AI models for cancer pathology often rely on shortcut correlations rather than detecting true biological signals, compromising reliability.
Key Details
- 1University of Warwick research analyzed 8,000+ patient samples across four cancer types using deep learning pathology models.
- 2Findings show many models predict biomarkers based on confounded visual cues, not causal biological features.
- 3AI models' headline accuracy can drop markedly when evaluated in stratified subgroups controlling for confounding factors.
- 4Reported accuracy for AI (just over 80%) was only modestly better than using tumour grade alone (~75%), already assessed by pathologists.
- 5Authors stress need for bias-aware evaluation and subgroup testing to prevent premature and unsafe clinical deployment.
Why It Matters
This research underscores significant risks of overrelying on AI diagnostic tools that may not generalize well in real-world settings. Robustness, causal modeling, and rigorous validation standards are essential before AI can safely supplement or replace human expertise in pathology and radiology.
Source
EurekAlert
Related News
•EurekAlert
NIH-Backed AI Model Predicts Cancer Survival Using Single-Cell Data
Researchers have developed scSurvival, a machine learning tool that uses single-cell tumor data to accurately predict cancer patient survival and identify high-risk cell populations.
•EurekAlert
AI Pathology Model Outperforms PD-L1 in Predicting NSCLC Immunotherapy Response
MD Anderson's Path-IO machine learning platform accurately predicts immunotherapy responses in metastatic non-small cell lung cancer, surpassing current biomarker standards.
•EurekAlert
Deep Learning Pathomics Platform Improves Immunotherapy Prediction in Lung Cancer
A deep learning pathomics platform accurately predicts immunotherapy response in metastatic NSCLC using routine pathology slides.