Pathology-Aware Generative Adversarial Networks for Medical Image Augmentation

TL;DR

This paper introduces pathology-aware GANs for medical image augmentation, enhancing data diversity for diagnosis and training, especially in oncology, by generating realistic pathological images to improve AI models and physician education.

Contribution

It presents novel pathology-aware GAN models tailored for medical image augmentation, addressing clinical relevance and supporting both AI training and physician education.

Findings

GAN-augmented data improves diagnostic accuracy

Generated images reflect realistic pathological variability

Supports physician training with diverse pathological images

Abstract

Convolutional Neural Networks (CNNs) can play a key role in Medical Image Analysis under large-scale annotated datasets. However, preparing such massive dataset is demanding. In this context, Generative Adversarial Networks (GANs) can generate realistic but novel samples, and thus effectively cover the real image distribution. In terms of interpolation, the GAN-based medical image augmentation is reliable because medical modalities can display the human body's strong anatomical consistency at fixed position while clearly reflecting inter-subject variability; thus, we propose to use noise-to-image GANs (e.g., random noise samples to diverse pathological images) for (i) medical Data Augmentation (DA) and (ii) physician training. Regarding the DA, the GAN-generated images can improve Computer-Aided Diagnosis based on supervised learning. For the physician training, the GANs can display novel desired pathological images and help train medical trainees despite infrastructural/legal constraints. This thesis contains four GAN projects aiming to present such novel applications' clinical relevance in collaboration with physicians. Whereas the methods are more generally applicable, this thesis only explores a few oncological applications.