Surgical skill evaluation is a field that attempts to improve patient outcomes by accurately assessing surgeon proficiency. An important application of the information gathered from skill evaluation is providing feedback to the surgeon on their performance. The most commonly utilized methods for judging skill all depend on some type of human intervention. Expert panels are considered the gold standard for skill evaluation, but are cost prohibitive and often take weeks or months to deliver scores. The Fundamentals of Laparoscopic Surgery (FLS) is a widely adopted surgical training regime. Its scoring method is based on task time and number of task-specific errors, which currently requires a human proctor to calculate. This scoring method requires prior information on the distribution of scores among skill levels, which creates a problem any time a new training module or technique is introduced. These scores are not normally provided while training for the FLS skills test, and [1] has shown that FLS scoring does not lend any additional information over sorting skill levels based on task time. Crowd sourced methods such as those in [2] have also been used to provide feedback and have shown concordance with patient outcomes, however it still takes a few hours to generate scores after a training session.
It is desired to find an assessment method that can deliver a score immediately following a training module (or even in real time) and depends neither on human intervention nor on task-specific probability distributions. It is hypothesized that isogony-based surgical tool motion analysis discerns surgical skill level independent of task time.
