Development of a machine learning algorithm for generating realistic synthetic electronic healthcare records (Prosate and Breast)
In the project PLCO-529, the algorithm for the creating of the realistic Lung Cancer Patients is created. Unfortunately, not all the time the sufficient amount of data about the specific cancer is available. The goal of this project is to validate whenever the algorithm created in PLCO-529 can be used to create realistic healthcare records, for patients with other types of cancer (for example for Prostate and Breast cancer). This project also aims to validate whenever it is possible to use a combination of the data (for example Lung and Prostate) to create realistic healthcare records if not enough patient data is available.
- Implement an algorithm to gather causal dependencies between patient characteristics, diagnostic & therapeutic procedures, and patient outcomes (complications, survival)
- Validation of the patient generator against real patient population from available data
- Integration of domain transfer between Lung- Prostate - Breast cancer
Daniel Stromer FAU Erlangen Nürnberg daniel.stromer@fau.de
Oliver Haas FAU Erlangen Nürnberg oliver.haas@fau.de