Explainsegnet: Interpretable Segmentation for Alzheimer's Diagnosis

Juntao Su, Syed Muhammad Anwar, Fang Jin

Abstract: This study focuses on enhancing the accuracy of predicting Alzheimer's disease, mild cognitive impairment, and normal control cases. We leverage the Alzheimer's Disease Neuroimaging Initiative (ADNI) dataset- comprising 3D structural magnetic resonance imaging brain scans, for our method development and evaluation. Our analysis utilizes a residual 3D convolutional neural network (equipped with 4 attention layers). The core contributions of our work are twofold: firstly, we have innovatively integrated clinical expertise into the initialization of the attention layer's weights through whole-brain segmentation technique; secondly, we have employed various state-of-the-art model interpretation techniques. These techniques effectively annotate influential brain regions and demonstrate promising results within neuroimaging analysis, as reflected in the key metrics and outcomes. Our findings underscore the potential of deep learning in neuroimaging, especially highlighting the critical role of comprehensive brain segmentation in enhancing diagnostic accuracy.