Prostate Imaging-Reporting and Data System (PI-RADS) suggests acquiring multiple apparent diffusion coefficient (ADC) maps including the lowest b-values between 50-100s/mm2 and highest b-values greater than 1400s/mm2. Radiomics is a novel field in medical imaging to advance decision support by utilizing large amount of quantitative features. In this study, we employed radiomics from ADC maps and a linear regression model to differentiate prostate cancer from benign tissues and evaluated the effect of various b-value combinations on ADC maps. We discovered that ADC with the b-values of 100 and 1000s/mm2 was most effective in discriminating prostate cancer with high accuracy.
This abstract and the presentation materials are available to members only; a login is required.