, MIC eMIC, features classification yields accurate results. Specifically, eMIC-based achieve high accuracy 94% classifying cognitively normal (CN) mild impairment (MCI) individuals, outperforming PCC (81%) (85%). For MCI classification, achieves higher compared (58%) eMIC (78%). CN exhibits best 95% (90%) (87%). These results underscore effectiveness derived from techniques in accurately differentiating individuals emphasizing potential neuroimaging machine methods improving diagnosis