Explainable harmonized cortical graph neural networks for risk prediction of Alzheimer’s disease

Research conducted during my internship at Wake Forest University, Center for Artificial Intelligence Research.

This paper extends our prior work on harmonized cortical GNNs by incorporating more comprehensive evaluation methods to examine model performance for prediction of Alzheimer’s disease risk. The work aims to make deep learning models more transparent and clinically actionable for healthcare practitioners.

Recommended citation: Liu, I., & Ma, D. (2026). Explainable harmonized cortical graph neural networks for risk prediction of Alzheimer's disease. Paper submitted to SPIE Medical Imaging 2026.