# Overcoming Barriers to Data Sharing with Medical Image Generation: A Comprehensive Evaluation

arXiv: https://arxiv.org/abs/2012.03769 · HTML: https://mlsanity.com/paper/2012.03769/

## Key facts

- Cited by 36 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 29 Nov 2020.

## Built on by (most cited first, 30 of 36)

- [Extracting Training Data from Diffusion Models](https://mlsanity.com/paper/2301.13188/) (2023)
- [Self-Consuming Generative Models Go MAD](https://mlsanity.com/paper/2307.01850/) (2023)
- [Generative Models as a Data Source for Multiview Representation Learning](https://mlsanity.com/paper/2106.05258/) (2021)
- [RoentGen: Vision-Language Foundation Model for Chest X-ray Generation](https://mlsanity.com/paper/2211.12737/) (2022)
- [A Multifaceted Benchmarking of Synthetic Electronic Health Record Generation Models](https://mlsanity.com/paper/2208.01230/) (2022)
- [Towards Theoretical Understandings of Self-Consuming Generative Models](https://mlsanity.com/paper/2402.11778/) (2024)
- [Generative AI for Synthetic Data Across Multiple Medical Modalities: A Systematic Review of Recent Developments and Challenges](https://mlsanity.com/paper/2407.00116/) (2024)
- [Augmenting medical image classifiers with synthetic data from latent diffusion models](https://mlsanity.com/paper/2308.12453/) (2023)
- [Conditional Generation of Medical Time Series for Extrapolation to Underrepresented Populations](https://mlsanity.com/paper/2201.08186/) (2022)
- [Assessing the ability of generative adversarial networks to learn canonical medical image statistics](https://mlsanity.com/paper/2204.12007/) (2022)
- [Unconditional Latent Diffusion Models Memorize Patient Imaging Data: Implications for Openly Sharing Synthetic Data](https://mlsanity.com/paper/2402.01054/) (2024)
- [What Does DALL-E 2 Know About Radiology?](https://mlsanity.com/paper/2209.13696/) (2022)
- [Assessing the capacity of a denoising diffusion probabilistic model to reproduce spatial context](https://mlsanity.com/paper/2309.10817/) (2023)
- [Synthetic Data in Radiological Imaging: Current State and Future Outlook](https://mlsanity.com/paper/2407.01561/) (2024)
- [A Generalist Model for Diverse Text-Guided Medical Image Synthesis](https://mlsanity.com/paper/2405.09806/) (2024)
- [Enhancing Fetal Plane Classification Accuracy with Data Augmentation Using Diffusion Models](https://mlsanity.com/paper/2501.15248/) (2025)
- [A Domain Translation Framework with an Adversarial Denoising Diffusion Model to Generate Synthetic Datasets of Echocardiography Images](https://mlsanity.com/paper/2403.04612/) (2024)
- [Report on the AAPM Grand Challenge on deep generative modeling for learning medical image statistics](https://mlsanity.com/paper/2405.01822/) (2024)
- [Scorecards for Synthetic Medical Data Evaluation and Reporting](https://mlsanity.com/paper/2406.11143/) (2024)
- [Clinical Evaluation of Medical Image Synthesis: A Case Study in Wireless Capsule Endoscopy](https://mlsanity.com/paper/2411.00178/) (2024)
- [Content-Aware Differential Privacy with Conditional Invertible Neural Networks](https://mlsanity.com/paper/2207.14625/) (2022)
- [Multi-omics Prediction from High-content Cellular Imaging with Deep Learning](https://mlsanity.com/paper/2306.09391/) (2023)
- [MedShift: identifying shift data for medical dataset curation](https://mlsanity.com/paper/2112.13885/) (2021)
- [RadImageGAN -- A Multi-modal Dataset-Scale Generative AI for Medical Imaging](https://mlsanity.com/paper/2312.05953/) (2023)
- [Semantics versus Identity: A Divide-and-Conquer Approach towards Adjustable Medical Image De-Identification](https://mlsanity.com/paper/2507.21703/) (2025)
- [Early Stopping Criteria for Training Generative Adversarial Networks in Biomedical Imaging](https://mlsanity.com/paper/2405.20987/) (2024)
- [Contrastive Learning from Synthetic Audio Doppelg"angers](https://mlsanity.com/paper/2406.05923/) (2024)
- [Generative models of MRI-derived neuroimaging features and associated dataset of 18,000 samples](https://mlsanity.com/paper/2407.12897/) (2024)
- [Minimum Data, Maximum Impact: 20 annotated samples for explainable lung nodule classification](https://mlsanity.com/paper/2508.00639/) (2025)
- [Medical Image De-Identification Resources: Synthetic DICOM Data and Tools for Validation](https://mlsanity.com/paper/2508.01889/) (2025)

## Abstract (arXiv)

Privacy concerns around sharing personally identifiable information are a major practical barrier to data sharing in medical research. However, in many cases, researchers have no interest in a particular individual's information but rather aim to derive insights at the level of cohorts. Here, we utilize Generative Adversarial Networks (GANs) to create derived medical imaging datasets consisting entirely of synthetic patient data. The synthetic images ideally have, in aggregate, similar statistical properties to those of a source dataset but do not contain sensitive personal information. We assess the quality of synthetic data generated by two GAN models for chest radiographs with 14 different radiology findings and brain computed tomography (CT) scans with six types of intracranial hemorrhages. We measure the synthetic image quality by the performance difference of predictive models trained on either the synthetic or the real dataset. We find that synthetic data performance disproportionately benefits from a reduced number of unique label combinations. Our open-source benchmark also indicates that at low number of samples per class, label overfitting effects start to dominate GAN training. We additionally conducted a reader study in which trained radiologists do not perform better than random on discriminating between synthetic and real medical images for intermediate levels of resolutions. In accordance with our benchmark results, the classification accuracy of radiologists increases at higher spatial resolution levels. Our study offers valuable guidelines and outlines practical conditions under which insights derived from synthetic medical images are similar to those that would have been derived from real imaging data. Our results indicate that synthetic data sharing may be an attractive and privacy-preserving alternative to sharing real patient-level data in the right settings.