An international study finds that AI surpasses average radiologists in detecting pancreatic cancer on CT scans using a newly developed benchmark and dataset.
Key Details
- 1The study was led by Radboudumc and published in The Lancet Oncology.
- 2A confidential dataset of nearly 400 patient CT scans was assessed by international experts and used to benchmark over 250 submitted AI models.
- 3The best AI model made correct assessments in 92% of scans, compared to 88% for radiologists, and produced 38% fewer false positives.
- 4The AI outperformed the average radiologist but requires further validation before clinical adoption.
- 5The research was funded by the Hanarth Fund and the EU’s Horizon Europe program.
Why It Matters
This study provides strong evidence for the potential of AI to support radiologists in diagnosing pancreatic cancer, a particularly deadly and hard-to-detect disease. The validated benchmark and international collaboration pave the way for further clinical development and adoption of AI in abdominal CT interpretation.
Source
EurekAlert
Related News
•EurekAlert
AI-Powered OCT Enables Rapid 'Optical Biopsy' for Early Endometrial Cancer Detection
A team at Washington University has developed a catheter-based 3D OCT system with AI to quickly and noninvasively detect early endometrial cancers.
•EurekAlert
AI Clinical Reasoning in Diagnostics and Digital Fatigue in Healthcare
Recent JMIR features explore large language models in clinical diagnostics and digital fatigue among healthcare professionals.
•EurekAlert
KAIST, MIT, Microsoft Develop Efficient AI Image Upsampling for Robotics
KAIST, MIT, and Microsoft have created 'Upsample Anything,' a training-free AI method to restore high-resolution visual data from compressed images with up to 16x improved GPU memory efficiency.