Estimating glioblastoma migration-to-proliferation ratio from single time-point MRI data via radial density distribution

A 1D reaction–diffusion fit to single time-point pre-treatment MRI segmentations estimates glioblastoma migration-to-proliferation ratio D/k with under 5% error and PCC = 0.996 on synthetic data.

• Why it matters: D/k is linked to treatment resistance, but estimating it from routine MRI is challenging, limiting studies of how tumor heterogeneity varies across patients.
• What they did: The method converts enhancing and non-enhancing segmentations into a 1D tumor density distribution and fits a 1D reaction–diffusion model to infer D/k from one MRI time point.
• The result: In 102 BraTS 2017 pre-treatment scans, elevated D/k is associated with reduced overall survival (log-rank p = 0.01), older age (p = 0.04), and lower Karnofsky Performance Status (p = 0.059), and mesenchymal-like phenotypes show higher D/k in 91 patients (p = 0.013, 0.055, 0.127).

AI summary of the abstract · may contain errors · not medical advice · read the paper.