Synthesizing Filamentary Structured Images with GANs

TL;DR

This paper introduces a GAN-based method for synthesizing realistic filamentary structured images from minimal training data, which enhances medical image analysis tasks.

Contribution

The approach effectively synthesizes diverse, realistic filamentary images from limited data and improves downstream image analysis performance.

Findings

Works well with as few as 10 training examples

Generates diverse and realistic images

Boosts image analysis accuracy

Abstract

This paper aims at synthesizing filamentary structured images such as retinal fundus images and neuronal images, as follows: Given a ground-truth, to generate multiple realistic looking phantoms. A ground-truth could be a binary segmentation map containing the filamentary structured morphology, while the synthesized output image is of the same size as the ground-truth and has similar visual appearance to what have been presented in the training set. Our approach is inspired by the recent progresses in generative adversarial nets (GANs) as well as image style transfer. In particular, it is dedicated to our problem context with the following properties: Rather than large-scale dataset, it works well in the presence of as few as 10 training examples, which is common in medical image analysis; It is capable of synthesizing diverse images from the same ground-truth; Last and importantly, the synthetic images produced by our approach are demonstrated to be useful in boosting image analysis performance. Empirical examination over various benchmarks of fundus and neuronal images demonstrate the advantages of the proposed approach.