Generative Adversarial Network for Medical Images (MI-GAN)

TL;DR

This paper introduces MI-GAN, a generative adversarial network designed to produce synthetic medical images and masks, improving training data availability and segmentation accuracy in medical imaging tasks.

Contribution

The paper presents a novel GAN architecture tailored for medical image synthesis, achieving state-of-the-art segmentation performance on retinal datasets.

Findings

Achieved dice coefficient of 0.837 on STARE dataset.

Achieved dice coefficient of 0.832 on DRIVE dataset.

Generated high-quality synthetic retinal images and masks.

Abstract

Deep learning algorithms produces state-of-the-art results for different machine learning and computer vision tasks. To perform well on a given task, these algorithms require large dataset for training. However, deep learning algorithms lack generalization and suffer from over-fitting whenever trained on small dataset, especially when one is dealing with medical images. For supervised image analysis in medical imaging, having image data along with their corresponding annotated ground-truths is costly as well as time consuming since annotations of the data is done by medical experts manually. In this paper, we propose a new Generative Adversarial Network for Medical Imaging (MI-GAN). The MI-GAN generates synthetic medical images and their segmented masks, which can then be used for the application of supervised analysis of medical images. Particularly, we present MI-GAN for synthesis of retinal images. The proposed method generates precise segmented images better than the existing techniques. The proposed model achieves a dice coefficient of 0.837 on STARE dataset and 0.832 on DRIVE dataset which is state-of-the-art performance on both the datasets.