How AI, VR will train the surgeons of tomorrow
AI can play a key role in the development of state-of-the-art training techniques for surgery, according to new findings published in PLOS One. The study’s authors used virtual reality (VR) technology and advanced machine learning techniques to develop both a simulation-based training system for surgeons and an automated feedback system called the Virtual Operative Assistant.
One consistent criticism of AI algorithms is that it can be unclear how, exactly, they reach their conclusions, creating a “black box” effect that leads to confusion and skepticism.
“This is problematic in the context of education because transparency and trust are vital components of ensuring a successful connection between teacher and learner,” wrote lead author Nykan Mirchi, McGill University in Quebec, Canada, and colleagues. “Transparency is also important for developing and implementing appropriate grading schemes and feedback mechanisms. Without such mechanisms, students report negative emotions, such as frustration and discomfort, when using technology for higher (post-secondary) education.”
Developing an AI-powered feedback system, the authors added, could help address such issues by putting such an emphasis on transparency.
To develop their training system, the researchers worked with 28 “skilled participants,” including 14 staff neurosurgeons, and 22 novice participants. Each person was asked to perform a brain tumor resection while wearing a VR headset, and then the team generated a list of performance metrics for training their algorithm.
It was then time to develop the Virtual Operative Assistant, which could provide feedback as soon as a participant had performed the surgical task at hand. The system was designed so that users perform the task at hand and then are automatically assessed as either ‘skilled’ or ‘novice’ via machine learning.
Overall, the feedback system had an accuracy of 92%, specificity of 82% and sensitivity of 100%.
“The Virtual Operative Assistant offers many advantages over non-AI methods of teaching in medical simulation,” the authors explained. “It allows educators to identify individual components of psychomotor expertise in tasks too multifaceted for instructors to adequately appreciate without this technology. The complexity of surgery makes it an ideal educational paradigm to apply these systems.”
Mirchi et al. concluded that their findings establish “a basis for the potential role of integrating AI and virtual reality simulation into surgical educational teaching.” However, they added, additional research is still required that that takes additional AI techniques into account and can provide more specific trainee assessments.
Michael has more than 18 years of experience as a professional writer and editor. He has written at length about cardiology, radiology, artificial intelligence and other key healthcare topics.