Modeling the Intra-class Variability for Liver Lesion Detection using a Multi-class Patch-based CNN

TL;DR

This paper introduces a multi-class CNN that explicitly models intra-class variability in liver CT images to improve automatic lesion detection, outperforming previous methods and aiding radiologists.

Contribution

It presents a novel multi-class CNN approach that categorizes image patches into sub-classes to enhance liver lesion detection accuracy.

Findings

Outperforms state-of-the-art fully convolutional networks

Uses 132 liver CT images with 498 lesions for validation

Shows significant improvement in detection results

Abstract

Automatic detection of liver lesions in CT images poses a great challenge for researchers. In this work we present a deep learning approach that models explicitly the variability within the non-lesion class, based on prior knowledge of the data, to support an automated lesion detection system. A multi-class convolutional neural network (CNN) is proposed to categorize input image patches into sub-categories of boundary and interior patches, the decisions of which are fused to reach a binary lesion vs non-lesion decision. For validation of our system, we use CT images of 132 livers and 498 lesions. Our approach shows highly improved detection results that outperform the state-of-the-art fully convolutional network. Automated computerized tools, as shown in this work, have the potential in the future to support the radiologists towards improved detection.