Meeting Banner
Abstract #4003

Deep learning identification of clear cell renal carcinoma cancer using MR imaging

Junyu Guo1, Keith Husley1, Yin Xi1, and Ivan Pedrosa1
1Radiology, UT southwestern medical center, Dallas, TX, United States


Clear cell renal carcinoma cancer (ccRCC) is the most common subtype among renal masses. ccRCC identification helps in decision making between active surveillance and definitive intervention. A clear cell likelihood score (ccLS) using subjective interpretation of multiparametric MRI by radiologists was proposed recently. In this study, we investigate whether deep learning (DL) using the three main MR sequences for ccLS can facilitate the diagnosis of ccRCC. We compared the results of twelve trained DL models with the reported ccLS performance. Our results demonstrate that DL may achieve a performance comparable to radiologists and provide useful information for identification of ccRCC.

This abstract and the presentation materials are available to members only; a login is required.

Join Here