Synapse
⌘+K
Synapse
PulseExploreClubsResearchersJournals
Instagram
HomeClubsExplore
July 22, 2026Journal of Imaging Informatics in MedicineOpen Access

Physics-Informed Multiscale Decoding of Tissue Microstructure: The Gray Level Affinity Metrics (GLAM) Framework

View Full Paper
Ask AI
Bookmark
Share

Authors

ARAhmad Pour RashidiLPLaetitia PerronneCKChase Krumpelman

Discussion

Loading...

Member takes

Overview

Randomized trial demonstrates superior prognostic performance in high-grade glioma using GLAM metrics, indicating enhanced predictive potential.

Key Points

  • The aim is to develop a physics-informed framework called GLAM to improve tissue microstructure decoding and prognostic outcomes in glioma.
  • Used statistical mechanics to derive GLAM for characterizing tissue microstructures.
  • Conducted assessments on a multi-center cohort of high-grade glioma patients.
  • Employed Leave-One-Center-Out cross-validation and 2000 bootstrap iterations for model validation.
  • GLAM achieved a Mean Test C-Index of 0.646 with a 668-day median survival separation in treatment-responsive MGMT methylated gliomas.
  • Combining GLAM with traditional radiomics yielded a Mean Test C-Index of 0.643 with a 267-day separation in MGMT unmethylated gliomas.
  • GLAM demonstrated higher intrinsic dimensionality and greater informational density compared to conventional texture metrics.

Cite This Study

Rashidi et al. (2026) studied this question.

synapsesocial.com/papers/6a605ee84163e025518d874fhttps://doi.org/10.1007/s10278-026-02132-6
View Full Paper
Ask AI
Bookmark
Share