{"title":"Using Gensim TFIDF and LSI Models to Retrieve Potential Answers to Clinical Questions from Clinical Practice Guidelines.","authors":"Mohammadreza Azarpira","doi":"10.3233/SHTI250101","DOIUrl":null,"url":null,"abstract":"<p><p>General Practitioners often encounter unanswered medical questions about patient symptoms or treatments at the point of care. Despite advances in information technology and the availability of the Internet, it is estimated that half of these questions remain unanswered. This proportion has remained stable over time. International Clinical Practice Guidelines (CPGs), which contain recently updated evidence, are an optimal source of information; more than 90% of relevant clinical questions can be answered using these guidelines. However, the large volume of these CPGs limits their accessibility at the point of care. We developed an Information Retrieval System using serialized Gensim TFIDF and LSI models to extract relevant answers to clinical questions. The true answer to clinical questions can be found in the first six answers of the algorithm in 98% of cases. This algorithm can be helpful for general practitioners to take greater advantage of CPGs at the point of care.</p>","PeriodicalId":94357,"journal":{"name":"Studies in health technology and informatics","volume":"323 ","pages":"307-311"},"PeriodicalIF":0.0000,"publicationDate":"2025-04-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Studies in health technology and informatics","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.3233/SHTI250101","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 0
Abstract
General Practitioners often encounter unanswered medical questions about patient symptoms or treatments at the point of care. Despite advances in information technology and the availability of the Internet, it is estimated that half of these questions remain unanswered. This proportion has remained stable over time. International Clinical Practice Guidelines (CPGs), which contain recently updated evidence, are an optimal source of information; more than 90% of relevant clinical questions can be answered using these guidelines. However, the large volume of these CPGs limits their accessibility at the point of care. We developed an Information Retrieval System using serialized Gensim TFIDF and LSI models to extract relevant answers to clinical questions. The true answer to clinical questions can be found in the first six answers of the algorithm in 98% of cases. This algorithm can be helpful for general practitioners to take greater advantage of CPGs at the point of care.