Geometrically-aggregated training samples: Leveraging summary statistics to enable healthcare data democratization
Preprint
Abstract
Healthcare data is highly sensitive and confidential, with strict regulations and laws to protect patient privacy and security. However, these regulations impede the access of healthcare data to a wider AI research community. As a result, AI healthcare research is often dominated by organisations with access to larger datasets or limited to silo-based development, where models are trained and evaluated on a limited population. Taking inspiration from the non-sensitive nature of the summary statistics (mean, variance, etc.) of healthcare data, this paper proposes geometrically-aggregated training samples (GATS) where each training sample is a convex combination of multiple patients’ characteristics. Thus, mappings from patients to any constructed sample are highly convoluted, preserving patient privacy. We demonstrate that these “summary training units” provide effective training on different tabular and time-series datasets (CURIAL, UCI Adult, and eICU), and indeed behave as a summary of the original training datasets. This approach takes important steps towards data accessibility and democratization.Keywords
This publication has 22 references indexed in Scilit:
- SynSigGAN: Generative Adversarial Networks for Synthetic Biomedical Signal GenerationBiology, 2020
- Empirical Analysis Of Overfitting And Mode Drop In Gan Training2020
- Benchmarking machine learning models on multi-centre eICU critical care datasetPLOS ONE, 2020
- Lost in Anonymization — A Data Anonymization Reference Classification Merging Legal and Technical ConsiderationsJournal of Law, Medicine & Ethics, 2020
- Seeing What a GAN Cannot Generate2019
- Generative Adversarial Network for Synthetic Time Series Data Generation in Smart Grids2018
- Unifying cancer and normal RNA sequencing data from different sourcesScientific Data, 2018
- Re-identification attacks—A systematic literature reviewInternational Journal of Information Management, 2016
- Anonymising and sharing individual patient dataBMJ, 2015
- PhysioBank, PhysioToolkit, and PhysioNetCirculation, 2000