Pathology-Aware Generative Adversarial Networks for Medical Image Augmentation

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.

arXiv commentPh.D. Thesis (The University of Tokyo) defended in February, 2020. Based on GAN-based synthetic brain MR image generation (ISBI 2018), arXiv:1902.09856, arXiv:1903.12564, arXiv:1905.13456, arXiv:1906.04962, arXiv:2001.03923

References in corpus (28)

• Very Deep Convolutional Networks for Large-Scale Image Recognition
• Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks
• A Survey on Deep Learning in Medical Image Analysis
• Auto-Encoding Variational Bayes
• Unsupervised Representation Learning with Deep Convolutional Generative Adversarial Networks
• Model-Agnostic Meta-Learning for Fast Adaptation of Deep Networks
• YOLOv3: An Incremental Improvement
• mixup: Beyond Empirical Risk Minimization
• Brain Tumor Segmentation with Deep Neural Networks
• Improved Regularization of Convolutional Neural Networks with Cutout
• The Effectiveness of Data Augmentation in Image Classification using Deep Learning
• GAN-based Synthetic Medical Image Augmentation for increased CNN Performance in Liver Lesion Classification
• Generative Adversarial Network in Medical Imaging: A Review
• Low Dose CT Image Denoising Using a Generative Adversarial Network with Wasserstein Distance and Perceptual Loss
• Validation, comparison, and combination of algorithms for automatic detection of pulmonary nodules in computed tomography images: the LUNA16 challenge
• Benchmark Analysis of Representative Deep Neural Network Architectures
• Data Augmentation Generative Adversarial Networks
• AutoAugment: Learning Augmentation Policies from Data
• DenseNet: Implementing Efficient ConvNet Descriptor Pyramids
• Med3D: Transfer Learning for 3D Medical Image Analysis
• Fully Convolutional Network Ensembles for White Matter Hyperintensities Segmentation in MR Images
• Learning What and Where to Draw
• Pedestrian-Synthesis-GAN: Generating Pedestrian Data in Real Scene and Beyond
• MC-GAN: Multi-conditional Generative Adversarial Network for Image Synthesis
• Generative Image Translation for Data Augmentation of Bone Lesion Pathology
• Enforcing constraints for interpolation and extrapolation in Generative Adversarial Networks
• Towards generative adversarial networks as a new paradigm for radiology education
• Machine learning and AI research for Patient Benefit: 20 Critical Questions on Transparency, Replicability, Ethics and Effectiveness