New AI framework enhances stroke assessment with uncertainty quantification

Researchers have developed Evidential Perfusion Physics-Informed Neural Networks (EPPINN) to improve the accuracy and reliability of computed tomography perfusion (CTP) imaging for acute ischemic stroke assessment. This new framework integrates evidential deep learning with physics-informed modeling to quantify uncertainty in physics constraints, addressing a key limitation of existing deterministic approaches. EPPINN models perfusion parameters using coordinate-based networks and places a Normal--Inverse--Gamma distribution over the physics residual to characterize voxel-wise uncertainty without requiring Bayesian sampling. Evaluations on phantom data, a benchmark dataset, and clinical data show EPPINN achieves lower error rates and provides more conservative uncertainty estimates than traditional methods, particularly under challenging conditions like sparse temporal sampling. AI

IMPACT This research could lead to more reliable and accurate diagnoses in critical medical scenarios like stroke assessment by providing quantifiable uncertainty in AI-driven analyses.

RANK_REASON The cluster contains a research paper detailing a new AI methodology for medical imaging analysis. [lever_c_demoted from research: ic=1 ai=1.0]

Read on arXiv cs.CV →

• Evidential Perfusion Physics-Informed Neural Networks
• ISLES 2018
• Junhyeok Lee
• Normal--Inverse--Gamma distribution
• Physics-informed neural networks

• paper
• model release
• other

AI-generated summary · Google Gemini · from 1 sources. How we write summaries →