Deep learning models outperform Brock score for incidental lung nodule malignancy risk

Two deep learning models predicted incidental lung nodule malignancy with higher specificity (60%) than the widely used Brock model (44%), at matched sensitivity, in a multicenter external validation study of 269 nodules.

• The screening-trained model and the combined screening+clinical model achieved AUCs of 0.74 and 0.72, respectively, versus 0.63 for Brock.
• Performance differences held across three centers with varying CT equipment and patient populations.

Automated summary

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