Optimal Sub-Sequence Matching for the Automatic Prediction of Surgical Tasks

Abstract

International audienceSurgery is one of the riskiest and most important medical acts that is performed today. The desires to improve patient outcomes, surgeon training, and also to reduce the costs of surgery, have motivated surgeons to equip their Operating Rooms with sensors that describe the surgical intervention. The richness and complexity of the data that is collected calls for new machine learning methods to support pre-, peri- and post-surgery (before, during and after). This paper introduces a new method for the prediction of the next task that the surgeon is going to perform during the surgery (peri). Our method bases its prediction on the optimal matching of the current surgery to a set of pre-recorded surgeries. We assess our method on a set of neurosurgeries (lumbar disc herniation removal) and show that our method outperforms the state of the art by providing a prediction (of the next task that is going to be performed by the surgeon) more than 85% of the time with a 95% accurac

Surgical motion analysis using discriminative interpretable patterns

“…In recent years, analysis of surgical motion has received a growing interest following the development of devices enabling automated capture of surgeon motions such as tracking, robotic and training systems. Surgical training programs now often include surgical simulators which are equipped with sensors for automatic surgical motions recording [1,2,3]. The ability to collect surgical motion data brings unprecedented opportunities for automated objective analysis and assessment of surgical trainees progression.…”

Offline identification of surgical deviations in laparoscopic rectopexy

“…The first is to develop an on-line multi-dimensional alignment method. Recently, Forestier et al [28] proposed a method designed to create an online one-dimensional alignment. The second aspect pertains to creating a reliable and automatic online activity recognition method [29,30,31].…”

Finding discriminative and interpretable patterns in sequences of surgical activities

“…In this case, a dictionary of good and bad patterns could be built. Finally, this method could also be used as an addition to surgical activities prediction systems [24,25,26,27] by providing frequencies of most frequent subsequences. Our system could also be used to identify the core set of subsequences activities that are performed by all the surgeons regardless of their countries or skill levels.…”