Data Mining Applications in Master Health Checkup: A Statistical Exploration

Data mining has been used exhaustively and widely by many organizations. In healthcare, data mining is becoming more and more popular, if not increasingly essential. Data mining applications can significantly benefit all parties involved in the healthcare industry. The huge amounts of data generated by healthcare transactions are too complex and voluminous to be processed and analyzed by traditional methods. Data mining provides the methodology and technology to transform these mounds of data into useful information for decision making. This study explores data mining applications in healthcare. In particular, this area was chosen as a model to study Master Health Checkup (MHC). The data were collected from secondary source containing 295 patients in St. Johns Hospital, Bangalore. The case sheet deals with socio demographic characteristic, Blood Pressure, Fat, Liver and diabetic related parameters. The salient feature of this study is the application of Factor Analysis, K-means clustering and Multivariate Discriminant Analysis (MDA) as data mining tools to develop the hidden structure present in the data. The scores from extracted factors are used to find initial groups by K-means clustering algorithm. A few outlier health care profiles, which could not be classified to any of the larger groups, are discarded as some of the parameters possessed higher values. Finally, DA is applied and the groups are identified as MHC patients belonging to O-Class (Obesity), N-Class (Normal) and UW-Class (Under Weight) in that order. The results of the study indicate that DA classification maps can be a feasible tool for the health care analysis of large amounts of master health checkup data.