Research On The Application Of Association Analysis In Clinical Data Of Coronary Heart Disease

With the popularization and deepening in construction of hospital informatization, almost all the large and medium-sized hospitals and many small-scale hospitals in China have build up hospital information systems. These systems have run for a long time, and stored mass clinical data. Although traditional database technique is able to provide functions for data query and statistics, it has little ability to translate data into clinical knowledge, which limits the deep level utilization of data. So, how to extract valuable knowledge from data to serve clinical research becomes one of the problems crying out for solutions faced by medical information workers.Data mining is the nontrivial process of identifying valid, novel, potentially useful, and ultimately understandable patterns or knowledge in data. And association analysis is one of the most important tasks in data mining, which emphasizes on analyzing associations or relationships between itemsets in data. So, data mining, especially association analysis, should be promising in deep level analysis and utilization of clinical data. But domestic researches in this area are still at theoretic level. Most of papers published are reviews, and the reports of application cases are inadequate. This paper have mapped an inpatient procedure to a transaction (market basket), items of diagnosis and treat to itemsets(basket data); built up the association analysis model so as to extract association rules form clinical data; applied data mining model to clinical data of coronary heart disease(CHD) patients in a certain general hospital; gained the association rules about diagnosis and operation of CHD; evaluated the validity and novelty of rules; discussed and concluded some problems come across during the course of data mining.