Using Artificial Intelligence To Improve Ventilator Settings For Intensive Care Patients

Research on Intelligent Optimization of Ventilator Parameters for Intensive Care Patients Based on Multimodal Large Models

This observational study aims to determine whether an AI-assisted decision support system can improve clinical outcomes for mechanically ventilated pediatric patients (aged 1 month to 18 years) in the PICU, compared to standard care provided by medical staff. The primary question addressed is: Do patients whose ventilator parameter optimization decisions are guided by AI assistance achieve a greater number of ventilator-free days within 28 days than those managed by medical staff? By utilizing clinical data collected following tracheal intubation to generate AI-driven recommendations-and comparing these against the actual adjustments made by physicians-this study seeks to assess whether the AI-assisted decision support system can effectively improve clinical outcomes for mechanically ventilated patients in the PICU.

Study Overview

• Status: Completed
• Conditions: Acute Respiratory Distress Syndrome; Pneumonia in Children; Respiratory Failure (Pediatric Patients)

• Study Type: Observational
• Enrollment (Actual): 2000

Contacts and Locations

Study Locations

• China
  • Zhejiang
    • Wenzhou, Zhejiang, China, 325000
      • The Second Affiliated Hospital of Wenzhou Medical University and Yuying Children's Hospital

Participation Criteria

Eligibility Criteria

• Ages Eligible for Study: Child; Adult
• Accepts Healthy Volunteers: No

• Sampling Method: Non-Probability Sample

Study Population

Pediatric patients aged 1 month to 18 years admitted to the Pediatric Intensive Care Unit (PICU) of the Second Affiliated Hospital of Wenzhou Medical University and Yuying Children's Hospital, who are receiving mechanical ventilation.

Description

Inclusion Criteria:

1. PICU patients aged 1 month to 18 years.
2. Receiving invasive mechanical ventilation, expected to last ≥ 48 hours.
3. Informed consent signed prior to enrollment.

Exclusion Criteria:

1. Expected survival < 24 hours
2. Irreversible brain injury (GCS = 3 + absence of brainstem reflexes)
3. Severe congenital cardiopulmonary malformations affecting ventilation assessment
4. Pregnancy (must be ruled out in adolescent girls)
5. Currently participating in other ventilation intervention trials
6. Guardian refusal to participate

Study Plan

How is the study designed?

Number of groups / cohorts

1

Cohorts and Interventions

Group / Cohort

AI-Assisted Ventilator Parameter Optimization in Pediatric ICU; Clinical data from key time points following tracheal intubation in each pediatric patient were input into an AI system to generate recommendations. These recommendations were then compared against the actual adjustments made by physicians, enabling a counterfactual assessment to determine whether-had the AI's suggestions been adopted-the number of ventilator-free days within a 28-day period would have been superior.

What is the study measuring?

Primary Outcome Measures

Outcome Measure	Measure Description	Time Frame
Number of ventilator-free days within 28 days	Days survived and free from invasive ventilation	From the start of tracheal intubation until 28 days after tracheal intubation.

Secondary Outcome Measures

Outcome Measure	Measure Description	Time Frame
mortality rate	All-cause mortality at 28 and 90 days following tracheal intubation	28 and 90 days after the initiation of tracheal intubation
Mechanical Ventilation-Related Complications	Cumulative duration of mechanical ventilation, reintubation rate (within 48 hours of extubation), ventilator-associated pneumonia (VAP), barotrauma.	From the start of tracheal intubation to Day 28
Length of Hospital Stay	PICU Length of Stay, Total Hospital Length of Stay	The duration from the time of admission to discharge for pediatric patients-up to a maximum of three months.
Artificial Intelligence System Evaluation	Physician Adoption Rates and Outcomes of Cases Involving Discrepancies Between AI Recommendations and Physician Decisions	From the start of tracheal intubation to Day 28
Health Economics	PICU Hospitalization Costs	The duration from the time of admission to discharge for pediatric patients-up to a maximum of three months.

Collaborators and Investigators

• Sponsor: Wu Rongzhou

Publications and helpful links

General Publications

• Gattinoni L, Tonetti T, Cressoni M, Cadringher P, Herrmann P, Moerer O, Protti A, Gotti M, Chiurazzi C, Carlesso E, Chiumello D, Quintel M. Ventilator-related causes of lung injury: the mechanical power. Intensive Care Med. 2016 Oct;42(10):1567-1575. doi: 10.1007/s00134-016-4505-2. Epub 2016 Sep 12.
• Acute Respiratory Distress Syndrome Network; Brower RG, Matthay MA, Morris A, Schoenfeld D, Thompson BT, Wheeler A. Ventilation with lower tidal volumes as compared with traditional tidal volumes for acute lung injury and the acute respiratory distress syndrome. N Engl J Med. 2000 May 4;342(18):1301-8. doi: 10.1056/NEJM200005043421801.
• Fleuren LM, Klausch TLT, Zwager CL, Schoonmade LJ, Guo T, Roggeveen LF, Swart EL, Girbes ARJ, Thoral P, Ercole A, Hoogendoorn M, Elbers PWG. Machine learning for the prediction of sepsis: a systematic review and meta-analysis of diagnostic test accuracy. Intensive Care Med. 2020 Mar;46(3):383-400. doi: 10.1007/s00134-019-05872-y. Epub 2020 Jan 21.
• Char DS, Shah NH, Magnus D. Implementing Machine Learning in Health Care - Addressing Ethical Challenges. N Engl J Med. 2018 Mar 15;378(11):981-983. doi: 10.1056/NEJMp1714229. No abstract available.
• Pirracchio R, Petersen ML, Carone M, Rigon MR, Chevret S, van der Laan MJ. Mortality prediction in intensive care units with the Super ICU Learner Algorithm (SICULA): a population-based study. Lancet Respir Med. 2015 Jan;3(1):42-52. doi: 10.1016/S2213-2600(14)70239-5. Epub 2014 Nov 24.
• Chen JH, Asch SM. Machine Learning and Prediction in Medicine - Beyond the Peak of Inflated Expectations. N Engl J Med. 2017 Jun 29;376(26):2507-2509. doi: 10.1056/NEJMp1702071. No abstract available.
• Komorowski M, Celi LA, Badawi O, Gordon AC, Faisal AA. The Artificial Intelligence Clinician learns optimal treatment strategies for sepsis in intensive care. Nat Med. 2018 Nov;24(11):1716-1720. doi: 10.1038/s41591-018-0213-5. Epub 2018 Oct 22.
• Esteva A, Robicquet A, Ramsundar B, Kuleshov V, DePristo M, Chou K, Cui C, Corrado G, Thrun S, Dean J. A guide to deep learning in healthcare. Nat Med. 2019 Jan;25(1):24-29. doi: 10.1038/s41591-018-0316-z. Epub 2019 Jan 7.
• Topol EJ. High-performance medicine: the convergence of human and artificial intelligence. Nat Med. 2019 Jan;25(1):44-56. doi: 10.1038/s41591-018-0300-7. Epub 2019 Jan 7.

Study record dates

Study Major Dates

• Study Start (Actual): January 1, 2021
• Primary Completion (Actual): December 31, 2025
• Study Completion (Actual): December 31, 2025

Study Registration Dates

• First Submitted: May 10, 2026
• First Submitted That Met QC Criteria: May 15, 2026
• First Posted (Actual): May 18, 2026

Study Record Updates

• Last Update Posted (Actual): May 18, 2026
• Last Update Submitted That Met QC Criteria: May 15, 2026
• Last Verified: May 1, 2026

More Information

Terms related to this study

• Keywords: Ventilator; Pediatric Intensive Care Unit; Multimodal Large Language Model
• Additional Relevant MeSH Terms: Respiratory Tract Diseases; Lung Diseases; Respiration Disorders; Respiratory Distress Syndrome; Respiratory Insufficiency
• Other Study ID Numbers: 2026-K-78-01

Plan for Individual participant data (IPD)

• Plan to Share Individual Participant Data (IPD)?: NO
• IPD Plan Description: Individual participant data will not be shared, the study's ethical approvals and consent agreements do not permit public data sharing. Access may be considered upon reasonable request to the corresponding author, subject to institutional review and data use agreements to ensure patient privacy and compliance with regulations.

Drug and device information, study documents

• Studies a U.S. FDA-regulated drug product: No
• Studies a U.S. FDA-regulated device product: No