A deep learning model named Sybil can predict future lung cancer risk from a single low-dose chest CT scan, as validated in a large Asian cohort.
Key Details
- 1The Sybil model was validated using data from over 21,000 individuals aged 50-80 who underwent LDCT screening between 2009 and 2021.
- 2Sybil demonstrated strong predictive performance for lung cancer diagnosis both at one and six years following the scan.
- 3The model was effective even for never-smokers, a group for whom conventional screening guidelines may not apply.
- 4The research was presented at the ATS 2025 International Conference.
- 5Continued prospective validation is planned to assess clinical use and prediction of lung cancer-specific mortality.
Why It Matters
This advancement could help tailor lung cancer screening strategies, especially benefiting populations (such as Asian never-smokers) not well-served by current guidelines. A validated AI risk tool from a single scan could make screening more efficient and personalized in routine radiological practice.
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.