Image Synthesis for Data Augmentation in Medical CT using Deep Reinforcement Learning

TL;DR

This paper introduces a novel deep reinforcement learning and style-transfer based method to generate diverse, high-resolution, anatomically accurate CT images, addressing data scarcity issues in medical imaging for improved deep learning applications.

Contribution

The paper presents a new integrated DRL and style-transfer approach for synthesizing diverse, high-quality CT images from small datasets, enhancing data augmentation for medical imaging.

Findings

Effective generation of anatomically accurate CT images

Works well with small datasets

Potential to improve deep learning training in medical imaging

Abstract

Deep learning has shown great promise for CT image reconstruction, in particular to enable low dose imaging and integrated diagnostics. These merits, however, stand at great odds with the low availability of diverse image data which are needed to train these neural networks. We propose to overcome this bottleneck via a deep reinforcement learning (DRL) approach that is integrated with a style-transfer (ST) methodology, where the DRL generates the anatomical shapes and the ST synthesizes the texture detail. We show that our method bears high promise for generating novel and anatomically accurate high resolution CT images at large and diverse quantities. Our approach is specifically designed to work with even small image datasets which is desirable given the often low amount of image data many researchers have available to them.