CT Image Enhancement Using Stacked Generative Adversarial Networks and Transfer Learning for Lesion Segmentation Improvement

TL;DR

This paper introduces a stacked GAN-based preprocessing method combined with transfer learning to enhance CT images, significantly improving lesion segmentation accuracy in medical imaging.

Contribution

It presents a novel SGAN approach for CT image enhancement and demonstrates its effectiveness in improving lesion segmentation performance.

Findings

SGAN preprocessing improves segmentation accuracy

HNN + SGAN outperforms other methods

Enhancement with SGAN surpasses original image segmentation results

Abstract

Automated lesion segmentation from computed tomography (CT) is an important and challenging task in medical image analysis. While many advancements have been made, there is room for continued improvements. One hurdle is that CT images can exhibit high noise and low contrast, particularly in lower dosages. To address this, we focus on a preprocessing method for CT images that uses stacked generative adversarial networks (SGAN) approach. The first GAN reduces the noise in the CT image and the second GAN generates a higher resolution image with enhanced boundaries and high contrast. To make up for the absence of high quality CT images, we detail how to synthesize a large number of low- and high-quality natural images and use transfer learning with progressively larger amounts of CT images. We apply both the classic GrabCut method and the modern holistically nested network (HNN) to lesion segmentation, testing whether SGAN can yield improved lesion segmentation. Experimental results on the DeepLesion dataset demonstrate that the SGAN enhancements alone can push GrabCut performance over HNN trained on original images. We also demonstrate that HNN + SGAN performs best compared against four other enhancement methods, including when using only a single GAN.