GAN Augmentation: Augmenting Training Data using Generative Adversarial Networks

Abstract

One of the biggest issues facing the use of machine learning in medical imaging is the lack of availability of large, labelled datasets. The annotation of medical images is not only expensive and time consuming but also highly dependent on the availability of expert observers. The limited amount of training data can inhibit the performance of supervised machine learning algorithms which often need very large quantities of data on which to train to avoid overfitting. So far, much effort has been directed at extracting as much information as possible from what data is available. Generative Adversarial Networks (GANs) offer a novel way to unlock additional information from a dataset by generating synthetic samples with the appearance of real images. This paper demonstrates the feasibility of introducing GAN derived synthetic data to the training datasets in two brain segmentation tasks, leading to improvements in Dice Similarity Coefficient (DSC) of between 1 and 5 percentage points under different conditions, with the strongest effects seen fewer than ten training image stacks are available.

References in corpus (6)

• Unsupervised Representation Learning with Deep Convolutional Generative Adversarial Networks
• Efficient Multi-Scale 3D CNN with Fully Connected CRF for Accurate Brain Lesion Segmentation
• Progressive Growing of GANs for Improved Quality, Stability, and Variation
• Data Augmentation Generative Adversarial Networks
• Playing for Data: Ground Truth from Computer Games
• Data Augmentation in Emotion Classification Using Generative Adversarial Networks

Cited by in corpus (72)

• Generative Adversarial Network in Medical Imaging: A Review
• AutoML: A Survey of the State-of-the-Art
• Global and Local Contrastive Self-Supervised Learning for Semantic Segmentation of HR Remote Sensing Images
• Selective Synthetic Augmentation with HistoGAN for Improved Histopathology Image Classification
• At the Dawn of Generative AI Era: A Tutorial-cum-Survey on New Frontiers in 6G Wireless Intelligence
• A survey of synthetic data augmentation methods in computer vision
• Segmentation of Lungs in Chest X-Ray Image Using Generative Adversarial Networks
• Controlling generative models with continuous factors of variations
• On Leveraging Pretrained GANs for Generation with Limited Data
• Inspect, Understand, Overcome: A Survey of Practical Methods for AI Safety
• Model Patching: Closing the Subgroup Performance Gap with Data Augmentation
• FedGAN: Federated Generative Adversarial Networks for Distributed Data
• Photoacoustic image synthesis with generative adversarial networks
• Regularization Methods for Generative Adversarial Networks: An Overview of Recent Studies
• Regularizing Deep Networks with Semantic Data Augmentation
• Sill-Net: Feature Augmentation with Separated Illumination Representation
• GANsfer Learning: Combining labelled and unlabelled data for GAN based data augmentation
• Augmenting correlation structures in spatial data using deep generative models
• Integrating GAN and Texture Synthesis for Enhanced Road Damage Detection
• Diffusion Based Augmentation for Captioning and Retrieval in Cultural Heritage
• Learning Stage-wise GANs for Whistle Extraction in Time-Frequency Spectrograms
• Medical Image Generation using Generative Adversarial Networks
• Teacher-Student Compression with Generative Adversarial Networks
• SAG-GAN: Semi-Supervised Attention-Guided GANs for Data Augmentation on Medical Images
• Data Augmentation with Manifold Exploring Geometric Transformations for Increased Performance and Robustness
• Enhancing the Performance of Practical Profiling Side-Channel Attacks Using Conditional Generative Adversarial Networks
• VAE-Info-cGAN: Generating Synthetic Images by Combining Pixel-level and Feature-level Geospatial Conditional Inputs
• Creating Realistic Anterior Segment Optical Coherence Tomography Images using Generative Adversarial Networks
• Data Augmentation for Object Detection via Differentiable Neural Rendering
• Constrained Generative Adversarial Network Ensembles for Sharable Synthetic Data Generation
• Measuring the Effectiveness of Voice Conversion on Speaker Identification and Automatic Speech Recognition Systems
• Synthetic Sample Selection via Reinforcement Learning
• Feeding the zombies: Synthesizing brain volumes using a 3D progressive growing GAN
• Synthesizing brain tumor images and annotations by combining progressive growing GAN and SPADE
• Diffusion-Weighted Magnetic Resonance Brain Images Generation with Generative Adversarial Networks and Variational Autoencoders: A Comparison Study
• Towards Better Long-range Time Series Forecasting using Generative Adversarial Networks
• On the Pitfalls of Learning with Limited Data: A Facial Expression Recognition Case Study
• Repaint: Improving the Generalization of Down-Stream Visual Tasks by Generating Multiple Instances of Training Examples
• Selective Synthetic Augmentation with Quality Assurance
• Unsupervised Text Embedding Space Generation Using Generative Adversarial Networks for Text Synthesis
• GAN Memory with No Forgetting
• Red-GAN: Attacking class imbalance via conditioned generation. Yet another perspective on medical image synthesis for skin lesion dermoscopy and brain tumor MRI
• A Survey of Techniques All Classifiers Can Learn from Deep Networks: Models, Optimizations, and Regularization
• Semi-supervised Semantic Segmentation of Prostate and Organs-at-Risk on 3D Pelvic CT Images
• tsBNgen: A Python Library to Generate Time Series Data from an Arbitrary Dynamic Bayesian Network Structure
• Semi-supervised Contrastive Learning for Label-efficient Medical Image Segmentation
• Adversarial Imitation Learning with Trajectorial Augmentation and Correction
• Ensembles of GANs for synthetic training data generation
• Learning Non-Unique Segmentation with Reward-Penalty Dice Loss
• Random Network Distillation as a Diversity Metric for Both Image and Text Generation
• OCT-GAN: Neural ODE-based Conditional Tabular GANs
• Cross-Domain Image Classification through Neural-Style Transfer Data Augmentation
• Evaluating Procedures for Establishing Generative Adversarial Network-based Stochastic Image Models in Medical Imaging
• PanDA: Panoptic Data Augmentation
• Learning Purified Feature Representations from Task-irrelevant Labels
• Man versus Machine: AutoML and Human Experts' Role in Phishing Detection
• Generative Residual Attention Network for Disease Detection
• Bringing Balance to Hand Shape Classification: Mitigating Data Imbalance Through Generative Models
• GAN based ball screw drive picture database enlargement for failure classification
• Data augmentation using generative networks to identify dementia
• Implicit Label Augmentation on Partially Annotated Clips via Temporally-Adaptive Features Learning
• Langevin Cooling for Domain Translation
• Synthetic Convolutional Features for Improved Semantic Segmentation
• Enhanced Magnetic Resonance Image Synthesis with Contrast-Aware Generative Adversarial Networks
• Towards All-around Knowledge Transferring: Learning From Task-irrelevant Labels
• Anonymization of labeled TOF-MRA images for brain vessel segmentation using generative adversarial networks
• Distance in Latent Space as Novelty Measure
• Regularising Deep Networks with Deep Generative Models
• Mining Domain Knowledge: Improved Framework towards Automatically Standardizing Anatomical Structure Nomenclature in Radiotherapy
• Safe Augmentation: Learning Task-Specific Transformations from Data
• Generative Imaging and Image Processing via Generative Encoder
• Partially fake it till you make it: mixing real and fake thermal images for improved object detection