Additional Look into GAN-based Augmentation for Deep Learning COVID-19 Image Classification
Oleksandr Fedoruk, Konrad Klimaszewski, Aleksander Ogonowski, Michał Kruk
TL;DR
The study addresses data scarcity in medical imaging by evaluating GAN-based augmentation for COVID-19 chest X-ray classification. Using StyleGAN2-ADA, the authors compare GAN-generated augmentation against classical augmentation and no augmentation on two small dataset sizes, observing that GAN augmentation matches classical augmentation on moderate/large datasets but underperforms on the smallest dataset. A clear positive correlation between original dataset size and classification performance emerges, independent of augmentation method. The work highlights substantial computational costs of GAN-based augmentation and suggests potential privacy advantages of synthetic data sharing, recommending further refinement for micro-dataset scenarios.
Abstract
The availability of training data is one of the main limitations in deep learning applications for medical imaging. Data augmentation is a popular approach to overcome this problem. A new approach is a Machine Learning based augmentation, in particular usage of Generative Adversarial Networks (GAN). In this case, GANs generate images similar to the original dataset so that the overall training data amount is bigger, which leads to better performance of trained networks. A GAN model consists of two networks, a generator and a discriminator interconnected in a feedback loop which creates a competitive environment. This work is a continuation of the previous research where we trained StyleGAN2-ADA by Nvidia on the limited COVID-19 chest X-ray image dataset. In this paper, we study the dependence of the GAN-based augmentation performance on dataset size with a focus on small samples. Two datasets are considered, one with 1000 images per class (4000 images in total) and the second with 500 images per class (2000 images in total). We train StyleGAN2-ADA with both sets and then, after validating the quality of generated images, we use trained GANs as one of the augmentations approaches in multi-class classification problems. We compare the quality of the GAN-based augmentation approach to two different approaches (classical augmentation and no augmentation at all) by employing transfer learning-based classification of COVID-19 chest X-ray images. The results are quantified using different classification quality metrics and compared to the results from the literature. The GAN-based augmentation approach is found to be comparable with classical augmentation in the case of medium and large datasets but underperforms in the case of smaller datasets. The correlation between the size of the original dataset and the quality of classification is visible independently from the augmentation approach.