Predicting Extubation Problems With Machine Learning
Official title: Prediction of Extubation Failure in Mechanically Ventilated Adult Patients Using a Machine Learning-Based Approach
Run by Universitas Indonesia · NCT07918924 · Data updated 2026-11-12
This study tests whether computer models can predict extubation failure in adults on ventilators in the ICU.
Quick facts
| Study type | Observational |
|---|---|
| Ages | 18 Years and up |
| Sex | All |
| Enrollment | 1065 (estimated) |
| Sponsor | Universitas Indonesia Academic / other |
| Locations | 2 sites (Jakarta, DKI Jakarta and 1 other locations) |
| Status | Recruiting |
| Started | September 4, 2024 |
| Main results expected | January 23, 2027 |
| Study ends | January 31, 2027 |
| Registry record updated | November 5, 2026 |
| Trial ID | NCT07918924 on ClinicalTrials.gov |
Is this trial a fit for you?
✓Likely a good fit
Adults 18 years or older in the ICU who are on invasive mechanical ventilation and are planned for extubation.
✕Probably not a fit
Patients who do not experience extubation or reintubation events may see no direct benefit from the predictive tools.
A plain-English reading of the study's rules. Only the study team can confirm whether you qualify.
What this trial studies
This is a prospective, multicenter observational study enrolling adults receiving invasive mechanical ventilation in the ICU. Researchers will collect clinical data before extubation and follow patients for seven days after planned extubation to see if they need reintubation, a tracheostomy, or die. Five machine learning methods will be built and compared to identify which model best predicts post-extubation problems. Data will be gathered from bedside monitors, electronic records, and interviews, using predictors from patient characteristics, clinical status, ventilation parameters, and labs. Importantly, extubation decisions are made as part of routine care and are not influenced by the models, which are developed after data collection is complete.
Why it matters
Potential benefit: If successful, the models could help clinicians identify patients at higher risk of extubation failure to guide monitoring or preventive strategies.
How similar studies have gone: While there is growing interest in ML-based prediction for airway management, this specific multicenter approach remains relatively novel in its prospective, post-extubation prediction framework.
Who can join: the full rules
Show full inclusion / exclusion criteria
Inclusion Criteria: * Aged 18 years or older * Receiving invasive mechanical ventilation through a breathing tube in the windpipe (endotracheal tube) in the intensive care unit (ICU) Exclusion Criteria: \- Used a breathing machine through a tracheostomy (a surgical opening in the neck) when admitted to the ICU
Where it's running
-
Cipto Mangunkusumo National General Hospital (RSCM)
Jakarta, DKI Jakarta, Indonesia Recruiting -
Universitas Indonesia Hospital (RS UI)
Depok, West Java, Indonesia Completed
Study contacts
- Study coordinator: Adhrie Sugiarto
- Email: adhrie@gmail.com
- Phone: +62213157008
Similar studies
Other studies for the same condition or treatment. Recruiting studies are listed first.
-
Levetiracetam Dosing in Critically Ill on CRRT
This study tests how levetiracetam behaves in the bodies of adults on continuous renal replacement therapy to help figure out the best dosing.
-
Respiratory Physiotherapy in Critically Ill Pneumonia Patients
This study tests whether a structured breathing and chest therapy program can change inflammatory biomarkers in adults with pneumonia in the ICU compared to usual care.
-
Immersive VR for improving upper limb function in critically ill ICU patients
This trial tests whether fully immersive virtual reality training added to standard ICU rehabilitation can help improve upper limb strength and coordination in adults with ICU-acquired weakness.
-
Cycling with immersive VR for ICU patients recovering from critical illness
In this crossover study, critically ill adults in the ICU will try three 10-minute cycling sessions to see if an immersive VR environment motivates them more than cycling with no visuals or cycling while watching a route on a TV.
-
Factors Influencing Outcomes in ECMO and Impella Patients
This study tests whether metabolic and thyroid hormone changes help predict survival for adults on ECMO or Impella at a single Hungarian cardiac ICU.
-
Intravenous Fish Oil for Critically Ill Patients with High Inflammation
This study tests whether giving IV fish oil at two different doses helps reduce inflammation in critically ill adults on breathing support.
How to take part
- Review the eligibility criteria above with your treating physician.
- Check the official listing on ClinicalTrials.gov for the latest contact details and recruitment status.
- Contact the study coordinator or principal investigator to ask for pre-screening. Pre-screening is free and never obligates you to enroll.