Personalized Morphological Brain Network Analysis for Temporal Lobe Epilepsy: Combining Cortical Thickness and Volume...

Area of Science:

• Neuroscience
• Medical Imaging
• Epilepsy Research

Background:

• Temporal lobe epilepsy (TLE) is a common, drug-resistant epilepsy type, often viewed as a brain network disorder.
• Current TLE diagnosis relies heavily on T1-weighted MRI morphological changes like cortical thickness and volume atrophy.
• Existing methods often neglect crucial inter-regional brain interactions.

Purpose of the Study:

• To develop novel biomarkers for TLE by integrating morphological and network features.
• To enhance diagnostic accuracy for TLE using a combined approach.
• To explore the utility of the Morphological Inverse Divergence (MIND) method in TLE network construction.

Main Methods:

• Constructed personalized brain networks using the Morphological Inverse Divergence (MIND) method, integrating cortical thickness and volume data.
• Analyzed both morphological and network-based features for TLE biomarker identification.
• Evaluated the diagnostic performance of morphological biomarkers alone versus combined network and morphological biomarkers.

Main Results:

• Network-based biomarkers significantly improved TLE diagnostic accuracy by over 10% compared to morphological biomarkers alone.
• The combination of network and morphological features yielded superior classification performance for TLE.
• Demonstrated the effectiveness of integrating brain network analysis with morphological data for TLE.

Conclusions:

• Integrated network and morphological biomarkers offer superior diagnostic power for TLE.
• This combined approach provides valuable insights into TLE pathology and network dysfunction.
• Findings support personalized clinical management strategies for TLE patients.