Quantitative hippocampal subfield volumetry using AID-HS in mesial temporal lobe epilepsy: correlation with

Nguyen Duy Hung1, Nguyen Thu Minh Chau2, Tran Dinh Van3

  • 1Department of Radiology, Hanoi Medical University, Ha Noi, Viet Nam; Department of Radiology, Viet Duc Hospital, Ha Noi, Viet Nam.

Abstract

Insights

The AID-HS pipeline accurately distinguishes hippocampal sclerosis (HS) subtypes in mesial temporal lobe epilepsy (mTLE) using MRI volumetric data. This tool aids in classifying HS types, with CA3 showing the highest diagnostic accuracy.

Area of Science:

  • Neuroimaging
  • Epilepsy research
  • Histopathology

Background:

  • Mesial temporal lobe epilepsy (mTLE) is often associated with hippocampal sclerosis (HS).
  • Accurate classification of HS subtypes is crucial for understanding disease mechanisms and guiding treatment.
  • Current diagnostic methods may have limitations in precisely characterizing HS subtypes.

Purpose of the Study:

  • To assess the diagnostic accuracy of the automated AID-HS pipeline for characterizing hippocampal subfield volumes.
  • To differentiate between typical HS type 1 and atypical HS (types 2 and 3) using volumetric data.
  • To validate the AID-HS pipeline against histopathological criteria in mTLE patients.

Main Methods:

  • Thirty mTLE patients undergoing surgery had 3.0 T MRI scans.
  • The AID-HS tool performed automated volumetric subfield analysis.
  • Hippocampal subfield volumes were compared between HS types, validated by 2013 ILAE histopathological criteria.

Main Results:

  • HS type 1 was the most common (83.3%).
  • Volume reduction in CA2 and CA3 was more frequent in typical HS type 1 compared to atypical HS.
  • MRI-detected volume reduction in CA2, CA3, and CA4 correlated significantly with histopathological sclerosis (p < 0.01).
  • The AID-HS pipeline demonstrated fair to good accuracy (AUC 0.73–0.83) in distinguishing HS subtypes.
  • The CA3 subfield achieved the highest accuracy (AUC 0.83) with 100% sensitivity.

Conclusions:

  • The AID-HS pipeline offers quantitative support for evaluating HS subtypes.
  • This automated tool may enhance the diagnostic process for HS in mTLE.
  • Further research can explore the clinical utility of AID-HS in epilepsy management.