Making Ventilators
See What They
Couldn't Before
Orbit AI develops machine-learning technology that gives mechanical ventilators the ability to understand patient respiratory effort in real time — enabling safer, more intelligent critical care.
AI Research Meets
Critical Care
Orbit AI Ventilation is a Canadian medical technology company combining deep clinical expertise in respiratory medicine with advanced machine learning. We make mechanical ventilation safer and more intelligent.
We provide consulting, research, and development services to hospitals, academic institutions, and medical device companies — contributing to Ontario's medical technology sector through clinically applicable intelligent ventilation solutions.
We bridge the gap between clinical insight and technological innovation. Our team combines physician-scientist experience in intensive care with expertise in machine-learning model development — translating research from the bench to the bedside.
Ontario, Canada
Respiratory intelligence
Research to clinical translation
- Hospitals & ICUs
- Academic Research Institutions
- Medical Device Companies
Computational
Ventilation
A machine-learning breakthrough that non-invasively predicts the respiratory muscle pressure waveform in real time — giving ventilators the ability to see what was previously invisible.
The Problem:
Ventilators Are Blind
Traditional ventilators monitor only machine-delivered pressure but cannot quantify the patient's own respiratory effort. This makes it impossible to match ventilator support to the patient's actual needs.
ARDS mortality remains approximately 40%. Uncontrolled spontaneous breathing can cause patient self-inflicted lung injury, while both excessive and insufficient respiratory effort can injure the diaphragm.
The gold standard — esophageal pressure monitoring — requires invasive catheter insertion, specialized training, and expensive consumables. Existing non-invasive alternatives provide only limited, unreliable snapshots.
What Makes CVent Different
Built with a fundamentally different approach to respiratory effort monitoring.
AI-Driven, No Assumptions
Trained on over one million breath waveforms via computational simulation. Unlike polynomial fitting methods, CVent makes no assumptions about Pmus waveform shape.
Real-Time Waveform Output
Provides a continuous Pmus waveform covering both inspiration and expiration — not just a single index value. Full temporal resolution of patient effort.
Multi-Mode Compatible
Works across PSV, PCV, and VCV ventilation modes. One model, all standard ventilation strategies.
Handles Asynchrony
Maintains accuracy during ineffective triggering, double triggering, reverse triggering, auto-triggering, and premature or delayed cycling.
Zero Invasiveness
No catheters, no consumables, no specialized training. Uses only airway pressure and flow signals already available from the ventilator.
Iteratively Upgradable
AI model architecture enables continuous improvement with new training data. The system evolves with clinical knowledge.
Clinical Value
Transpulmonary Driving Pressure
Enables estimation of transpulmonary driving pressure during spontaneous breathing for lung-protective ventilation.
Dual Lung-Diaphragm Protection
Guides both lung protection against overdistension and diaphragm protection against atrophy or injury.
Patient-Ventilator Synchrony
Full Pmus waveform reveals asynchrony types that are otherwise invisible to the clinician.
Weaning Prediction
Pmus and pressure-time product directly reflect respiratory muscle strength and work of breathing.
Validation Evidence
From Research to Reality
CVent technology has been validated and translated into a commercial intelligent ventilator platform through industry partnership. The technology is compatible with standard ventilation modes and delivers real-time Pmus waveform output after approximately one minute of AI learning at the inference stage.
Multiple patent applications have been filed protecting the core innovation.
Competitive Advantage
How CVent compares to existing approaches for monitoring respiratory effort.
| Feature | CVentOrbit AI | End-Inspiratory HoldForeign Brands | Polynomial FittingDomestic Brands |
|---|---|---|---|
| Pmus Index (Peak Effort) | ✓ | ✓ | ✓ |
| Real-Time Continuous Pmus Waveform | ✓ | — | — |
| Fully Non-Invasive | ✓ | — | ✓ |
| Multi-Mode (PSV / PCV / VCV) | ✓ | — | ✓ |
| Respiratory Phase Information | ✓ | — | — |
| Works During Asynchrony | ✓ | — | — |
| No Waveform Shape Assumptions | ✓ | ✓ | — |
| AI Model, Iteratively Upgradable | ✓ | — | — |
Dr. Lu Chen
R&D Lead Scientist
PhD
University of Toronto
MD
Central South University
Specialist Physician
Beijing Tiantan Hospital, Critical Care
Research Fellow
St Michael's Hospital, University of Toronto
Supervised by Dr. Laurent Brochard
Co-Supervised by Dr. Haibo Zhang
Publication Record
22+ articles · 8 reviews · 4 book chaptersFirst-author publications in top-tier critical care journals:
Airway closure in ARDS
AJRCCM 2018 — First Author
Recruitment-to-Inflation Ratio
AJRCCM 2020 — First Author
Respiratory mechanics partition in ARDS
ICM 2022 — First Author
Bedside esophageal pressure monitoring
Critical Care 2017 — First Author
Manuscript Reviewer
Key Scientific Contributions
Airway Closure in ARDS
Identified and characterized airway closure as an underestimated phenomenon in ARDS (AJRCCM 2018)
Recruitment-to-Inflation Ratio
Developed a bedside method for rapid quantification of PEEP benefit-risk balance (AJRCCM 2020)
Respiratory Mechanics Partition
Partitioned respiratory mechanics in ARDS patients across multiple centers (ICM 2022)
CVent Technology
Co-inventor of Computational Ventilation — ML-based non-invasive Pmus prediction
Invited Speaking
- European Society of Intensive Care Medicine (ESICM) — 2016, 2017, 2018
- ESICM Live Forum, Madrid — 2018
- Canadian Critical Care Society / Critical Care Canada Forum — 2016, 2019
- University of Toronto Mechanical Ventilation Symposium — 2016–2022
Research Grants
CAVIARDS Randomized Controlled Trial
Co-applicant — Funded by CIHR and University of Toronto
CAVIARDS-19 Trial
Co-applicant — Funded by University of Toronto
Services
Specialized consulting, research, and development for organizations advancing respiratory care through technology.
Hospitals & ICUs
Clinical consulting and research collaboration for intensive care units seeking to advance respiratory care practices.
- Respiratory mechanics consulting
- Clinical study design and support
- Device evaluation and validation
- Physiological data interpretation
- Esophageal pressure monitoring implementation
- Advanced monitoring integration (EIT, respiratory mechanics)
Academic Research
Collaborative research partnerships for institutions advancing respiratory physiology and critical care science.
- Respiratory physiology interpretation
- Research protocol development
- Collaborative study design
- Data analysis and interpretation
- EIT monitoring integration research
- Publication and grant collaboration
Device Companies
End-to-end R&D services for medical device manufacturers developing next-generation ventilation technology.
- ML model development for ventilators
- Algorithm validation and testing
- Clinical integration consulting
- Bench and in silico testing
- Regulatory compliance support
- Plug-in module development for existing systems
Start a
Conversation
Whether you're exploring AI-driven ventilation solutions, seeking research collaboration, or evaluating our consulting services — we'd like to hear from you.
Unit 4 – 117 Wellington St. E.
Aurora, ON L4G 1H9, Canada
Orbit AI Ventilation Inc.
Founded 2023 — Ontario, Canada