From Bench to Bedside: A Machine Learning Tool for the Detection of Inspiratory Leak

Status

Recruiting

Phases

Unknown

Study type

Observational

Source

ClinicalTrials.gov

Registry ID

NCT07428694

Enrollment

Registered

2026-02-24

Start date

2025-10-01

Completion date

2026-10-01

Last updated

2026-02-24

For informational purposes only — not medical advice. Sourced from public registries and may not reflect the latest updates. Terms

Conditions

Chronic Respiratory Failure

Brief summary

Study of the applicability of machine learning tools in detecting inspiratory leakage in longterm non-invasive ventilation. The study was conducted in two stages. Firstly the ML model was trained on both bench model created scenarios and then ten patients. And secondly the success of the model was assessed in a proof of concept pilot study of ten patients.

Interventions

None listed

Study design

Observational model

CASE_ONLY

Time perspective

PROSPECTIVE

Eligibility

Sex/Gender

ALL

Age

18 Years to No maximum

Healthy volunteers

Inclusion criteria

* elective hospitalisation for control of non-invasive ventilation * use of ResMedLumis 100/150 ventilator * treatment for \>3 months

Exclusion criteria

* current exacerbation

Design outcomes

Primary

Measure	Time frame	Description
Correct interpretation of inspiratory leak by machine learning tool	one year	Measured in comparison with god standard method of polygraphy

Countries

Norway

Outcome results