Architecture Overview
System components
ELADAIS is designed to support large-scale research by integrating and analysing health data. It consists of three key components:
ETL Processes
Extract, transform, and load raw data into a standardized OMOP database, ensuring compatibility and enabling federated data harmonization.
Microservices Inventory
The core analytical engine provides tools for data search, cohort definition, patient analysis, and predictive modeling, with a modular design for scalability.
User Interface
A user-friendly front-end for researchers and stakeholders to access, explore, and analyze health data seamlessly.
Technical Architecture
01
Exposure Layer
- Handles API calls, authentication, and access control using OpenAPI specifications.
- Ensures secure and standardized interactions between users/external systems and ELADAIS services.
- Supports API logging and runtime settings, managed through Python-based configuration tools.
02
Service Layer
- Hosts core business logic for cohort creation and analysis.
- Uses Dockerized microservices for modularity and scalability.
- Manages user requests from the Exposure Layer and coordinates interactions with the Data Interoperability Layer.
03
Data Interoperability Layer
- Oversees ETL workflows to transform and integrate healthcare data.
- Executes SQL and Python-based transformations to standardize and curate datasets.
- Facilitates the seamless integration of heterogeneous datasets within the OMOP framework.
Visual representation
ELADAIS services
A number of services will be developed which are independent of each other, but which can be run together through pipelines (cohorts, visualisation, nlp, etc).
FRONT END – Access to services
SECURITY LAYER – Double authenticator
MANAGEMENT MODULE
API rest ELADAIS
Search services
Exploratory data services
Generic services
OHDSI module
Design cohorts
Visualisation tool
Analytic module