Alzheimer’s Disease Classification in Functional MRI With 4D Joint Temporal-Spatial Kernels in Novel 4D CNN Model

Motivation: Previous works in the literature apply 3D spatial-only models on 4D functional MRI data leading to possible sub-par feature extraction to be used for downstream tasks like classification. Goal(s): In this work, we aim to develop a novel 4D convolution network to extract 4D joint temporal-spatial kernels that not only learn spatial information but in addition also capture temporal dynamics. Approach: We apply our novel approach on the ADNI dataset with data augmentations such as circular time shifting to enforce time-invariant results. Results: Experimental results show promising performance in capturing spatial-temporal data in functional MRI compared to 3D models. Impact: The 4D CNN model improves Alzheimer's disease diagnosis for rs-fMRI data, enabling earlier detection and better interventions. Future research could explore task-based fMRI applications and regression tasks, enhancing understanding of cognitive performance and disease progression.