medigan: a Python library of pretrained generative models for medical image synthesis

Abstract

Synthetic data generated by generative models can enhance the performance and capabilities of data-hungry deep learning models in medical imaging. However, there is (1) limited availability of (synthetic) datasets and (2) generative models are complex to train, which hinders their adoption in research and clinical applications. To reduce this entry barrier, we propose medigan, a one-stop shop for pretrained generative models implemented as an open-source framework-agnostic Python library. medigan allows researchers and developers to create, increase, and domain-adapt their training data in just a few lines of code. Guided by design decisions based on gathered end-user requirements, we implement medigan based on modular components for generative model (i) execution, (ii) visualisation, (iii) search & ranking, and (iv) contribution. The library's scalability and design is demonstrated by its growing number of integrated and readily-usable pretrained generative models consisting of 21 models utilising 9 different Generative Adversarial Network architectures trained on 11 datasets from 4 domains, namely, mammography, endoscopy, x-ray, and MRI. Furthermore, 3 applications of medigan are analysed in this work, which include (a) enabling community-wide sharing of restricted data, (b) investigating generative model evaluation metrics, and (c) improving clinical downstream tasks. In (b), extending on common medical image synthesis assessment and reporting standards, we show Fréchet Inception Distance variability based on image normalisation and radiology-specific feature extraction.

32 pages, 7 figures

References in corpus (15)

• PyTorch: An Imperative Style, High-Performance Deep Learning Library
• Conditional Generative Adversarial Nets
• GANs Trained by a Two Time-Scale Update Rule Converge to a Local Nash Equilibrium
• NICE: Non-linear Independent Components Estimation
• MONAI: An open-source framework for deep learning in healthcare
• Towards Faster and Stabilized GAN Training for High-fidelity Few-shot Image Synthesis
• CrossMoDA 2021 challenge: Benchmark of Cross-Modality Domain Adaptation techniques for Vestibular Schwannoma and Cochlea Segmentation
• DLTK: State of the Art Reference Implementations for Deep Learning on Medical Images
• Carbon Footprint of Selecting and Training Deep Learning Models for Medical Image Analysis
• High-resolution synthesis of high-density breast mammograms: Application to improved fairness in deep learning based mass detection
• Unsupervised Brain Anomaly Detection and Segmentation with Transformers
• Brain Imaging Generation with Latent Diffusion Models
• Mimicry: Towards the Reproducibility of GAN Research
• StudioGAN: A Taxonomy and Benchmark of GANs for Image Synthesis
• Sharing Generative Models Instead of Private Data: A Simulation Study on Mammography Patch Classification

Cited by in corpus (5)

• High-resolution synthesis of high-density breast mammograms: Application to improved fairness in deep learning based mass detection
• Feature Extraction for Generative Medical Imaging Evaluation: New Evidence Against an Evolving Trend
• Report of the Medical Image De-Identification (MIDI) Task Group -- Best Practices and Recommendations
• Pre- to Post-Contrast Breast MRI Synthesis for Enhanced Tumour Segmentation
• Fréchet Radiomic Distance (FRD): A Versatile Metric for Comparing Medical Imaging Datasets