Two rigorous phases.
One validated platform.
A prospective validation study conducted at the Simulation Laboratory, Queen's University. This is how we build a training tool the academic and clinical community can trust.
The Research Timeline
Enrollment begins August 2026. Two sequential phases, each building on the last.
Expert–Novice Discrimination Study
Before we can train anyone with AI feedback, we need to know what expert performance actually looks like and which XR-captured metrics meaningfully separate experts from novices. Phase 1 builds that foundation.
Expert vs. Novice Discrimination
Compare experienced clinical practitioners to health sciences students with no prior IV experience. Identify which XR-captured performance metrics reliably tell the difference.
Validate Performance Indices
From 13 candidate metrics captured by the Meta Quest 3 — insertion angle, needle trajectory, movement smoothness, hazard events, workflow sequence — identify which ones meet rigorous discrimination thresholds.
Build the Expertise Index (MREI)
Construct a composite Mixed Reality Expertise Index, a single validated score that predicts expert-level performance, built from the most discriminating metrics.
Calibrate AI Feedback Thresholds
Use expert performance distributions to set the benchmarks the AI will use in Phase 2. When feedback is calibrated against real experts, it means something.
The output of Phase 1 is a validated list of performance indices, a composite MREI score, and the expert-calibrated AI feedback thresholds that power Phase 2.
Randomized Controlled Trial
Does AI-augmented XR training actually produce better procedural skills than training without feedback, or better than video? Phase 2 answers that with a three-arm randomized controlled trial.
XR + AI Feedback
Learners practice in the XR environment with the full AI system active. After each attempt: real-time safety alerts for critical errors, and a detailed post-attempt scorecard comparing performance against Phase 1 expert benchmarks.
XR Only (No Feedback)
Learners practice in the same XR environment, without any AI feedback or scorecard. This arm isolates the effect of the XR environment itself versus the added value of AI coaching.
Video Instruction
Learners study a standardized expert-performed IV venipuncture instructional video. This is the current standard of scalable procedural training and serves as the comparison baseline.
What Phase 2 Will Measure
"This isn't a commercial product first — it's a validated educational tool."
Results from both phases will be submitted for publication in peer-reviewed medical education journals.
Why XR + AI Feedback Works
Our approach is grounded in peer-reviewed evidence. Here's what the literature says.
HoloLens 2 for Arteriotomy Training
HoloLens MR guidance produced greater and more consistent skill progression than video. Proficiency gain: 10.1 vs. 6.89 (p=0.0076).
Immersive XR guidance with interactive feedback improves procedural outcomes beyond video.
AR-Guided External Ventricular Drain Placement
AR-guided novices achieved accuracy comparable to trained freehand users (12.2mm vs. 13.5mm), with a significantly reduced learning curve.
Real-time spatial guidance enables novices to perform at trained levels without constant supervision.
Mixed Reality for Trauma Management
MR simulation outperformed traditional didactic teaching. Higher clinical knowledge scores and positive supervisor appraisal sustained.
XR-based learning transfers to real clinical performance — the gold standard for training modality evaluation.
Immersive VR for Chest Tube Insertion
Knowledge improved 46.7% → 86.7% post-training (p<0.001). System Usability Scale 82.5/100. Strong correlation between usability and technical skills (r=0.51).
Immersive VR is both effective and highly usable — usability and outcomes move together.
HMD Immersive Laparoscopy
80% user preference for immersive HMD. Immersion drove engagement and intent for ongoing use.
HMD-delivered training increases learner engagement and likelihood of continued skill development.
HTC Vive for ENT Procedural Training
VR group OSCE scores significantly higher (26.9 vs. 21.5, p=0.005). Confidence higher (p=0.008). VR rated 9.6/10 for effectiveness.
Commercial HMD platforms deliver high-fidelity training with minimal cost and excellent experience.
Orthognathic Surgery VR Training
Fidelity rating 4.35/5 across virtual environment, instruments, anatomy, and procedures. Both surgeons and novice students rated system highly.
Immersive VR teaches complex surgical procedures across experience levels — novice to expert.
Synthesis
Across these studies, one pattern emerges: XR with intelligent feedback produces faster learning, better skill transfer, and higher engagement than traditional methods. OPERIO is built directly on this evidence base.