We use cookies to understand how you use our site and to improve your experience. This includes personalizing content and advertising. To learn more, click here. By continuing to use our site, you accept our use of cookies. Cookie Policy.

Features Partner Sites Information LinkXpress
Sign In
Advertise with Us
Greiner Bio-One

Download Mobile App




Virtual Reality Simulators Help Determine Neurosurgeon Expertise

By HospiMedica International staff writers
Posted on 14 Aug 2019
Print article
Image: A new study claims VR simulators can help categorize neurosurgeon expertise (Photo courtesy of Helmut Bernhard/ NEURO).
Image: A new study claims VR simulators can help categorize neurosurgeon expertise (Photo courtesy of Helmut Bernhard/ NEURO).
Virtual reality (VR) simulators may soon be capable of classifying surgical expertise with high precision, claims a new study.

Researchers at McGill University (Montreal, Canada) and Amirkabir University of Technology (Tehran, Iran) conducted a study that included 50 participants in order to identify surgical and operative factors--as selected by a machine learning algorithm--that could be used to quantify psychomotor skills and generate data sets that could be used classify levels of expertise in a VR surgical procedure. For the study, the participants conducted tumor resections using the NeuroVR, a VR simulator that records all instrument movements in 20 millisecond intervals.

Study participants were recruited from four stages of neurosurgical training. They were classified as expert (neurosurgery staff), seniors (neurosurgical fellows and senior residents), juniors (neurosurgical junior residents), and medical students, all of whom participated in 250 simulated resections. Through an iterative process, performance metrics associated with instrument movement and force, resection of tissues, and bleeding generated from the raw simulator data output were selected to most accurately determine group membership.

The results showed that a K-nearest neighbor algorithm had an accuracy of 90% (45 of 50), a naive Bayes algorithm had an accuracy of 84%, a discriminant analysis algorithm had an accuracy of 78%, and a support vector machine algorithm had an accuracy of 76%. The K-nearest neighbor algorithm used six performance metrics to classify participants, the naive Bayes algorithm used 9 performance metrics, the discriminant analysis algorithm used 8 performance metrics, and the support vector machine algorithm used 8 performance metrics. The study was published on August 2, 2019, in the Journal of the American Medical Association (JAMA).

“Physician educators are facing increased time pressure to balance their commitment to both patients and learners,” said senior author Rolando Del Maestro, PhD, of the McGill Neurosurgical Simulation and Artificial Intelligence Learning Centre (NEURO). “Our study proves that we can design systems that deliver on-demand surgical assessments at the convenience of the learner and with less input from instructors. It may also lead to better patient safety by reducing the chance for human error both while assessing surgeons and in the operating room.”

Current training for surgeons is largely confined to classroom lessons and viewing cadaver-based teaching, with limited hands-on time actually spent on cadavers by students themselves.

Related Links:
McGill University
Amirkabir University of Technology


Print article

Channels

Copyright © 2000-2019 Globetech Media. All rights reserved.