Colorectal cancer detection using non-contrast CT and deep learning: a multicenter and international cohort study
AI analysis of routine CT scans spots colorectal cancers better than radiologists alone without extra imaging, turning existing scans into screening opportunities.
COCA, a deep learning model applied to non-contrast CT scans, demonstrated robust colorectal cancer detection with AUC up to 0.996 across international validation centers and two real-world consecutive-patient cohorts totaling over 27,000 cases. The system improved radiologist sensitivity by 20.4% and maintained specificity above 99%, positioning it as a potential opportunistic CRC screening tool in routine CT workflows without additional imaging burden.
What the study was
- Study design
- Retrospective multicenter international cohort + real-world validation
- Population
- CRC patients and normal controls; development (1321 CRC + 1357 controls), 6-center validation (2053 patients), 2 real-world cohorts (9016 + 18427 consecutive patients)
- Sample size
- 29796
- Category
- Early Detection
- Maturity
- Validated
- Journal
- Annals of Oncology
Why it surfaced
Large-scale multicenter international validation (n>29,000 total) of a non-invasive AI CRC screening tool on non-contrast CT. Real-world consecutive patient performance confirms clinical-grade accuracy. Ann Oncol publication adds credibility. Highly near-term implementable in routine CT workflows.
A plain-language summary of published research — not medical advice. Talk to a clinician about your own care.