Respiratory failure & ventilation
Enhancing mechanical ventilation management with artificial intelligence (AI): Computer vision for automated detection of ventilatory modes, parameters and asynchrony
Authors' conclusion
NexoVent accurately detects ventilatory data and multiple forms of PVA using non-invasive, image-based computer vision.
Study context
- Population
- ICU of a tertiary care hospital; Servo-i and Servo-s ventilators in PCV mode, images acquired with standard mobile devicesrespiratory cycles analysed
- Participants
- 621 participants
- Comparison
- NexoVent AI computer vision vs expert consensus
- Outcome
- Accuracy of NexoVent in detecting ventilator mode, alphanumeric ventilator parameters and patient-ventilator asynchrony (PVA)
Methodological detailsobservational, single-centre
- Observational
- Single-centre
Performance
As reported in the abstract
Detection accuracy vs expert consensus
| Test or model | Accuracy, ventilator parameters (overall) | Accuracy, ventilator modes | Asynchrony detection, delayed cycle (lowest) | Asynchrony detection, ineffective effort (highest) |
|---|---|---|---|---|
| NexoVent | 95.4 % | 94.0 % | 81.6 % | 97.8 % |
NexoVent
Accuracy, ventilator parameters (overall)95.4 %
Accuracy, ventilator modes94.0 %
Asynchrony detection, delayed cycle (lowest)81.6 %
Asynchrony detection, ineffective effort (highest)97.8 %
Original publication
Enhancing mechanical ventilation management with AI: Computer vision for automated detection of ventilatory modes, parameters and asynchrony. J Crit Care. Published 22 August 2025. doi:10.
Composed by AI from the published abstract and checked against it. How Briefs are made →