Marcus Owen Tsang 
and Sui Ching Katherine Dai
Medical Student, School of Medicine, Newcastle University, Newcastle Upon Tyne, UK 
Correspondence to: Marcus Owen Tsang, marcuscholar@gmail.com
Additional information
- Ethical approval: N/a
- Consent: N/a
- Funding: No industry funding
- Conflicts of interest: N/a
- Author contribution: Marcus Owen Tsang and Sui Ching Katherine Dai – Conceptualization, Writing – original draft, review and editing
- Guarantor: Sui Ching Katherine Dai
- Provenance and peer-review: Unsolicited and externally peer-reviewed
- Data availability statement: N/a
Keywords: AI-driven personalised feedback, Virtual reality surgical simulation, Automated performance assessment, Adaptive learning algorithms, Undergraduate robotic suturing training
Peer Review
Received: 23 November 2025
Accepted: 23 November 2025
Version accepted: 1
Published: 27 December 2025
VRiMS Inaugural Conference Abstract
Plain Language Summary Infographic
Abstract
Introduction: Artificial Intelligence (AI) and immersive technologies such as Virtual Reality (VR) have the ability to personalise undergraduate surgical education by delivering adaptive feedback, standardised assessment, and resource-efficient training.
Methods: A focused review of 75 studies from a pool of >140 on the PubMed database explored AI-driven feedback, automated assessment, VR integration, and adaptive learning in undergraduate surgical training.
Results: Results across multiple trials demonstrate significant improvements in surgical training using AI-driven personalised approaches. These benefits were seen across the learning pathway – from initial skill acquisition to targeted feedback and final performance outcomes. A randomised pilot study reported high classification accuracies compared with human assessors for suturing (89%) and knot-tying (91%), lending to the credibility of AI-based performance assessments. Further trials involving participants without previous robotic surgery experience demonstrated significantly greater improvements in robotic suturing skills for those who received teaching videos based on AI-personalised feedback compared to those that did not (improvement 0.30 vs –0.02, p = 0.018). A Cochrane review involving a further 8 trials on AI-personalised surgical training showed similarly promising results, reporting reduced operative time by an average of 11.76 minutes (95% CI: 15.23 to 8.30 minutes), with adaptive AI feedback particularly beneficial for underperformers (p = 0.02).
Conclusions: AI-assisted feedback accurately identifies performance levels, whilst AI-personalised curricula and VR simulations show clear short-term improvements in novice surgical skill acquisition. These findings support the need for larger, multi-centre randomised controlled trials to assess long-term benefits. Integrating AI as a faculty-supervised adjunct offers a pragmatic and scalable path to transforming surgical education.
