WGAN-GP for Synthetic Retinal Image Generation: Enhancing Sensor-Based Medical Imaging for Classification Models
Open Access
Abstract
Accurate synthetic image generation is crucial for addressing data scarcity challenges in medical image classification tasks, particularly in sensor-derived medical imaging. In this work, we propose a novel method using a Wasserstein Generative Adversarial Network with Gradient Penalty (WGAN-GP) and nearest-neighbor interpolation to generate high-quality synthetic images for diabetic retinopathy classification. Our approach enhances training datasets by generating realistic retinal images that retain critical pathological features. We evaluated the method across multiple retinal image datasets, including Retinal-Lesions, Fine-Grained Annotated Diabetic Retinopathy (FGADR), Indian Diabetic Retinopathy Image Dataset (IDRiD), and the Kaggle Diabetic Retinopathy dataset. The proposed method outperformed traditional generative models, such as conditional GANs and PathoGAN, achieving the best performance on key metrics: a Fréchet Inception Distance (FID) of 15.21, a Mean Squared Error (MSE) of 0.002025, and a Structural Similarity Index (SSIM) of 0.89 in the Kaggle dataset. Additionally, expert evaluations revealed that only 56.66% of synthetic images could be distinguished from real ones, demonstrating the high fidelity and clinical relevance of the generated data. These results highlight the effectiveness of our approach in improving medical image classification by generating realistic and diverse synthetic datasets.Keywords
Funding Information
- National Council of Humanities, Sciences, Technologies, and Innovation of Mexico (CONAHCyT)
This publication has 25 references indexed in Scilit:
- Explainable Diabetic Retinopathy Detection and Retinal Image GenerationIEEE Journal of Biomedical and Health Informatics, 2021
- A Benchmark for Studying Diabetic Retinopathy: Segmentation, Grading, and TransferabilityIEEE Transactions on Medical Imaging, 2021
- DR-GAN: Conditional Generative Adversarial Network for Fine-Grained Lesion Synthesis on Diabetic Retinopathy ImagesIEEE Journal of Biomedical and Health Informatics, 2020
- Not-so-supervised: A survey of semi-supervised, multi-instance, and transfer learning in medical image analysisMedical Image Analysis, 2019
- Supervised Segmentation of Un-Annotated Retinal Fundus Images by SynthesisIEEE Transactions on Medical Imaging, 2018
- A survey on deep learning in medical image analysisMedical Image Analysis, 2017
- Automated Identification of Diabetic Retinopathy Using Deep LearningOphthalmology, 2017
- Development and Validation of a Deep Learning Algorithm for Detection of Diabetic Retinopathy in Retinal Fundus PhotographsJAMA, 2016
- ImageNet Large Scale Visual Recognition ChallengeInternational Journal of Computer Vision, 2015
- Image quality assessment: from error visibility to structural similarityIEEE Transactions on Image Processing, 2004