Analysis of surgical event logs in a hospital by using heuristics miner technique

The main objective of the research was to analyze and investigate the behavior of the patients and surgeons based on an event log previously collected from a private hospital in Bangkok, Thailand. We applied Heuristics Mining algorithm supported by ProM (as a process mining tool) to discover the extent of the significance metric (i.e., with respect to the relative importance of behaviors) as well as the correlation metric (i.e., with respect to how closely the behaviors are related with each other) amongst of the process instances within the MXML-formatted event log. The results showed that the activity named “Implantation of diaphragmatic pacemaker” was the most significant activity within the event log with a total of 2,903 cased frequency of repetition as a significance metric of 0.999 (i.e., the highest) and a correlation metric of 1358 (i.e., very firm and robust). The findings of the study can be used by hospitals administrators, managers, surgeons and staff in order to benchmark the patients' treatment processes in a more efficient, effective and timely manner.