Abstract #4943

Fine-tuned Deep Convolutional Neural Network for Automatic Detection of Clinically Significant Prostate Cancer with Multi-parametric MRI

Xinran Zhong^1,2, Hung Le Minh³, Holden Wu^1,2, Michael Kuo¹, Steven Raman¹, William Hsu¹, Xin Yang³, and Kyunghyun Sung^1,2

¹Department of Radiological Sciences, University of California, Los Angeles, Los Angeles, CA, United States, ²Physics and Biology in Medicine IDP, University of California, Los Angeles, Los Angeles, CA, United States, ³School of Electronics Information and Communications, Huazhong University of Science and Technology, Wuhan, People's Republic of China

A deep convolutional neural network (CNN) based automatic classification system to distinguish between indolent and clinically significant prostate carcinoma using multi-parametric MRI (mp-MRI) is proposed. By applying data augmentation, 138 lesions were used to fine-tune the pre-trained CNN model called Overfeat. Those fine-tuned models were then shown to provide better performance than existing pre-trained CNN method, texture features based system as well as PI-RADS standards on a separate 40 testing cases.

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

Join Here