AI model detects esophageal cancer and precancerous lesions on noncontrast chest CT

An AI model (EAGLE) analyzing noncontrast chest CT achieved 90.0% sensitivity for esophageal cancer and 98.5% specificity in external multicenter testing (11,466 patients), and 99.94% specificity in low-dose screening. Could enable opportunistic esophageal cancer screening.

• In external opportunistic screening cohorts (n=11,466), specificity was 98.5%, sensitivity 90.0% for cancer and 52.5% for precancerous lesions; low-dose CT validation (n=1,607) showed comparable performance.
• In a real-world cohort (n=35,402), calibration reduced false positives by 72.7% while preserving sensitivity; prospective hospital validation (n=17,446) achieved 42.2% positive predictive value.
• In paired CT-endoscopy cohorts (n=702), sensitivities were 65.0% for precancerous lesions and 78.4% for stage I esophageal cancer at a higher-sensitivity operating point.

Automated summary

RadPigeon summaries are original and for information only. They are not clinical advice.