Integrating Health Care Data in an Informatics for Integrating Biology & the Bedside (i2b2) Model Persisted Through Elasticsearch: Design, Implementation, and Evaluation in a French University Hospital
Bibliographic record
Abstract
BACKGROUND: The volume of digital data in health care is continually growing. In addition to its use in health care, the health data collected can also serve secondary purposes, such as research. In this context, clinical data warehouses (CDWs) provide the infrastructure and organization necessary to enhance the secondary use of health data. Various data models have been proposed for structuring data in a CDW, including the Informatics for Integrating Biology & the Bedside (i2b2) model, which relies on a relational database. However, this persistence approach can lead to performance issues when executing queries on massive data sets. OBJECTIVE: This study aims to describe the necessary transformations and their implementation to enable i2b2's search engine to perform the phenotyping task using data persistence in a NoSQL Elasticsearch database. METHODS: This study compares data persistence in a standard relational database with a NoSQL Elasticsearch database in terms of query response and execution performance (focusing on counting queries based on structured data, numerical data, and free text, including temporal filtering) as well as material resource requirements. Additionally, the data loading and updating processes are described. RESULTS: We propose adaptations to the i2b2 model to accommodate the specific features of Elasticsearch, particularly its inability to perform joins between different indexes. The implementation was tested and evaluated within the CDW of Bordeaux University Hospital, which contains data on 2.5 million patients and over 3 billion observations. Overall, Elasticsearch achieves shorter query execution times compared with a relational database, with particularly significant performance gains for free-text searches. Additionally, compared with an indexed relational database (including a full-text index), Elasticsearch requires less disk space for storage. CONCLUSIONS: We demonstrate that implementing i2b2 with Elasticsearch is feasible and significantly improves query performance while reducing disk space usage. This implementation is currently in production at Bordeaux University Hospital.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.007 | 0.011 |
| Meta-epidemiology (narrow) | 0.001 | 0.001 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.001 | 0.002 |
| Science and technology studies | 0.001 | 0.001 |
| Scholarly communication | 0.003 | 0.003 |
| Open science | 0.004 | 0.003 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.003 | 0.001 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".