Validating Vergo: Why Independent Testing Mattered—And What We Learned

A closer look at our product validation process and the confidence it built

When we set out to build Vergo, we didn’t just want to create another workplace safety tool—we wanted to develop a system that organizations could trust with real decision-making. That meant putting our platform to the test—not internally, but through independent, university-led validation using research-grade methods.

Too often, AI-based products make bold claims without external evidence to back them up. At Vergo, we believe in earning trust through transparency, which is why we partnered with the Faculty of Health at Dalhousie University to run a validation study using optical motion capture—the gold standard for biomechanical analysis.

Here’s why that validation mattered to us, how it was conducted, and what the results told us.

Why We Prioritized Independent Validation

Ergonomics is a science. And when you’re dealing with workplace safety and injury prevention, accuracy matters. We wanted to know:

• How closely does Vergo’s AI posture tracking align with the most advanced human motion capture systems?

• Where does our system perform strongest—and what can be improved?
  

This wasn’t just a checkbox exercise. It was a way to pressure-test our core technology and build confidence among future users: safety managers, clinical researchers, physiotherapists, and frontline workers who rely on accurate feedback.

How the Study Was Designed

Dalhousie University’s validation study, funded through the National Research Council’s IRAP program, involved 10 participants performing three common workplace lifting tasks:

• A stoop lift

• A squat lift

• A lift-and-twist
  

Each participant performed two trials of each task using a 10 lb milk crate. Their movements were captured using both:

• Optitrack motion capture cameras (recording at 120Hz), and

• Standard smartphone cameras (recording at 30Hz), which powered Vergo’s analysis.
  

The researchers compared key joint and segment angles—including torso, shoulder, and elbow—across both systems, using Bland-Altman analysis and curve comparisons to evaluate how closely Vergo matched the motion capture baseline.

The Results: Strong Torso Tracking and Positive Correlation in Movement Curves

The independent review highlighted several strengths in Vergo’s performance:

• Torso tracking showed strong alignment with Optitrack across both squat and stoop lifts. This confirms that Vergo can reliably detect the degree of spinal flexion—a major risk factor in ergonomic injuries.
  

• Movement curves for all joint angles showed similar patterns, meaning Vergo was successfully capturing the shape and timing of posture changes, even during complex movements like the lift-and-twist.
  

• Visual comparison graphs demonstrated consistency between Vergo and motion capture, reinforcing that our AI model tracks real-world body mechanics in a way that maps to gold-standard systems.
  

In short: The overall validation showed that Vergo’s system captures ergonomic movement accurately enough for meaningful workplace use.

“For each segment/joint, Vergo appears to track the motion similar to Optitrack, with the strongest performance being torso angle outputs.”— Dr. Ryan Frayne, School of Health and Human Performance, Dalhousie University

Why This Matters for Our Customers

This validation gives Vergo users something rare in the AI safety space: independent, peer-reviewed assurance that the technology works as intended. It means the posture data you see—especially for spinal flexion—is rooted in biomechanical rigour, not just computer vision hype.

For safety teams, this means more confidence in your assessments. For researchers and clinicians, it offers a scalable alternative to lab-based analysis. And for workers, it means the feedback they’re getting is grounded in real-world biomechanics—not assumptions.