AI Detects Pancreatic Cancer Earlier Than Doctors
A new AI system called REDMOD detected pancreatic cancer earlier than current medical methods in a study, identifying early signs an average of 475 days before clinical diagnosis. The model analyzed CT scans from 219 patients and showed higher sensitivity than radiologists, correctly identifying 73% of true cases.
Researchers developed the Radiomics-based Early Detection Model (REDMOD) to detect pancreatic cancer earlier. REDMOD analyzes patterns in tissue texture to detect early cancer signals not visible on standard CT scans. The model was tested using abdominal CT scans from 219 patients who were initially considered disease-free but later diagnosed with pancreatic cancer. REDMOD identified early signs of cancer an average of 475 days before clinical diagnosis, showing higher sensitivity than radiologists. The model correctly identified 73% of true cases compared to 39% by radiologists. Further testing in high-risk patients is required before it can be used widely in clinical practice.
This content was automatically generated and/or translated by AI. It may contain inaccuracies. Please refer to the original sources for verification.