Unlocking Neural Transparency: Jacobian Maps for Explainable AI in Alzheimer's Detection
By: Yasmine Mustafa, Mohamed Elmahallawy, Tie Luo
Potential Business Impact:
Helps doctors find Alzheimer's earlier and trust results.
Alzheimer's disease (AD) leads to progressive cognitive decline, making early detection crucial for effective intervention. While deep learning models have shown high accuracy in AD diagnosis, their lack of interpretability limits clinical trust and adoption. This paper introduces a novel pre-model approach leveraging Jacobian Maps (JMs) within a multi-modal framework to enhance explainability and trustworthiness in AD detection. By capturing localized brain volume changes, JMs establish meaningful correlations between model predictions and well-known neuroanatomical biomarkers of AD. We validate JMs through experiments comparing a 3D CNN trained on JMs versus on traditional preprocessed data, which demonstrates superior accuracy. We also employ 3D Grad-CAM analysis to provide both visual and quantitative insights, further showcasing improved interpretability and diagnostic reliability.
Similar Papers
Cross-Attention Fusion of MRI and Jacobian Maps for Alzheimer's Disease Diagnosis
Image and Video Processing
Finds Alzheimer's early using brain scans.
Enhancing Alzheimer's Diagnosis: Leveraging Anatomical Landmarks in Graph Convolutional Neural Networks on Tetrahedral Meshes
Image and Video Processing
Finds Alzheimer's risk using brain scans, not costly tests.
Deep Learning Approaches with Explainable AI for Differentiating Alzheimer Disease and Mild Cognitive Impairment
Image and Video Processing
Finds Alzheimer's early using brain scans.