Development of an automated decision system for selection of preoperative investigations based on cartesian product and gaussian distribution

To evaluate patient risk(s) before anesthesia and surgery, physicians often order a battery of routine preoperative investigations for all surgical patients. The utility and cost-effectiveness of such practice have been questioned because of abnormal results. In this study, the Cartesian product theory with multiple sets and elements under rule based on a Gaussian distribution aims for functionality of a prototype decision support system for selecting individualized and clinically relevant investigations. The complex medial databases in relevant to patient's risks are categorized into multiple sets with nary relation denoted by Cartesian product. Preliminary results show that the system suggests fewer preoperative investigations in comparison with the issued guidelines, hence lessening healthcare expenditures. Furthermore, the graph of biochemical measurement values could represent the trend of each investigation for homogeneity group of patients.