Abstract #0704

A Comparison between Radiologists versus Deep Learning for Prostate Cancer Detection in Multi-parameter MRI

Ruiming Cao¹, Xinran Zhong², Sohrab Afshari Mirak³, Ely Felker³, Voraparee Suvannarerg^3,4, Teeravut Tubtawee^3,5, Fabien Scalzo⁶, Steve Raman³, and Kyunghyun Sung³

¹Bioengineering, UC Berkeley, Berkeley, CA, United States, ²UT Southwestern, Dallas, TX, United States, ³Radiology, UCLA, Los Angeles, CA, United States, ⁴Radiology, Faculty of Medicine Siriraj Hospital, Mahidol University, Bangkok, Thailand, ⁵Radiology, Faculty of Medicine, Prince of Songkla University, Songkhla, Thailand, ⁶Computer Science, UCLA, Los Angeles, CA, United States

We evaluated our recently developed deep learning system, FocalNet, for prostate cancer detection in multi-parametric MRI (mpMRI). This study performed a head-to-head comparison between FocalNet and four genitourinary radiologists in an independent evaluation cohort consisting of 126 mpMRI scans untouched during the development. FocalNet demonstrated similar detection performance to radiologists under the high specificity condition or the high sensitivity condition, while radiologists outperformed FocalNet in moderate specificity and sensitivity.

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

Join Here