<abstract xmlns="http://www.w3.org/1999/xhtml"><p>A semantic-based search engine for clinical data would be a substantial aid for hospitals to provide support for clinical practitioners. Since electronic medical records of patients contain a variety of information, there is a need to extract meaningful patterns from the Patient Medical Records (PMR). The proposed work matches patients to relevant clinical practice guidelines (CPGs) by matching their medical records with the CPGs. However in both PMR and CPG, the information pertaining to symptoms, diseases, diagnosis procedures and medicines is not structured and there is a need to pre-process and index the information in a meaningful way. In order to reduce manual effort to match to the clinical guidelines, this work automatically extracts the clinical guidelines from the PDF documents using a set of regular expression rules and indexes them with a multi-field index using Lucene. We have attempted a multi-field Lucene search and ontology-based advanced search, where the PMR is mapped to SNOMED core subset to find the important concepts. We found that the ontology-based search engine gave more meaningful results for specific queries when compared to term based search.</p></abstract>

A semantic-based search engine for clinical data would be a substantial aid for hospitals to provide support for clinical practitioners. Since electronic medical records of patients contain a variety of information, there is a need to extract meaningful patterns from the Patient Medical Records (PMR). The proposed work matches patients to relevant clinical practice guidelines (CPGs) by matching their medical records with the CPGs. However in both PMR and CPG, the information pertaining to symptoms, diseases, diagnosis procedures and medicines is not structured and there is a need to pre-process and index the information in a meaningful way. In order to reduce manual effort to match to the clinical guidelines, this work automatically extracts the clinical guidelines from the PDF documents using a set of regular expression rules and indexes them with a multi-field index using Lucene. We have attempted a multi-field Lucene search and ontology-based advanced search, where the PMR is mapped to SNOMED core subset to find the important concepts. We found that the ontology-based search engine gave more meaningful results for specific queries when compared to term based search.

A Semantic Multi-Field Clinical Search for Patient Medical Records

Linguistics and Multilingual Studies, School of Humanities

Cybernetics and Information Technologies

This work is licensed under the Creative Commons Attribution-NonCommercial-NoDerivatives 3.0 License.

{"article-title":"A Semantic Multi-Field Clinical Search for Patient Medical Records"}

A semantic-based search engine for clinical data would be a substantial aid for hospitals to provide support for clinical practitioners. Since electronic...

A Semantic Multi-Field Clinical Search for Patient Medical Records

Online veröffentlicht: 30. März 2018

Seitenbereich: 171 - 182

Eingereicht: 31. Mai 2017

Akzeptiert: 31. Jan. 2018

DOI: https://doi.org/10.2478/cait-2018-0014

Schlüsselwörter
Semantic similarity, application to NLP, SNOMED ontology, information extraction and text simplification

© 2018 E Umamaheswari Vasanthakumar et al., published by De Gruyter Open

This work is licensed under the Creative Commons Attribution-NonCommercial-NoDerivatives 3.0 License.

A Semantic Multi-Field Clinical Search for Patient Medical Records

Online veröffentlicht: 30. März 2018

Seitenbereich: 171 - 182

Eingereicht: 31. Mai 2017

Akzeptiert: 31. Jan. 2018

DOI: https://doi.org/10.2478/cait-2018-0014

SchlüsselwörterSemantic similarity, application to NLP, SNOMED ontology, information extraction and text simplification

© 2018 E Umamaheswari Vasanthakumar et al., published by De Gruyter Open

This work is licensed under the Creative Commons Attribution-NonCommercial-NoDerivatives 3.0 License.

Schlüsselwörter
Semantic similarity, application to NLP, SNOMED ontology, information extraction and text simplification