CT-based machine learning model characterizes CK7 expression in clear cell renal cell carcinoma

In an exploratory multicenter study, a logistic regression model using clinical and CT features predicted CK7 expression in ccRCC with validation AUC 0.714 (95% CI 0.589–0.828).

• The final model used age, cystic component, corticomedullary attenuation, corticomedullary enhancement ratio, nephrographic attenuation, and tumor size.
• The model showed moderate discrimination; further prospective validation is needed before clinical use.

Automated summary

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