System and method for computer-aided synthesis of simulated health data
Patent Information
- Application Number
- DE602019074959
- Authority / Receiving Office
- DE · DE
- Patent Type
- Patents
- Current Assignee / Owner
- Priority Date
- 2018-12-19
- Filing Date
- 2019-12-19
- Publication Date
- 2025-08-27
- Estimated Expiration
- 2039-12-19
AI Technical Summary
Existing healthcare research faces challenges with Electronic Health Record (EHR) data due to Protected Health Information (PHI) privacy regulations, inconsistent data formats, limited data availability for rare diseases, and insufficient information for complex health outcomes, hindering effective analysis and sharing of data across healthcare providers.
A system utilizing machine learning techniques to generate synthetic EHR data through state transition machines, simulating health-related events over a patient's lifespan, and adjusting models to ensure similarity with actual EHR data, thereby creating representative synthetic datasets without PHI.
This approach enables the generation of representative synthetic EHR data that mirrors population health trends, reducing PHI risks and facilitating healthcare research and policy analysis while overcoming data integration and consistency issues.