CT Image Enhancement Using Stacked Generative Adversarial Networks and Transfer Learning for Lesion Segmentation Improvement
This work addresses the challenge of high noise and low contrast in CT images for medical image analysis, offering an incremental improvement in lesion segmentation.
The paper tackled the problem of automated lesion segmentation from CT images by proposing a preprocessing method using stacked generative adversarial networks (SGAN) to reduce noise and enhance contrast, resulting in improved segmentation performance where SGAN enhancements alone allowed GrabCut to outperform HNN on original images, with HNN + SGAN achieving the best results compared to other methods.
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.