MétaCan
Menu
Back to cohort
Record W2145892679 · doi:10.5339/qfarc.2014.hbpp0784

Linked Data Based Semantically Enabled Electronic Medical Record Systems

2014· article· en· W2145892679 on OpenAlexaff
Newres Al Haider, William William Van Woensel, Ahmad Marwan Ahmad, Syed Sibte Raza Abidi

Bibliographic record

VenueQatar Foundation Annual Research Conference Proceedings Volume 2014 Issue 1 · 2014
Typearticle
Languageen
FieldComputer Science
TopicSemantic Web and Ontologies
Canadian institutionsDalhousie University
Fundersnot available
KeywordsComputer scienceImplementationIdentifierSemantics (computer science)Clinical decision support systemRDFLinked dataResource (disambiguation)Unique identifierSemantic WebInformation retrievalDecision support systemData miningSoftware engineeringProgramming language

Abstract

fetched live from OpenAlex

Electronic Medical Record (EMR) systems are information systems keeping electronic versions of patients€' medical records. The use of EMR systems has been steadily increasing in recent years, due to many potential benefits. A fully functional EMR system can record patient demographic and chart data, keep track of vital signs, current medications, drug allergies and many other important facets of the patient's medical record. In an ideal scenario, such a system would also be able to handle and implement complex decision support tasks, such as clinical guideline implementations, drug interaction checking and critical alerts. One large issue with existing EMR system implementations is that the semantics of the information elements are not made explicit. Internal identifiers are often used as a placeholder for clinical concepts. This is a large problem when aiming to interact with the patient record, whether it is by physicians, new decision support implementations or by other health information systems. Without explicit semantics, both understanding and accessing the right information can require additional effort to adapt to these identifiers. We propose a Linked Data based approach to implement an EMR system to solve these issues. Linked Data, and in particular the Resource Description Framework (RDF) form the basis of the Semantic Web, which is designed to make the semantics of information both human and machine accessible. With RDF knowledge is represented as a set of triples, where each element of the triple can be an explict Uniform Resource Identifier (URI) with which internal and external resources can be linked. By linking to well defined medical terminologies, such as SNOMED CT, an RDF based approach can explicitly refer to a formalized set of concepts. Using databases for RDF documents, called triple stores, multiple large records can be stored as triple sets. With query languages that make use of triple based patterns, such as the SPARQL query language, the necessary clinical guidelines and other decision support can be implemented in a scalable way. We have evaluated this approach by implementing a Linked Data based version of an EMR system for Atrial Fibrillation (AF) patients. We have populated this system with automatically generated data that takes into account clinically feasible parameters. In addition a number of AF specific queries and decision support tasks were implemented to evaluate the scalability of the whole approach. Our results show that such a system has adequate performance for EMR systems deployed for a single small scale clinic, even on desktop level hardware. A key limiting factor is that an EMR can theoretically hold multiple years worth of very fine grained patient data, which can slow down the execution of various decision support tasks. However we have found that the portion of the dataset that is relevant to the decision support queries is often a very small subset of the overall record. A system that is dynamically able to divide the dataset over multiple data stores is needed to keep the system scalable for larger records with a higher number of patients.

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 imitation

Not 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.

metaresearch head score (Codex)0.012
metaresearch head score (Gemma)0.012
Version: codex-gemma-dda1882f352aValidation status: machine_predicted_unvalidated
Candidate categoriesMetaresearch, Meta-epidemiology (narrow), Scholarly communication, Open science, Insufficient payload (model declined to judge)
Consensus categoriesInsufficient payload (model declined to judge)
DomainCandidate signal: none · Consensus signal: none
Study designCandidate signal: Not applicable · Consensus signal: none
GenreCandidate signal: Empirical · Consensus signal: none
Teacher disagreement score0.967
Threshold uncertainty score1.000

Codex and Gemma teacher scores by category

CategoryCodexGemma
Metaresearch0.0120.012
Meta-epidemiology (narrow)0.0000.000
Meta-epidemiology (broad)0.0010.000
Bibliometrics0.0010.001
Science and technology studies0.0010.001
Scholarly communication0.0020.003
Open science0.0070.002
Research integrity0.0000.001
Insufficient payload (model declined to judge)0.0010.004

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.

Opus teacher head0.065
GPT teacher head0.358
Teacher spread0.293 · how far apart the two teachers sit on this one work
Validation statusscore_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from it

Classification

machine, unvalidated

Machine predicted; both teacher heads agree on what is shown here.

Study designNot applicable
Domainnot available
GenreEmpirical

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".

Quick stats

Citations0
Published2014
Admission routes1
Has abstractyes

Explore more

Same venueQatar Foundation Annual Research Conference Proceedings Volume 2014 Issue 1Same topicSemantic Web and OntologiesFrench-language works237,207