Abstract #2449

Automated diagnosis of prostate cancer from dynamic contrast-enhanced MRI using a Convolution Neural Network–based deep learning approach

Ming Deng¹, Haibo Xu², Xiaoyong Zhang³, and Yingao Zhang⁴

¹Department of Radiology, Zhongnan Hospital of Wuhan University, Wuhan University,, Wuhan, China, ²Department of Radiology, Zhongnan Hospital of Wuhan University, Wuhan University, Wuhan, China, ³MR Collaborations, Siemens Healthcare Ltd, Shenzhen, China, ⁴Zhongnan Hospital of Wuhan University, Wuhan University, Wuhan, China

The aim of this study was to evaluate the diagnostic performance of a convolutional neural network (CNN)-based deep learning technique for the differentiation of prostate cancer (PC) using dynamic contrast agent–enhanced magnetic resonance imaging (DCE-MRI) data. Our patient study demonstrated that the quantitative image features derived from the DCE-MR images based on the self-defined CNN model can be effective in distinguishing PC from the normal, and the automated extraction of Ktrans, TDC, DR, and DY features can significantly promote PC diagnosis. The high performance of the proposed CNN-based deep learning method statistical analysis demonstrated its potential for improving PC diagnosis.

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

Join Here