Effectiveness of an Artificial Intelligent Tutoring System in Simulation Training
Study Details
Study Description
Brief Summary
Brief Summary:
Background:
Although surgical experience and technical skill are associated with better patient outcomes, quantitating surgical ability in the operating room is challenging. In surgical education, large datasets generated by high-fidelity virtual reality simulators can be employed by machine learning algorithms to objectively measure trainee performance and competence on expert benchmarks. This allows repetitive practice of surgical skills in safe and risk-free environments with immediate feedback.
Our group developed and has a patent pending for an intelligent tutoring system called the Virtual Operative Assistant (VOA). Utilizing an Artificial Intelligence (AI) support vector machine algorithm, the VOA assesses data derived from the NeuroVR (CAE Healthcare) simulator platform and provides individualized audiovisual feedback to improve learner performance during simulated brain tumor resections. The effectiveness of intelligent tutoring systems such as the VOA to the human surgical apprenticeship pedagogy remains to be elucidated.
The aim of this study is to compare the effectiveness and educational impact of personalized VOA feedback to expert instruction on medical student's technical skills learning of a virtual reality tumor resection procedure.
Specific Aims: 1) To assess if medical students receiving personalized VOA feedback statistically improve their surgical performance when compared to those having (a) no expert instructor feedback or (b) expert instructor-mediated feedback. 2) To outline if different emotions are elicited by the VOA intelligent tutoring system in medical students while performing this achievement task as compared to human instruction
| Condition or Disease | Intervention/Treatment | Phase |
|---|---|---|
|
N/A |
Detailed Description
Design: A three-arm partially blinded randomized controlled trial of VOA training versus remote-based expert instruction versus control.
Setting: Neurosurgical Simulation and Artificial Intelligence Learning Centre, Montreal Neurological Institute.
Participants: Eligible first- and second-year medical students from across the province of Quebec.
Task: Complete removal of a simulated tumour - distinguishable by colour and haptic properties - with minimal bleeding and damage to surrounding healthy brain using two surgical instruments (Cavitron Ultrasonic Aspirator and Bipolar pincers) of the NeuroVR (CAE Healthcare) surgical simulator.
Intervention: A single 75-minute training session, including six virtual subpial tumour resection attempts (five simple practice scenarios and one complex realistic scenario) with assessment and feedback from either:
-
the VOA intelligent tutoring system (Group 2) or
-
a remote-based expert instructor (Group 3)
Both compared to:
- control group (Group 1) that receives no assessment or performance feedback.
To our knowledge this will be the first study to compare the effectiveness of an AI-powered intelligent tutoring system to expert instruction in the context of medical and surgical virtual reality simulation and assess the emotional response to such instruction. This study aims to begin to identify successful approaches to use this innovative technology in the medical educational curriculum and improve patient outcomes by augmenting safety, efficiency and competency of surgeons and other healthcare providers.
Study Design
Arms and Interventions
| Arm | Intervention/Treatment |
|---|---|
| No Intervention: Control Group Control Group - Baseline Training 25 Participants allocated. Individuals receive introductory information on using the simulator and the scenario. They perform 5 simple subpial tumour resections for practice and have 5 minutes per trial. After each attempt, the student takes a 5-minute break with no assessment or feedback on their performance. On their 6th attempt they have 13 minutes to perform a different realistic scenario. |
|
| Experimental: Experimental Group - Virtual Operative Assistance Training Experimental Group - Virtual Operative Assistance Training 25 participants allocated. Individuals receive the same information, have the same amount of time and perform the same scenarios as the control group. In the 5-minutes between attempts, participant receive the Virtual Operative Assistance Training assessment of their performance and audiovisual feedback. |
Behavioral: Virtual Operative Assistant Training
Individuals receive the same basic information, have the same amount of time and perform the same scenarios as the control group. In the 5-minutes between attempts, participant receive the Virtual Operative Assistant assessment of their performance and audiovisual feedback.
|
| Experimental: Experimental Group - remote-based expert Instructor Training 25 participants allocated. Individuals receive the same information, have the same amount of time and perform the same scenarios as the control group. Meanwhile, a trained instructor observes the participant's on-screen performance, that is live-streamed, remotely. Instructors are senior neurosurgery residents with extensive experience in performing and assessing this scenario. During the 5-minute feedback session, they chat with the student, discussing the performance and help in setting goals for the next trial. |
Behavioral: Remote-Based Expert Instructor Training
Individuals receive the same information, have the same amount of time and perform the same scenarios as the control group. Meanwhile, a trained instructor observes the participant's on-screen performance, that is live-streamed, remotely. Instructors are senior neurosurgery residents with extensive experience in performing and assessing this scenario. During the 5-minute feedback session, they chat with the student, discussing the performance and help in setting goals for the next trial.
|
Outcome Measures
Primary Outcome Measures
- Change in procedural performance . [Day of Study]
Performance in each practice attempt is measured utilizing raw data from the simulator that is used for assessment by previously established AI algorithms on validated metrics.
- Change in learning [Day of Study]
Performance on the complex realistic scenario is evaluated by expert instructors using the Objective Structured Assessments of technical Skills (OSATS) Visual Rating Scale (weighted at 50%) and the AI assessment algorithms (weighted at 50%) creating a composite performance score.
Secondary Outcome Measures
- Difference in the strength of emotions elicited [Day of Study]
Measured using Duffy's Medical Emotions Scale (MES), before, during and after the intervention.
- Difference in cognitive load [Day of Study]
Measured using Leppink's Cognitive Load Index (CLI) after the intervention.
Eligibility Criteria
Criteria
Inclusion Criteria:• First- and second-year medical students from any Canadian institution who do not meet the exclusion criteria.
-
Exclusion Criteria: • Participation in any of our group's previous trials involving the NeuroVR (CAE Healthcare) simulator.
-
Contacts and Locations
Locations
| Site | City | State | Country | Postal Code | |
|---|---|---|---|---|---|
| 1 | Neurosurgical Simulation and Artificial Intelligence Learning Centre | Montreal | Quebec | Canada | H3A 2B4 |
Sponsors and Collaborators
- McGill University
Investigators
- Principal Investigator: Rolando Del Maestro, MD, McGill
Study Documents (Full-Text)
None provided.More Information
Publications
- Birkmeyer JD, Finks JF, O'Reilly A, Oerline M, Carlin AM, Nunn AR, Dimick J, Banerjee M, Birkmeyer NJ; Michigan Bariatric Surgery Collaborative. Surgical skill and complication rates after bariatric surgery. N Engl J Med. 2013 Oct 10;369(15):1434-42. doi: 10.1056/NEJMsa1300625.
- Birkmeyer JD, Stukel TA, Siewers AE, Goodney PP, Wennberg DE, Lucas FL. Surgeon volume and operative mortality in the United States. N Engl J Med. 2003 Nov 27;349(22):2117-27.
- Duffy, M.C., et al., Emotions in medical education: Examining the validity of the Medical Emotion Scale (MES) across authentic medical learning environments. Learning and Instruction, 2020. 70: p. 101150.
- Leppink J, Paas F, Van der Vleuten CP, Van Gog T, Van Merriënboer JJ. Development of an instrument for measuring different types of cognitive load. Behav Res Methods. 2013 Dec;45(4):1058-72. doi: 10.3758/s13428-013-0334-1.
- Mirchi N, Bissonnette V, Yilmaz R, Ledwos N, Winkler-Schwartz A, Del Maestro RF. The Virtual Operative Assistant: An explainable artificial intelligence tool for simulation-based training in surgery and medicine. PLoS One. 2020 Feb 27;15(2):e0229596. doi: 10.1371/journal.pone.0229596. eCollection 2020.
- Stulberg JJ, Huang R, Kreutzer L, Ban K, Champagne BJ, Steele SR, Johnson JK, Holl JL, Greenberg CC, Bilimoria KY. Association Between Surgeon Technical Skills and Patient Outcomes. JAMA Surg. 2020 Oct 1;155(10):960-968. doi: 10.1001/jamasurg.2020.3007. Erratum in: JAMA Surg. 2020 Oct 1;155(10):1002. Erratum in: JAMA Surg. 2021 Jul 1;156(7):694.
- Winkler-Schwartz A, Marwa I, Bajunaid K, Mullah M, Alotaibi FE, Bugdadi A, Sawaya R, Sabbagh AJ, Del Maestro R. A Comparison of Visual Rating Scales and Simulated Virtual Reality Metrics in Neurosurgical Training: A Generalizability Theory Study. World Neurosurg. 2019 Jul;127:e230-e235. doi: 10.1016/j.wneu.2019.03.059. Epub 2019 Mar 15.
- Winkler-Schwartz A, Yilmaz R, Mirchi N, Bissonnette V, Ledwos N, Siyar S, Azarnoush H, Karlik B, Del Maestro R. Machine Learning Identification of Surgical and Operative Factors Associated With Surgical Expertise in Virtual Reality Simulation. JAMA Netw Open. 2019 Aug 2;2(8):e198363. doi: 10.1001/jamanetworkopen.2019.8363.
- 2010-270, NEU-09-042