Management of a difficult airway
Translating clinical workflows into an intuitive simulation.
Company
Lumeto
Role
UX Designer
UX Researcher
Interaction Designer
Duration
6 months
Team
1 PO
1 Scrum master
1 Designer (me)
2 Engineers
Team
1 Technical Artist
2 Content Artists
Clinical partners
The Problem
❌
Users struggled with key VR procedures, initially attributed to lack of VR experience.
❌
Lack of tactile feedback made error detection difficult.
❌
Clinical partners demanded realism to prevent negative learning habits.
Learning efficacy
The Solution
✅
Introduced automatic tool alignment to reduce errors.
✅
Enhanced feedback to aid user understanding.
✅
Balanced usability and realism, leading to improved learner success and satisfaction.
Learning efficacy
Impact
⚠️
-80%
User Error
⏱️
-50%
Time-to-task completion
🎓
+26%
Procedural knowledge
🎉
98%
Learning efficacy
Context
Challenge: Users struggled to perform key procedures in our VR-based training scenario, initially thought to be due to unfamiliarity with VR.
Partnering with CHEST (American College of Chest Physicians), we developed a pulmonary critical care scenario for residents. The scenario involved managing a patient with declining oxygen levels, requiring timely intubation to save their life.
This marked our first clinical simulation. Our team collaborated closely with experienced doctors from NYU and the Mayo Clinic to meticulously recreate patient physiology, vital fluctuations, drug interactions, and medical tools for an authentic learning experience.
Despite this effort, learners encountered significant challenges performing procedures, which led us to explore deeper causes beyond VR unfamiliarity.
Discovery
Testing
During Alpha 0.1 testing, users exhibited high error rates, struggled with task completion, and faced a steep learning curve involving VR navigation and tool manipulation.
To diagnose the problem, I coordinated usability tests with clinical partners across the US, collecting data from 39 users over 30 hours of testing at three institutions.
Analysis
Analysis revealed that the main obstacle was the interaction design of two critical tools: the Video Laryngoscope and the Endotracheal Tube (ETT). Learners struggled with delicate maneuvers required for intubation, particularly in opening the airway and correctly inserting the tube.
Key Insight: The primary issue wasn’t VR novelty, but flawed tool interaction design.
Drag to explore – N+1 User Journey Map in FigJam
How it works
Critical Requirement: Accurate tool manipulation under stress is essential for effective training.
In real-life scenarios, clinicians intubate patients by manually opening the airway with a hyperangulated laryngoscopic blade and inserting a tube to deliver oxygen. This delicate process requires precise handling to prevent serious harm.
Our VR design required learners to use the blade and guide the tube based on a monitor display. Incorrect blade angles or tube insertion led to errors, such as pumping oxygen into the stomach instead of the lungs.
Timely, accurate execution was crucial, but our design hindered learners, making successful intubation difficult.
Insights
Findings: Lack of tactile feedback in VR led to poor error recovery.
Without the physical sensation of real tools, learners couldn’t detect subtle errors. Normally, tactile cues help clinicians adjust their technique instinctively. In VR, using bulky controllers and gyroscopic sensors, learners felt disconnected, leading to slow or failed error recovery.
Learners often misinterpreted failures as software bugs, unable to discern what they were doing wrong. The lack of feedback resulted in frustration and significantly hindered learning outcomes.
🔎
Low
Error recognition
⏰
Slow
Error recovery
❌
Insufficient
Feedback
Exploration
Sharing our findings with clinical stakeholders
While our team proposed solutions such as X-ray views, success popups, and technique hints, stakeholders rejected them, fearing that unrealistic elements would create "training scars"—habits that wouldn’t translate to real-world practice.
Challenge and Constraints
Challenge: Clinical stakeholders demanded high realism to avoid negative learning outcomes.
Moreover, we faced time constraints and couldn’t overhaul the engine. Delivering a pilot-ready platform quickly was crucial to maintaining our partnership.
🎯
Acuracy
was paramount
🧑💻
Minimal
Development effort
⚡️
Quick
turnaround time
Product Pillars
Goal: Develop a realistic, intuitive simulation within VR’s technical limits.
Our task was to design a solution that balanced realism and usability while fostering an environment where learners could practice without fear of failure.
We outlined key principles:
💬
Encourage teamwork and communication.
🤔
Enable learners to connect theoretical knowledge with practical skills.
✨
Ensure a smooth transition for users familiar with physical simulations.
📈
Promote productive failure by providing psychological safety.
Final Design
Proposed Solution
Solution: Streamline interactions and enhance feedback.
We redefined our design pillars by removing interactions reliant on physical senses learners lacked. Instead, we simplified procedures and increased feedback layers to guide users.
Updated Interaction
Automatic tool alignment: The VL snapped into place when near the airway, ensuring correct positioning.
Updated Controls
Reduced interactions: We minimized the steps needed for tool manipulation, focusing on decision-making and communication.
Enhanced feedback: Clear visual and auditory cues were added to help learners understand their actions' outcomes.
Mocap studio
New Version Shipped
These changes aligned with clinical partners’ expectations, leading to a successful pilot release.
Outcome: Simplified interactions led to higher learner success and satisfaction.
Users achieved significantly better results with the updated design. Reduced friction allowed deeper immersion, resulting in improved post-test scores and higher knowledge retention.
⚠️
-80%
User Error
⏱️
-50%
Time-to-task completion
🎓
+26%
Procedural knowledge
✅
98%
Learning efficacy
This success allowed us to ship a stable build for CHEST’s pilot study. Shortly after, we launched version 1.0 and showcased the product at IMSH, the largest medical simulation conference.
Our efforts led to a partnership with eCampus Ontario to develop 13 more scenarios for undergraduate and early-career nurses, accessible across all colleges and universities in Ontario.
Testimonials
"Johnny’s can-do attitude and quick problem-solving make him indispensable. From untangling UX inconsistencies across platforms to handling engineering challenges with flexibility and composure, he consistently delivers polished, thoughtful solutions. Empathetic and collaborative, he fosters smooth decision-making and energizes those around him."
Belinda Darcy
Director of Design
Johnny's aptitude for design stands out because of its customer centricity. He was very quick to understand a complex domain such as healthcare simulation that is both vast and varied. He listened intently to users and came up with very intuitive and novel ways to solve for training in a relatively new medium such as VR. Oftentimes he went beyond his role to actively take part in customer engagements and be a champion for the product. I would highly recommend Johnny to any organization that is looking for a multidimensional design specialist. Best wishes!
