Automated Quantification of Lateral and Medial Temporal Lobe Volumes for Improved Diagnosis of Early Alzheimer’s Disease

Abstract

The purpose of this study was to evaluate the importance of automated lateral and medial temporal volume measurement technique for the early diagnosis of Alzheimer's disease (AD). A cross-sectional T1-weighted magnetic resonance image was obtained from 39 healthy adults and 39 patients with mild AD. The study demonstrates significant volume loss in the lateral temporal lobe (LTL) and medial temporal lobe (MTL) regions of the brain in early cases of AD, suggesting that volume loss could be used as a viable biomarker for mild AD diagnosis. Using a deep learning-based auto-segmentation network (CINet), the study accurately estimates the volumes of various LTL and MTL brain regions. Notably, higher volume loss is observed in the left MTL and LTL regions compared to the right, indicating an asymmetric impact in mild AD. The study underscores the significance of automated technique for AD diagnosis and monitoring disease progression, contributing valuable insights for potential early interventions.