Physics-Based Reaction-Diffusion Model Improves Stroke Lesion Prediction

A team from the University of Zurich and GIK Institute embedded the Fisher-KPP reaction-diffusion equation into patient-specific 3D MRI grids to predict stroke lesion growth. The physics-driven approach nearly doubled Dice scores over thresholding methods, presenting an alternative to pure deep learning.

2026-09-11 ~ 2026-09-11 · 2 related posts

• Physics-based reaction-diffusion model predicts stroke infarct growth from routine MRI — maier_ak · 2026-09-11
• Physics-based Fisher-KPP stroke model beats rCBF thresholding: Dice 0.46 vs 0.25 — maier_ak · 2026-09-11