Related Experiment Videos
Practical Implications for Using Laboratory Data: Research over Federated Networks
David Rubio Ruiz1,2, Aída Muñoz Monjas1,2, Paula Bermejo Bernardo1
1Biomedical Informatics Group, Universidad Politécnica de Madrid, Madrid, Spain.
Introduction:
Semantic standardization is essential, but not sufficient, to enable research over real-world data networks. Even when harmonized, differences in data availability, granularity, and clinical practice can introduce subtle biases that affect downstream analyses and AI model development.
Methods:
This work explores laboratory data availability and testing practices using LOINC-coded data from TriNetX, a global federated research network. Significant LOINC part variations were identified using statistical dispersion assessment.
Results:
Data from 172 healthcare institutions was evaluated. Representative examples were further analysed demonstrating the relevance of specimen and method availability, as well as property and component country-specific differences.
Conclusions:
Acknowledging data variability is key to minimizing biases introduced during AI model development and research over clinical real-world data.
Related Concept Videos
02:26Protein Networks
These interactions can be represented through maps depicting protein-protein interaction networks, represented as nodes and edges. Nodes are circles that are representative of a protein,...
01:29Issues And Trends In Healthcare Delivery System
Cost Containment
Payment for healthcare services has historically promoted adoption of costly and often unnecessary or inefficient...