A Semantic Web Framework To Computerize And Execute Clinical Guidelines: Towards The Handling Of Co-Morbidities In Clinical Decision Support Systems
Bibliographic record
Abstract
Background: A Canadian study recommends General Practitioners (GP) to use evidence based Clinical Guidelines (CG) when dealing with co-morbid cardiovascular diseases, in particular for the diagnosis and preliminary management of co-morbid Chronic Heart Failure (CHF) and Atrial Fibrillation (AF). Although paper-based Canadian CG exist for the management of CHF and AF, the challenge for physicians is to simultaneously apply multiple independent CG when dealing with patients having cardiovascular co-morbidities. Objective: The objective of this inter-disciplinary research program is to assist physicians in handling co-morbidities through a computerized clinical decision support framework that recommends evidence-based interventions based on the patient's health profile. We target decision support for the diagnosis and treatment of CHF, AF and co-morbid CHF-AF. Approach: We take a healthcare knowledge management approach to develop a Clinical Decision Support System (CDSS) for handling comorbid diseases. Our solution involves the development of institution-specific CP from a combination of CG, and then generate a CP knowledge model using a semantically-rich formalism—i.e. a CG ontology. The CG ontology semantically defines the clinical concepts in order to establish semantic interoperability between multiple CG. Next, we systematically align the ontologically-modeled CG of different diseases to realize a unified knowledge model that derives the evidence based recommendations for handling both single and co-morbid diseases. Our methodology entails the following steps: (i) knowledge identification to derive specialized disease-specific CG from existing evidence-based sources; (b) knowledge modeling to abstract medical and procedural knowledge from the CG; (c) knowledge representation to computerize the CG in terms of a semantically-rich CG ontology; (d) knowledge alignment to systematically synthesize multiple ontologically-modeled CG to develop a unified ontology-based CG knowledge model representing comorbid diseases; (e) knowledge execution to generate patient-specific recommendations, based on patient data, by reasoning over the aligned CG model; and (f) evaluation of the knowledge model and the recommendations produced in response to a range of clinical scenarios. Results: We present the COMET (Co-morbidity Ontological Modeling & ExecuTion) system to provide clinical decision support for three scenarios: (i) Cardiac Heart Failure (CHF); (ii) Atrial Fibrillation (AF); and (iii) co-morbidity of either AF or CHF. COMET is designed for GP in Nova Scotia and is accessible over the web. Evaluation: A pilot study was conducted to assess how well COMET meets the physician's needs to manage co-morbid CHF-AF. Conclusion: In conclusion, this project provides a solution for the complex problem of handling co-morbidities in a CDSS. Our solution is based on semantic modeling of disease-specific knowledge which extends the possibility of scaling up to include additional diseases and aligning their knowledge models to handle even further co-morbid situations. Our CG alignment approach helps (a) avoiding duplication of clinical tasks; (b) re-usability of diagnostic results; (c) determine compatibility of different clinical activities; and (d) standardization of care across multiple institutions. We believe that this project achieves knowledge translation whereby we have successfully computerized and translated paper-based CG so that they can now be operationalized at the point-of-care by GP to handle comorbid diseases.
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 distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.006 | 0.007 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.000 | 0.001 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.000 | 0.000 |
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 teacher head, 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".