GANs 'N Lungs: improving pneumonia prediction
This addresses the challenge of imbalanced data in medical imaging for pneumonia prediction, but it is incremental as it applies an existing GAN method to a specific domain.
The authors tackled the problem of improving deep learning model performance on highly-imbalanced tasks by using CycleGAN for data augmentation to balance datasets, resulting in improved accuracy for pneumonia binary classification.
We propose a novel method to improve deep learning model performance on highly-imbalanced tasks. The proposed method is based on CycleGAN to achieve balanced dataset. We show that data augmentation with GAN helps to improve accuracy of pneumonia binary classification task even if the generative network was trained on the same training dataset.