EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

SmartAlert: Machine learning-based patient-ventilator asynchrony detection system in intensive care units

Jaroslav Pažout, Milan Němý, Jakub Mikeš, Jan Jirman, Jan Kubr, Eliška Niebauerová, Miroslav Macík, Michal Pech, Michal Štajnrt, Jakub Vaněk, Petr Waldauf, Václav Zvoníček und 4 weitere

Computer Methods and Programs in Biomedicine · 2025

DOI: 10.1016/j.cmpb.2025.108927 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Computer Methods and Programs in Biomedicine, 2025. Verfasst von Jaroslav Pažout und 15 weiteren Autor:innen. Beteiligte Einrichtung: Department of Anesthesiology and Intensive Care, 3rd Faculty of Medicine, Charles University and Kralovske Vinohrady University Hospital in Prague, Š…. Hauptthemen laut MEDLINE: Intensive Care Units, Machine Learning, Respiration, Artificial, Ventilators, Mechanical. Die Arbeit wurde bisher 7-mal zitiert (Crossref) und nennt 27 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Czech Technical University in Prague Research Center for Informatics, European Commission, Ministry of Education Youth and Sports of the Czech Republic.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

BACKGROUND AND OBJECTIVE: Patient-ventilator asynchronies (PVA) are associated with ventilator-induced lung injury and increased mortality. Current detection methods rely on static thresholds, extensive preprocessing, or proprietary ventilator data. This study aimed to develop and validate a fully online, real-time system that detects and classifies PVAs directly from ventilator screen data while alerting clinicians based on severity. METHODS: The SmartAlert system was developed using ventilator screen recordings from ICU patients. It extracts pressure and flow waveforms from video recordings, converts them into time-series data, and employs deep neural networks to classify asynchronies and assign alarm levels from no urgency to most urgent. A dataset of 381,280 double-breath units was independently annotated by two expert intensivists. Two deep learning models were trained: one for alarm prediction and another for asynchrony classification (ineffective triggering, double cycling, high inspiratory effort, no asynchrony). Performance was evaluated using accuracy, sensitivity, specificity, and AUC-ROC, compared to expert consensus. RESULTS: SmartAlert demonstrated strong performance for alarm level prediction (overall accuracy: 83.8 %, weighted AUC-ROC: 0.943 [95 % CI: 0.941-0.945]) and PVA classification (weighted accuracy: 89.3 %, weighted AUC-ROC: 0.951 [95 % CI: 0.950-0.953]). It showed high specificity for urgent alarms (99.9 % for level 3) and PVA types (98.5 % for ineffective triggering, 96.9 % for double cycling, 94.8 % for high inspiratory effort). CONCLUSIONS: We developed and internally validated SmartAlert, an automated system that detects PVAs, classifies severity, and alerts clinicians in real time. Its potential to reduce alarm fatigue, optimize ventilator settings, and improve patient outcomes remains to be tested in clinical trials.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Intensive Care UnitsMachine LearningRespiration, ArtificialVentilators, MechanicalHumansROC CurveClinical AlarmsNeural Networks, ComputerDeep LearningReproducibility of ResultsPatient-Ventilator Asynchrony

Stichwörter der Autor:innen

Deep neural networksMechanical ventilationPatient-ventilator asynchronyReal-time monitoring

Autor:innen und Institutionen

  • Department of Anesthesiology and Intensive Care, 3rd Faculty of Medicine, Charles University and Kralovske Vinohrady University Hospital in Prague, Šrobárova 50, Prague 100 34, Czech Republic.
  • Division of Clinical Geriatrics, Center for Alzheimer Research, Department of Neurobiology, Care Sciences and Society, Karolinska Institute, Blickagången 16, Huddinge 14183, Sweden; Department of Cognitive Systems and N…
  • Department of Cognitive Systems and Neurosciences, Czech Institute of Informatics, Robotics and Cybernetics, Czech Technical University in Prague, Jugoslávských partyzánů 1580/3, Prague 160 00, Czech Republic.
  • Department of Computer Graphics and Interaction, Faculty of Electrical Engineering, Czech Technical University in Prague, Karlovo náměstí 13, Prague 121 35, Czech Republic.
  • Department of Artificial Intelligence, Czech Institute of Informatics, Robotics, and Cybernetics, Czech Technical University in Prague, Jugoslávských partyzánů 1580/3, Prague 160 00, Czech Republic; Cognitive Robotics,…

Förderung

  • Czech Technical University in Prague Research Center for Informatics
  • European Commission
  • Ministry of Education Youth and Sports of the Czech Republic
  • Ministry of Health of the Czech Republic

Zitieren

APA

Pažout, J., Němý, M., Mikeš, J., Jirman, J., Kubr, J., Niebauerová, E., Macík, M., Pech, M., Štajnrt, M., Vaněk, J., Waldauf, P., Zvoníček, V., Vysloužilová, L., Babuška, R., Duška, F., & VentConnect Study group (2025). SmartAlert: Machine learning-based patient-ventilator asynchrony detection system in intensive care units. Computer Methods and Programs in Biomedicine. https://doi.org/10.1016/j.cmpb.2025.108927

Vancouver

Pažout J, Němý M, Mikeš J, Jirman J, Kubr J, Niebauerová E, et al. SmartAlert: Machine learning-based patient-ventilator asynchrony detection system in intensive care units. Computer Methods and Programs in Biomedicine. 2025. doi:10.1016/j.cmpb.2025.108927

BibTeX
@article{Paout2025,
  author = {Pažout, Jaroslav and Němý, Milan and Mikeš, Jakub and Jirman, Jan and Kubr, Jan and Niebauerová, Eliška and Macík, Miroslav and Pech, Michal and Štajnrt, Michal and Vaněk, Jakub and Waldauf, Petr and Zvoníček, Václav and Vysloužilová, Lenka and Babuška, Robert and Duška, František and VentConnect Study group, VentConnect Study group},
  title = {SmartAlert: Machine learning-based patient-ventilator asynchrony detection system in intensive care units},
  journal = {Computer Methods and Programs in Biomedicine},
  year = {2025},
  doi = {10.1016/j.cmpb.2025.108927}
}
Kostenlos starten →