Prostate Cancer Diagnosis using Deep Learning with 3D Multiparametric MRI

TL;DR

This paper introduces XmasNet, a deep learning model that effectively classifies prostate cancer lesions using 3D multiparametric MRI data, outperforming traditional methods and achieving high accuracy in a competitive challenge.

Contribution

The study presents a novel 3D CNN architecture, XmasNet, specifically designed for prostate cancer MRI classification, demonstrating superior performance over existing models.

Findings

XmasNet achieved an AUC of 0.84 in the PROSTATEx challenge.

XmasNet outperformed 69 methods from 33 groups.

Deep learning shows strong potential for cancer imaging.

Abstract

A novel deep learning architecture (XmasNet) based on convolutional neural networks was developed for the classification of prostate cancer lesions, using the 3D multiparametric MRI data provided by the PROSTATEx challenge. End-to-end training was performed for XmasNet, with data augmentation done through 3D rotation and slicing, in order to incorporate the 3D information of the lesion. XmasNet outperformed traditional machine learning models based on engineered features, for both train and test data. For the test data, XmasNet outperformed 69 methods from 33 participating groups and achieved the second highest AUC (0.84) in the PROSTATEx challenge. This study shows the great potential of deep learning for cancer imaging.