Federated Study Data Translation for Privacy-Safe Aggregation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing federated studies face challenges in ensuring data consistency and privacy across multiple sites due to variations in data production, collection, and legal restrictions, leading to inaccuracies and privacy risks.
Innovation Solution
A computer-implemented system with a predefined common data model, data translation, and a central repository to unify data translations, ensuring data compatibility and privacy, while using a site of analysis to process and generate reports.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If data is transmitted between multiple sites for federated studies, then data consistency and collaboration are improved, but privacy risks and legal compliance issues worsen
Solution Approach 1:
The patent introduces a central repository as an intermediary that stores data translation elements (mappings, extensions, validations) and provides them to translation clients at each site. This mediator enables consistent data translation across sites without requiring direct data transmission between sites, thus maintaining data consistency while protecting privacy.
Solution Approach 2:
The patent extracts the sensitive individual participant data from the data transmission process. Instead of transmitting raw data between sites, only translated aggregate data is transmitted. The translation process extracts and removes identifying information while preserving research value, resolving the contradiction between data sharing and privacy protection.
2Adaptability or versatility
If data translation is performed at each site independently, then data processing flexibility is improved, but data compatibility and consistency worsen
Solution Approach 1:
The patent implements local quality by allowing each site to have its own translation client that can handle site-specific data formats and requirements. Each translation client is customized to the local site's data structure while still producing output that conforms to the common data model, thus maintaining both local flexibility and global compatibility.
Solution Approach 2:
The central repository acts as a mediator that provides standardized data translation elements to all sites. These elements ensure that regardless of local variations in data collection, all sites translate their data consistently according to the same mappings and validations, thereby ensuring data compatibility across diverse local environments.
3Measurement precision
If individual participant data is transmitted for analysis, then analysis accuracy is improved, but legal compliance and privacy protection worsen
Solution Approach 1:
The patent extracts identifying information from individual participant data through the translation process. The translation client removes or anonymizes personally identifiable information while retaining the clinical and research data needed for analysis. This extraction enables analysis to proceed with sufficient accuracy while complying with privacy laws.
Solution Approach 2:
Instead of transmitting original individual participant data, the system creates translated copies that contain the essential research information in a standardized format. These copies are used for analysis at the central site, providing sufficient accuracy for research purposes while eliminating privacy and legal compliance issues associated with transmitting original data.
Data Source
AI summary
A computer-implemented federated studies system and method are disclosed. The system includes: a predefined common data model associated with a study; a set of sites of origin, each associated with a set of input devices for acquiring study data from subjects and a study data device. Each study data device includes: a database for storing study data derived from input devices; a data model translator for translating study data to a common data model format; a common model database for storing, in the data model, individual participant values derived from study data; a study compute module for determining site aggregate values from individual participant values; and an aggregate database for storing site aggregate values. A site of analysis is associated with a study analysis device that includes: an analysis compute module for processing site aggregate values received from sites of origin; and a reporting module for generating study reports.