Integrating existing large scale medical laboratory data into the semantic web framework
Bibliographic record
Abstract
Semantic Web technologies have shown to have great potential in many different domains, to facilitate knowledge representation, exchange and reasoning, in a formal and yet both human and machine understandable way. In particular, within the health domain, they enable knowledge integration and understanding by explicitly defining and linking concepts and relationships using ontologies to information within clinical knowledge bases. This additional metadata also allows for automated decision support and semantic based analytics to be implemented, that facilitate improved healthcare at a lower cost. Unfortunately many existing datasets in healthcare environments are still stored in relational databases, as opposed to using semantic technologies. Due to this, the link with explicit metadata is often lacking or non-existent. Furthermore, both the databases and the clinical terminologies can be considerably large, making the mapping and subsequent uses of the information a difficult process. In a full fledged decision support system the level and accuracy of the mapping can greatly influence the effectiveness of any subsequent analysis and decision support tasks. This is especially true in clinical scenarios, where very large and complex sets of terms need to be mapped to relational databases. In this paper we aim to provide a general approach for interlinking relational data with clinical ontology based metadata that allows for a fine grade evaluation, with respect to the mapping's impact on analytics. We evaluate our approach by mapping information from clinical terminologies, such as SNOMED CT, to a large laboratory dataset contained in a relational database, with the goal of creating a full fledged, semantically enabled, analytics and decision support system.
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 | 0.009 | 0.012 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.002 |
| Bibliometrics | 0.007 | 0.007 |
| Science and technology studies | 0.001 | 0.002 |
| Scholarly communication | 0.006 | 0.008 |
| Open science | 0.003 | 0.006 |
| Research integrity | 0.002 | 0.002 |
| Insufficient payload (model declined to judge) | 0.001 | 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".