Meeting Banner
Abstract #4284

Diffusion-kurtosis based texture features for differential diagnosis between prostate cancer and benign prostatic hyperplasia

Yunsong Liu1, Ailian Liu1, Lihua Chen1, Jiazheng Wang2, Liangjie Lin2, and Xiaoxiao Zhang2
1The First Affiliated Hospital of Dalian Medical University, Dalian, China, 2Philips Healthcare, Beijing, China

Synopsis

The differential diagnosis between prostate cancer (PCa) and benign prostatic hyperplasia (BPH) remains challenging due to their overlapping atypical clinical symptoms and MRI features. This study aims to explore the performance of mean kurtosis (MK) based texture features in differentiation between PCa and BPH. Results show that the MK texture features can help distinguish PCa and BPH. The 90Percentile MK value show the best diagnostic performance (AUC: 0.93; sensitivity: 88%; specificity: 100%).

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

Join Here