Improving Time-series Generative Adversarial Networks (GANs) for Generating Electronic Health Records (EHRs)
| Title | Improving Time-series Generative Adversarial Networks (GANs) for Generating Electronic Health Records (EHRs) |
|---|---|
| Summary | Synthetic Electronic Health Records |
| Keywords | Time-series, GANs, EHR, Tabular data |
| TimeFrame | |
| References | [[References::[1] Yoon, Jinsung, Daniel Jarrett, and Mihaela Van der Schaar. "Time-series generative adversarial networks." Advances in neural information processing systems 32 (2019).
[2] Brophy, Eoin, et al. "Generative adversarial networks in time series: A systematic literature review." ACM Computing Surveys 55.10 (2023): 1-31. [3] Johnson, Alistair, et al. "Mimic-iv." PhysioNet. Available online at: https://physionet. org/content/mimiciv/1.0/(accessed August 23, 2021) (2020).]] |
| Prerequisites | |
| Author | |
| Supervisor | Amira Soliman, Atiye Sadat Hashemi |
| Level | Master |
| Status | Open |
The shift towards Electronic Health Records (EHRs) in both research and clinical practice is making it easier to give patients the care they need without adding extra costs [3]. Accessing highly qualified big EHR datasets for improving deep learning-based models is a big challenge. Generative adversarial networks (GANs) are a class of deep learning models that can generate realistic and high-quality synthetic data that can be used for various applications. A time-series GAN is required for generating patient data since this type of data is considered temporal and is being measured over time [2]. In this master thesis, we aim to improve time series generative adversarial networks [1] in terms of EHR’s temporal and spatial aspects.
