Research & Validation

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.

August 2026
Enrollment Start
Q3 2026
Phase 1 Results
Expert–Novice Discrimination
Q2 2027
Phase 2 Results
Randomized Controlled Trial
1
PHASE 1 · Q3 2026

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.

Two-Phase Research Protocol Overview

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.

2
PHASE 2 · Q2 2027

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.

AI System Architecture — Data Input, Processing Core, and Feedback Loop
Arm A

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.

Arm B

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.

Arm C

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

Immediate Skill
Blinded OSATS scores right after training — does AI-guided practice produce better technique than video?
Retention
Participants are tracked over time to measure whether skills stick and whether differences emerge between training arms across multiple timepoints.
Cognitive Load
NASA Task Load Index across all three arms: does AI feedback help or overload learners?
Procedural Knowledge
Pre/post written knowledge test: does XR training transfer to conceptual understanding?

"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.

Published Evidence

Why XR + AI Feedback Works

Our approach is grounded in peer-reviewed evidence. Here's what the literature says.

HoloLens 2 for Arteriotomy Training

Key Finding

HoloLens MR guidance produced greater and more consistent skill progression than video. Proficiency gain: 10.1 vs. 6.89 (p=0.0076).

Implication

Immersive XR guidance with interactive feedback improves procedural outcomes beyond video.

AR-Guided External Ventricular Drain Placement

Key Finding

AR-guided novices achieved accuracy comparable to trained freehand users (12.2mm vs. 13.5mm), with a significantly reduced learning curve.

Implication

Real-time spatial guidance enables novices to perform at trained levels without constant supervision.

Mixed Reality for Trauma Management

Key Finding

MR simulation outperformed traditional didactic teaching. Higher clinical knowledge scores and positive supervisor appraisal sustained.

Implication

XR-based learning transfers to real clinical performance — the gold standard for training modality evaluation.

Immersive VR for Chest Tube Insertion

Key Finding

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).

Implication

Immersive VR is both effective and highly usable — usability and outcomes move together.

HMD Immersive Laparoscopy

Key Finding

80% user preference for immersive HMD. Immersion drove engagement and intent for ongoing use.

Implication

HMD-delivered training increases learner engagement and likelihood of continued skill development.

HTC Vive for ENT Procedural Training

Key Finding

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.

Implication

Commercial HMD platforms deliver high-fidelity training with minimal cost and excellent experience.

Orthognathic Surgery VR Training

Key Finding

Fidelity rating 4.35/5 across virtual environment, instruments, anatomy, and procedures. Both surgeons and novice students rated system highly.

Implication

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.

FAQ

Frequently Asked Questions

Have a question not answered here?

Contact Us
OPERIO

XR-based, AI-powered feedback platform for procedural medical skills training. Developed at the Chung Lab, Queen's University.

Get In Touch

Chung Lab, Faculty of Health Sciences
Queen's University
Kingston, Ontario, Canada

Schedule a Conversation

© 2026 Operio. All rights reserved.