Healthcare Decision Support System for Administration of Chronic Diseases / 대한의료정보학회지

OBJECTIVES: A healthcare decision-making support model and rule management system is proposed based on a personalized rule-based intelligent concept, to effectively manage chronic diseases. METHODS: A Web service was built using a standard message transfer protocol for interoperability of personal health records among healthcare institutions. An intelligent decision service is provided that analyzes data using a service-oriented healthcare rule inference function and machine-learning platform; the rules are extensively compiled by physicians through a developmental user interface that enables knowledge base construction, modification, and integration. Further, screening results are visualized for the self-intuitive understanding of personal health status by patients. RESULTS: A recommendation message is output through the Web service by receiving patient information from the hospital information recording system and object attribute values as input factors. The proposed system can verify patient behavior by acting as an intellectualized backbone of chronic diseases management; further, it supports self-management and scheduling of screening. CONCLUSIONS: Chronic patients can continuously receive active recommendations related to their healthcare through the rule management system, and they can model the system by acting as decision makers in diseases management; secondary diseases can be prevented and health management can be performed by reference to patient-specific lifestyle guidelines.

Comparison of Machine Learning Algorithms for Classification of the Sentences in Three Clinical Practice Guidelines / 대한의료정보학회지

OBJECTIVES: Clinical Practice Guidelines (CPGs) are an effective tool for minimizing the gap between a physician's clinical decision and medical evidence and for modeling the systematic and standardized pathway used to provide better medical treatment to patients. METHODS: In this study, sentences within the clinical guidelines are categorized according to a classification system. We used three clinical guidelines that incorporated knowledge from medical experts in the field of family medicine. These were the seventh report of the Joint National Committee (JNC7) on Prevention, Detection, Evaluation, and Treatment of High Blood Pressure from the National Heart, Lung, and Blood Institute; the third report of the National Cholesterol Education Program (NCEP) Expert Panel on Detection, Evaluation, and Treatment of High Blood Cholesterol in Adults from the same institution; and the Standards of Medical Care in Diabetes 2010 report from the American Diabetes Association. Three annotators each tagged 346 sentences hand-chosen from these three clinical guidelines. The three annotators then carried out cross-validations of the tagged corpus. We also used various machine learning-based classifiers for sentence classification. RESULTS: We conducted experiments using real-valued features and token units, as well as a Boolean feature. The results showed that the combination of maximum entropy-based learning and information gain-based feature extraction gave the best classification performance (over 98% f-measure) in four sentence categories. CONCLUSIONS: This result confirmed the contribution of the feature reduction algorithm and optimal technique for very sparse feature spaces, such as the sentence classification problem in the clinical guideline document.

A Data Warehouse Based Retrospective Post-marketing Surveillance Method: A Feasibility Test with Fluoxetine / 대한의료정보학회지

OBJECTIVE: Post-marketing surveillance (PMS) is an adverse events monitoring practice of pharmaceutical drugs on the market. Traditional PMS methods are labor intensive and expensive to perform, because they are largely based on manual work including phone-calling, mailing, or direct visits to relevant subjects. The objective of this study was to develop and validate a PMS methodology based on the clinical data warehouse (CDW). METHODS: We constructed a archival DB using a hospital information system and a refined CDW from three different hospitals. Fluoxetine hydrochloride, an antidepressant, was selected as the target monitoring drug. Corrected QT prolongation on ECG was selected as the target adverse outcome. The Wilcoxon signed rank test was performed to analyze the difference in the corrected QT interval before and after the target drug administration. RESULTS: A refined CDW was successfully constructed from three different hospitals. Table specifications and an entity-relation diagram were developed and are presented. A total of 13 subjects were selected for monitoring. There was no statistically significant difference in the QT interval before and after target drug administration (p=0.727). CONCLUSION: The PMS method based on CDW was successfully performed on the target drug. This IT-based alternative surveillance method might be beneficial in the PMS environment of the future.