Cascaded deep learning system on MRI classifies benign vs malignant small renal masses

A cascaded deep learning pipeline that automatically segments and classifies small renal masses on MRI achieved an external-test AUC of 0.788, comparable to a senior radiologist. On the internal test set (AUC 0.936), it significantly outperformed the senior radiologist and junio…

• The automated segmentation model achieved comparable classification performance to manual segmentation (internal AUC: 0.936 vs 0.944, p=0.671; external AUC: 0.788 vs 0.832, p=0.629).
• In subgroup analysis of tumors <3 cm, the model still significantly outperformed the junior radiologist.
• The study trained on 712 patients from one institution and was externally validated on 77 patients from two other centers.

Automated summary

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