Coevolutionary generative adversarial networks for medical image augumentation at scale
Open Access
- 8 July 2022
- conference paper
- Published by Association for Computing Machinery (ACM)
- p. 367-376
Abstract
Medical image processing can lack images for diagnosis. Generative Adversarial Networks (GANs) provide a method to train generative models for data augmentation. Synthesized images can be used to improve the robustness of computer-aided diagnosis systems. However, GANs are difficult to train due to unstable training dynamics that may arise during the learning process, e.g., mode collapse and vanishing gradients. This paper focuses on Lipizzaner, a GAN training framework that combines spatial coevolution with gradient-based learning, which has been used to mitigate GAN training pathologies. Lipizzaner improves performance by taking advantage of its distributed nature and running at scale. Thus, the Lipizzaner algorithm and implementation robustness can be scaled to high-performance computing (HPC) systems to provide more accurate generative models. We address medical imaging data augmentation to create chest X-Ray images by using Lipizzaner on the HPC infrastructure provided by Oak Ridge National Labs' Summit Supercomputer. The experimental analysis shows improved performance by increasing the scale of the Lipizzaner GAN training. We also demonstrate that distributed coevolutionary learning improves performance even when using suboptimal neural network architectures due to hardware constraints.Keywords
Funding Information
- MCIN/AEI/10.13039/501100011033 (PID 2020-116727RB-I00)
- EU Horizon 2020 research and innovation programme
- Univ. de Malaga (UMA18-FEDERJA-003)
- TAILOR ICT-48 Network (952215)
This publication has 28 references indexed in Scilit:
- Coevolution of Generative Adversarial Networks2019
- Recent Progress on Generative Adversarial Networks (GANs): A SurveyIEEE Access, 2019
- Chest Radiographs in Congestive Heart Failure: Visualizing Neural Network LearningRadiology, 2019
- Evolutionary Generative Adversarial NetworksIEEE Transactions on Evolutionary Computation, 2019
- Designing neural networks through neuroevolutionNature Machine Intelligence, 2019
- Evolving Deep Neural Networks2019
- Evolved GANs for generating pareto set approximations2018
- SGAN: An Alternative Training of Generative Adversarial Networks2018
- Parallel metaheuristics: recent advances and new trendsInternational Transactions in Operational Research, 2012
- Coevolutionary Principles2012