Meeting Banner
Abstract #0832

Added Value of DCE in Machine Learning-based Tumor Probability Maps for Predicting Clinically Significant Cancer Foci in Pre-biopsy MR images

Gabriel Addio Nketiah1, Léo Pallas2, Adrian L Breto 3, Radka Stoyanova3, Mattijs Elschot 1, and Tone F. Bathen1,4
1Circulation and Medical Imaging, Norwegian University of Science and Technology, Trondheim, Norway, 2Department of Digital Science, CPE Lyon, Lyon, Norway, 3Sylvester Comprehensive Cancer Center, University of Miami, Miami, FL, United States, 4Clinic of Radiology and Nuclear Medicine, St. Olavs University Hospital, Trondheim, Norway

The utility of multiparametric (mp) versus biparametric (bp) MRI protocol in prostate cancer diagnosis has been compared in several large studies, but mainly using manual qualitative evaluation. In this study we employed machine learning models to investigate the added value of DCE (i.e. mpMRI) in predicting significant cancer foci in pre-biopsy MR images. Whereas both protocols had comparable results in the whole prostate and transition zone analyses, we found mpMRI model to be more useful in the peripheral zone, where significant differences (p < 0.05) were found for all performance measures i.e. area under the curve, accuracy, sensitivity and specificity.

This abstract and the presentation materials are available to 2020 meeting attendees and eLibrary customers only; a login is required.

Join Here