A Deep Learning Prediction Model for Predicting Alzheimer'S Disease And Quantifying Asymmetric Degeneration of the Hippocampus

Lateral asymmetry in the hippocampal volume and shape and its association with Alzheimer’s disease (AD) have been studied extensively and deep learning (DL) models built on the hippocampal MRI data have demonstrated promising performance for early prediction of AD. However, it remains unclear if the left and right hippocampal MRI data contain information with different prediction power and how the bilateral hippocampal information can be integrated for improving the AD prediction. To address these issues, we propose an instance based DL model to learn informative features from the hippocampal MRI data, regularized by an autoencoder, for predicting the AD conversions in a time-to-event prediction modeling framework, quantify lateral asymmetry in the hippocampal prediction power of AD conversion, and identify the optimal strategy to integrate the bilateral hippocampal MRI data for predicting AD. Experimental results on MRI scans of 1307 subjects (817 for training and 490 for validation) have demonstrated that the left hippocampus can better predict AD than the right hippocampus, and an integration of the bilateral hippocampal data with the instance based DL method improved AD prediction, compared with alternative predictive modeling strategies.