In Part I of interoperability, I touched upon federated and centralized models and RDF and NoSQL paradigms for data persistence in centralized models. So what exactly is NoSQL in eHealth? NoSQL or ‘Not Only SQL’, unlike the structured tabular data storage method of relational database, use data storage methods such as key-value, graph, or document.
If you, like me, were first introduced to the world of EHRs through open source products such as McMaster’s OSCAR or WHO’s OpenMRS, it would be difficult to think beyond a relational database system with the household names of Oracle, SQL Server and MySQL. As we discussed before, structured relational database systems are not suitable for clinical workflows. This has led to the rejection of most EHR systems by the clinicians.
Whether EWD.js succeeds in reigniting interest in M, or the new generation NoSQL databases make their inevitable foray into eHealth, it is high time for a major overhaul in the existing EHR data models.
- MongoDB Update Addresses User Criticism (datacenterknowledge.com)
- Modern SQL in PostgreSQL [and other databases] (Use-The-Index-Luke.com)
- NoSQL – Creating Waves In The Database Community (spec-india.com)
- Named Entity Recognition using LLMs: a cTakes alternative? - September 1, 2023
- Distilling LLMs to small task-specific models - August 24, 2023
- Kedro for multimodal machine learning in healthcare - January 25, 2023