Font Size: a A A

Based On Association Rules Mining Applications Of Electronic Medical Records

Posted on:2013-11-11Degree:MasterType:Thesis
Country:ChinaCandidate:Y CengFull Text:PDF
GTID:2248330374975013Subject:Software engineering
Abstract/Summary:PDF Full Text Request
With the deepening of the hospital informatization construction, information system hascovered the hospital business development of various departments, especially in the processof patient’s diagnosis and treatment. Registration, writing of medical records’ documents, aswell as disbursement and settlement all are electronically stored in the database ofinformation system which accumulates abundant medical information. Brain specializdhospital especially accumulates rich information of Electronic Medical Records ofspecialized diseases of nervous system. In recent years, Data Mining from electronic medicalrecords has gradually become a hot topic in research, therefore, actively explore the scientificand practical data mining techniques to find out valuable rules in the plentiful data ofelectronic medical records to provide scientific evidence for clinical experts in diseasediagnosis, treatment and clinical scientific research to improve the level of diagnosis andtreatment is of great significance and broad prospect.Based on the research of data mining with association rules technique, this paper hascombined with the hospital information system of Guangdong999Brain Hospital to carry outthe application research of aassociation rules-based data mining from electronic medicalrecords. The main contents include:1. analysis of the application of data mining technique athome and abroad in the medical field. It also has introduced of the architecture andapplication progress of hospital information system. For the analysis of data of electronicmedical records, this paper has stated the methods and process of data mining, as well asseveral issues needed to pay attention.2. Introduction of basic concepts, classification andprocess of data mining using association rules. Description with application examples of twomain association rules: Apriori algorithm and FP-growth algorithm.3. This paper emphasizedthe description of the implementation of the Apriori algorithm with examples, as well asdescription of limitation and optimization methods. As the electronic medical records withthe features of complex data structure, big data capacity, fast data update, to decrease thevolume of scanned data and reduce scan of transaction database, compares to classic Apriorialgorithm, using improved Apriori algorith and has made experimental analysis of actual datain mining frequent itemsets has improved the efficiency of data mining.4. Take the2011 medical records data of discharged from hospital of electronic medical records system ofGuangdong999Brain Hospital as an example. This paper has analyzed the methods of datapreprocessing, the process of operating and the law between epilepsy and related diseases, aswell as the evaluation of the result of data mining.
Keywords/Search Tags:association rules, data mining, electronic medical records, Apriori algorithm
PDF Full Text Request
Related items