Clinical factor nomogram predicts AI-defined cardiopulmonary abnormality risk on LDCT screening

A nomogram based on smoking, metabolic status, BMI, age, and sex predicts cardiopulmonary abnormalities on LDCT with AUC 0.833 (95% CI 0.787-0.879) in validation, outperforming age/sex alone (AUC 0.647).

• Validation AUC 0.833 (95% CI 0.787-0.879) vs. age/gender AUC 0.647 (ΔAUC 0.186, p<0.001).
• Excellent calibration (Hosmer-Lemeshow p=0.591) and positive clinical net benefit.
• Key limitation: single-center retrospective study lacking external validation; prospective studies warranted.

Automated summary

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