Leveraging artificial intelligence and data science techniques in harmonizing, sharing, accessing and analyzing...

Area of Science:

• Health Informatics
• Artificial Intelligence
• Data Science

Background:

• COVID-19 pandemic in Rwanda generated fragmented SARS-COV-2/COVID-19 data.
• Data silos hinder disease understanding, progress monitoring, and evidence generation for prevention.

Purpose of the Study:

• Leverage AI and data science to harmonize fragmented datasets.
• Support Rwandan government needs in monitoring and predicting COVID-19 burden, including hospital admissions and infection rates.

Main Methods:

• Gathering electronic health records (EHRs) and COVID-19 testing data, linked with community survey data.
• Utilizing Observational Health Data Sciences and Informatics (OHDSI) open-source tools to harmonize EHRs to the Observational Medical Outcomes Partnership (OMOP) Common Data Model (CDM).
• Establishing a federated data network across 15 health facilities in Rwanda.

Main Results:

• A technical infrastructure with harmonized EHR data in OMOP CDM format at 15 participating health facilities.
• A central portal with a data catalogue and OHDSI tools for distributed studies.
• Integration of national COVID-19 registry and community survey data.

Conclusions:

• The project establishes Africa's first AI-leveraged OMOP CDM-based federated data network for data harmonization.
• The infrastructure is scalable for monitoring other pandemics, predicting outcomes, and planning tailored responses.