Summary of Protect and Extend — Using Gans For Synthetic Data Generation Of Time-series Medical Records, by Navid Ashrafi et al.
Protect and Extend – Using GANs for Synthetic Data Generation of Time-Series Medical Records
by Navid Ashrafi, Vera Schmitt, Robert P. Spang, Sebastian Möller, Jan-Niklas Voigt-Antons
First submitted to arxiv on: 21 Feb 2024
Categories
- Main: Machine Learning (cs.LG)
- Secondary: Artificial Intelligence (cs.AI); Cryptography and Security (cs.CR)
GrooveSquid.com Paper Summaries
GrooveSquid.com’s goal is to make artificial intelligence research accessible by summarizing AI papers in simpler terms. Each summary below covers the same AI paper, written at different levels of difficulty. The medium difficulty and low difficulty versions are original summaries written by GrooveSquid.com, while the high difficulty version is the paper’s original abstract. Feel free to learn from the version that suits you best!
| Summary difficulty | Written by | Summary |
|---|---|---|
| High | Paper authors | High Difficulty Summary Read the original abstract here |
| Medium | GrooveSquid.com (original content) | Medium Difficulty Summary The abstract discusses the importance of preserving private user data, particularly in sensitive services like IT-based health services. To achieve this, synthetic data generation using Generative Adversarial Networks (GANs) has become a popular approach over traditional anonymization techniques. The paper compares state-of-the-art GAN-based models for generating synthetic time-series medical records of dementia patients, focusing on privacy preservation and quality. Predictive modeling, autocorrelation, and distribution analysis are used to evaluate the generated data’s Quality of Generating (QoG). Membership inference attacks are applied to assess potential data leakage risks. The results show that the PPGAN model outperforms others in terms of privacy preservation while maintaining an acceptable level of QoG. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary Synthetic medical records for dementia patients need to be generated without privacy concerns. This paper compares different models that use Generative Adversarial Networks (GANs) to create these records. They check how well the models work and if they keep patient data private. The best model, called PPGAN, keeps sensitive information safe while still making good predictions. |
Keywords
* Artificial intelligence * Gan * Inference * Synthetic data * Time series
