Analysis of medical and healthcare data based on positive and negative association rules

Abstract

Researchers pointed out that the integrated health care system has gradually become an important part of modern health care information system. This is similar to the enterprise information system, mainly for the needs of the medical industry sector. The medical and healthcare data contains a lot of rich information, yet it is generally considered as being poor. However, there is no effective way to analyze and model, and you can draw hidden relationships and patterns from these information. Data mining technology in the commercial, scientific and other fields applied to many practical applications. Thus, the importance of data in the medical and healthcare data (human genetic code, medical records and prescriptions, hospital management information, pathogenic factors, etc.) more and more attentions which makes the association rules mining in medical applications more and more widely. To date, association rules mining has been widely used in people's social life, including transportation, health care, education, etc. most applications focus on positive association rules, neglecting negative effects among various diagnosis processes (e.g. a specific symptom wo't occur when some symptoms occur). Hence, negative association rules are sometimes much more informative than the positive ones. Due to that, this paper aims to analyze medical and healthcare data comprehensively from both positive and negative association rules. The analysis was performed on the medical and healthcare data collected from a Person's Hospital. Using medical profiles such as age, sex, medication records and disease to predict the possibility of a patient or some disease. It can be significant information, such as patterns, and the relationship between disease-related medical and physiological indicators. These association rules build bridges among different diseases and medicines and provide significant information for doctors and social organizations. The association rules we found can provide important reference value for medical and healthcare research and development, like potential complications, preventive medicine, disease diagnosis and disease prevention.