Patient-Ventilator Dyssynchrony Detection With a Machine Learning Algorithm

Automated Detection and Classification of Patient-Ventilator Dyssynchrony With a Machine Learning Algorithm

This is a diagnostic study aiming to compare accuracy to detect and classify patient-ventilator dyssynchronies by a machine learning algorithm, compared to the gold-standard defined as dyssynchronies diagnosed and classified by mechanical ventilator and esophageal pressure waveforms analyzed by experts.

The main question of this study is:

• Are patient-ventilator dyssynchronies accurately detected and classified by an artificial intelligence algorithm when compared to experts analyzing esophageal pressure and mechanical ventilator waveforms?

Study Overview

• Status: Recruiting
• Conditions: Respiratory Failure
• Intervention / Treatment: Device: Artificial Intelligence Detection and Classification of Patient-Ventilator Dyssynchronies

Detailed Description

This is a diagnostic, observational study, aiming to assess patient-ventilator dyssynchrony automated detection and classification by a machine learning algorithm. Accuracy of the machine learning algorithm will be compared with the gold-standard, defined as dyssynchronies detected and classified by mechanical ventilation experts.

Experts will analyzed airway pressure, flow, volume and esophageal pressure waveforms to detect and classify dyssynchronies.

• Study Type: Observational
• Enrollment (Estimated): 80

Contacts and Locations

• Study Contact: Name: Glauco M Plens, MD; Phone Number: +5511982213020; Email: glaucomplens@gmail.com

Study Locations

• Brazil
  • SP
    • Sao Paulo, SP, Brazil, 05403900
      • Recruiting
      • Heart Institute, University of São Paulo
      • Contact: Glauco M Plens, MD; Phone Number: +5511982213020; Email: glaucomplens@gmail.com

Participation Criteria

Eligibility Criteria

• Ages Eligible for Study: Adult; Older Adult
• Accepts Healthy Volunteers: N/A

• Sampling Method: Non-Probability Sample

Study Population

Adult subjects under mechanical ventilation, with an assisted or assist-controlled mode, who are monitored with an esophageal pressure balloon due to clinical indication are eligible.

Description

Inclusion Criteria:

• Subjects under assisted or assist-controlled mechanical ventilation and monitored with esophageal pressure balloon.

Exclusion Criteria:

• Refusal from patient's family or attending physician

Study Plan

How is the study designed?

Number of groups / cohorts

1

Cohorts and Interventions

Group / Cohort

Intervention / Treatment

Artificial Intelligence Detection and Classification of Patient-Ventilator Dyssynchronies; This is a single arm study, since all subjects included will be exposed to both diagnostic methods (artificial intelligence and experts). The proposed diagnostic method is a machine learning algorithm integrated in the mechanical ventilator FlexiMag Max 700 (Magnamed, Brazil), which will continuously record data from mechanical ventilation of included subjects for a time period of up to 72 hours. The gold-standard involves esophageal pressure waveform recording and offline analysis by experts.

Device: Artificial Intelligence Detection and Classification of Patient-Ventilator Dyssynchronies; Machine learning algorithm to detect and classify patient-ventilator dyssynchronies, which is integrated in the mechanical ventilator (Fleximag Max, Magnamed, Brazil).

What is the study measuring?

Primary Outcome Measures

Outcome Measure	Measure Description	Time Frame
Diagnostic Accuracy of the Artificial Intelligence algorithm	Sensitivity, specificity, positive predictive value, negative predictive value of the artificial intelligence algorithm to detect and classify patient-ventilator dyssynchronies. These accuracy indexes will be estimated for each kind of dyssinchrony: ineffective effort, autotriggering, double triggering, reverse triggering, reverse triggering with a double cycle	3 days

Secondary Outcome Measures

Outcome Measure	Measure Description	Time Frame
Pendelluft detection	Percentage of cycles with pendelluft detected with the artificial intelligence algorithm compared to the percentage of cycles with pendelluft detected with the esophageal pressure	3 days

Collaborators and Investigators

• Sponsor: University of Sao Paulo General Hospital
• Collaborators: Magnamed Tecnologia Medica S/A
• Investigators: Study Director: Eduardo LV Costa, MD, PhD, University of Sao Paulo

Publications and helpful links

General Publications

• Amato MB, Barbas CS, Medeiros DM, Magaldi RB, Schettino GP, Lorenzi-Filho G, Kairalla RA, Deheinzelin D, Munoz C, Oliveira R, Takagaki TY, Carvalho CR. Effect of a protective-ventilation strategy on mortality in the acute respiratory distress syndrome. N Engl J Med. 1998 Feb 5;338(6):347-54. doi: 10.1056/NEJM199802053380602.
• Blanch L, Villagra A, Sales B, Montanya J, Lucangelo U, Lujan M, Garcia-Esquirol O, Chacon E, Estruga A, Oliva JC, Hernandez-Abadia A, Albaiceta GM, Fernandez-Mondejar E, Fernandez R, Lopez-Aguilar J, Villar J, Murias G, Kacmarek RM. Asynchronies during mechanical ventilation are associated with mortality. Intensive Care Med. 2015 Apr;41(4):633-41. doi: 10.1007/s00134-015-3692-6. Epub 2015 Feb 19.
• 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.
• LeCun Y, Bengio Y, Hinton G. Deep learning. Nature. 2015 May 28;521(7553):436-44. doi: 10.1038/nature14539.
• Sousa MLEA, Magrans R, Hayashi FK, Blanch L, Kacmarek RM, Ferreira JC. Clusters of Double Triggering Impact Clinical Outcomes: Insights From the EPIdemiology of Patient-Ventilator aSYNChrony (EPISYNC) Cohort Study. Crit Care Med. 2021 Sep 1;49(9):1460-1469. doi: 10.1097/CCM.0000000000005029.
• Sousa MLA, Magrans R, Hayashi FK, Blanch L, Kacmarek RM, Ferreira JC. Predictors of asynchronies during assisted ventilation and its impact on clinical outcomes: The EPISYNC cohort study. J Crit Care. 2020 Jun;57:30-35. doi: 10.1016/j.jcrc.2020.01.023. Epub 2020 Jan 21.

Study record dates

Study Major Dates

• Study Start (Actual): May 25, 2024
• Primary Completion (Estimated): May 24, 2025
• Study Completion (Estimated): December 24, 2025

Study Registration Dates

• First Submitted: May 24, 2024
• First Submitted That Met QC Criteria: July 10, 2024
• First Posted (Actual): July 17, 2024

Study Record Updates

• Last Update Posted (Actual): July 17, 2024
• Last Update Submitted That Met QC Criteria: July 10, 2024
• Last Verified: May 1, 2024

More Information

Terms related to this study

• Keywords: mechanical ventilation; artificial intelligence
• Additional Relevant MeSH Terms: Respiratory Tract Diseases; Respiration Disorders; Signs and Symptoms, Respiratory; Respiratory Insufficiency; Patient-Ventilator Asynchrony
• Other Study ID Numbers: CAAE 78855824.7.0000.0068

Plan for Individual participant data (IPD)

• Plan to Share Individual Participant Data (IPD)?: NO

Drug and device information, study documents

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