Energy-based Tissue Manifolds for Longitudinal Multiparametric MRI Analysis
Researchers have developed a novel geometric framework for analyzing longitudinal multiparametric MRI data. This approach uses patient-specific energy modeling in sequence space, representing each voxel by its intensity vector across multiple MRI sequences. An implicit neural representation is trained to learn an energy function, which then describes tissue regimes without requiring segmentation labels. AI
IMPACT Introduces a novel geometric framework for MRI analysis using energy modeling, potentially improving longitudinal tracking of tissue changes in neuro-oncology.