Surgical Intent

From Distributed Information and Intelligence Analysis Group

Numerous complex problems are faced in critical care environment, most often in the operating room and trauma environment. Such environments require a team to work together under severe time constraints. A typical scenario is the handoff from one team to another in a complex medical procedure. Due to incomplete information or misunderstanding/miscommunication between doctors, a wrong decision might be made which could result in loss of life or limb. In this project we propose to develop a formal approach that can enhance decision making and problem solving in trauma environments. The goal is to come up with a model of the behaviors and intentions of each member in a surgical team, in order to prevent operating room mishaps ranging from simple miscommunications to incorrect amputations. This involves gathering data from surgeons about various OR scenarios, putting it into Bayesian Knowledge Bases (BKB, Santos 99), using BKB to make an inference about intentions and behaviors over time, and generating an alert if someone deviates from the intended course.