Meeting Banner
Abstract #1064

Comparative Study of Four Widely used Classifiers for Prostate Cancer Detection with Multi-parametric MRI

Chengyan Wang 1 , Shuangjuan Cheng 2 , Juan Hu 2 , He Wang 2 , Xuedong Yang 2 , Jue Zhang 1,3 , Xiaoying Wang 1,2 , and Jing Fang 1,3

1 Academy for Advanced Interdisciplinary Studies, Peking University, Beijing, Beijing, China, 2 Department of Radiology, Peking University First Hospital, Beijing, Beijing, China, 3 College of Enigneering, Peking University, Beijing, Beijing, China

Prostate cancer (PCa) is the second most frequently diagnosed cancer and the sixth leading cause of cancer death among men worldwide [1]. Several studies [2-4] have proven that the diagnostic accuracy of PCa detection can be significantly improved by combining different MR sequences, and several computer-aided diagnosis (CAD) systems have been proposed to integrate the MR information. However, no comparison has been made to find out which system performs better. In this study, we evaluate the performance of four widely used classifiers using leave-one-out (LOO) method. MLP and SVM classifiers which have been successfully applied in many branches of medical diagnostics seem promising here. Because of their abilities to resolve nonlinear complex relations among input variables, without the need for any prior assumptions about these relations, MLP and SVM models are more advisable for PCa detection.

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

Join Here