• HOME
  • NEWS
  • EXPLORE
    • CAREER
      • Companies
      • Jobs
    • EVENTS
    • iGEM
      • News
      • Team
    • PHOTOS
    • VIDEO
    • WIKI
  • BLOG
  • COMMUNITY
    • FACEBOOK
    • INSTAGRAM
    • TWITTER
Thursday, November 19, 2026
BIOENGINEER.ORG
No Result
View All Result
  • Login
  • HOME
  • NEWS
  • EXPLORE
    • CAREER
      • Companies
      • Jobs
        • Lecturer
        • PhD Studentship
        • Postdoc
        • Research Assistant
    • EVENTS
    • iGEM
      • News
      • Team
    • PHOTOS
    • VIDEO
    • WIKI
  • BLOG
  • COMMUNITY
    • FACEBOOK
    • INSTAGRAM
    • TWITTER
  • HOME
  • NEWS
  • EXPLORE
    • CAREER
      • Companies
      • Jobs
        • Lecturer
        • PhD Studentship
        • Postdoc
        • Research Assistant
    • EVENTS
    • iGEM
      • News
      • Team
    • PHOTOS
    • VIDEO
    • WIKI
  • BLOG
  • COMMUNITY
    • FACEBOOK
    • INSTAGRAM
    • TWITTER
No Result
View All Result
Bioengineer.org
No Result
View All Result
Home NEWS Science News Health

Machine Learning Model Predicts High-Flow Nasal Cannula Failure in COVID-19 Patients

by
November 9, 2026
in Health
Reading Time: 3 mins read
0
Machine Learning Model Predicts High-Flow Nasal Cannula Failure in COVID-19 Patients

Machine Learning Model Predicts High-Flow Nasal Cannula Failure in COVID-19 Patients

Share on FacebookShare on TwitterShare on LinkedinShare on RedditShare on Telegram

In a prospective observational cohort study of critically ill patients with COVID-19-related acute hypoxemic respiratory failure, machine learning models integrating baseline clinical and metabolomic data accurately predicted the failure of high-flow nasal cannula therapy.

High-flow nasal cannula support is widely used as a first-line non-invasive respiratory therapy in acute hypoxemic respiratory failure, a role that expanded substantially during the COVID-19 pandemic. However, treatment failure remains frequent among treated patients, and delayed intubation is associated with worse clinical outcomes, underscoring the clinical need for reliable early predictive tools to guide decision-making in the intensive care unit.

Conducted at a single tertiary intensive care unit by researchers in Barcelona, Spain, the study analyzed 111 adult patients with COVID-19-related acute hypoxemic respiratory failure treated with high-flow nasal cannula. The researchers divided the cohort into a training cohort of 89 patients treated between March 2020 and April 2021, and a temporally independent validation cohort of 22 patients treated between May 2021 and October 2021.

At baseline, investigators recorded a total of 164 variables for each patient, encompassing 25 clinical parameters and 139 metabolomic features. Predictive models were developed using Random Forest machine learning algorithms combined with recursive feature elimination and bootstrap resampling, with performance evaluated by calculating the area under the receiver operating characteristic curve through internal and temporally independent validation.

A six-variable model combining clinical and metabolomic features achieved an area under the receiver operating characteristic curve of 0.87 (95% confidence interval: 0.79 to 0.94) in the training cohort and 0.91 (95% confidence interval: 0.78 to 1.00) in the validation cohort. A simplified three-variable model incorporating the ratio of oxygen saturation to fraction of inspired oxygen, anthranilic acid, and octanoic acid achieved an area under the curve of 0.71 (95% confidence interval: 0.61 to 0.82) in training and 0.91 (95% confidence interval: 0.77 to 1.00) in validation, with both models demonstrating improved training performance relative to the ROX index.

The investigators noted that the predictive performance observed in the validation cohort must be interpreted cautiously because of its limited sample size of 22 patients. While the simplified three-variable model retained performance comparable to the broader model and supports further evaluation for clinical utility, the authors concluded that validation in larger and more diverse patient populations remains necessary.

While high-flow nasal cannula therapy became a cornerstone of initial respiratory management throughout the COVID-19 pandemic, treatment failure remained a frequent occurrence in intensive care units. When high-flow oxygen fails to stabilize a patient, postponing invasive mechanical ventilation can lead to worse clinical outcomes.

To construct their predictive framework, the researchers gathered an extensive baseline profile comprising 164 distinct variables from each participant upon admission. To navigate this high-dimensional feature space, the investigators implemented Random Forest machine learning coupled with recursive feature elimination and bootstrap resampling.

While the comprehensive six-variable model achieved an area under the receiver operating characteristic curve of 0.87 in the training cohort and 0.91 in validation, the simplified configuration relied on just three specific parameters: the ratio of pulse oxygen saturation to fraction of inspired oxygen, anthranilic acid, and octanoic acid.

Patients recruited between March 2020 and April 2021 formed the primary training cohort of 89 individuals, whereas 22 patients admitted between May 2021 and October 2021 constituted the validation cohort. Although the simplified model maintained an area under the curve of 0.91 in this prospective validation sample, the authors underscored that the limited cohort size warrants careful interpretation.

The published findings reflect an interdisciplinary effort combining intensive care medicine, metabolomics, and biomedical signal processing across institutions in Barcelona, including Hospital del Mar, Universitat Pompeu Fabra, and the Institute for Bioengineering of Catalonia.

Subject of Research: Predicting high-flow nasal cannula failure in COVID-19 acute hypoxemic respiratory failure

Article Title: Prediction of high-flow nasal cannula failure in COVID-19 acute hypoxemic respiratory failure using combined clinical and metabolomic data: a machine learning study

Article References: Parrilla-Gómez, F. J., Nebot, P., Oller-Moreno, S., Marin-Corral, J., Castellví-Font, A., Vilà-Vilardell, C., Pérez-Terán, P., Granado-García, A., Muñoz-Bermúdez, R., Ahuir-Bauset, L., Soriano Rodriguez, C., Camañes Mayordomo, C., Haro, N., Gomez-Gomez, A., Martín-López, E., Pozo, O. J., Marco, S., & Masclans Enviz, J. R. (2026). Prediction of high-flow nasal cannula failure in COVID-19 acute hypoxemic respiratory failure using combined clinical and metabolomic data: a machine learning study. Frontiers in Medicine, 13, Article 1946899. https://doi.org/10.3389/fmed.2026.1946899

Image Credits: AI Generated

DOI: 10.3389/fmed.2026.1946899

Keywords: COVID-19, high-flow nasal cannula, acute hypoxemic respiratory failure, metabolomics, machine learning, Random Forest, intensive care unit, predictive modeling, Prediction, high-flow, nasal, cannula

News Source: Teresa Odom. (November 9, 2026). Machine Learning Model Predicts High-Flow Nasal Cannula Failure in COVID-19 Patients. Scienmag.

Tags: acute hypoxemic respiratory failurecannulaCOVID-19High-flowhigh-flow nasal cannulaintensive care unitMachine LearningMetabolomicsNasalpredictionpredictive modelingRandom Forest
Share12Tweet7Share2ShareShareShare1

Related Posts

Pune Cash Aid Beneficiaries Report High Perceived Independence but Limited Asset Ownership

Pune Cash Aid Beneficiaries Report High Perceived Independence but Limited Asset Ownership

November 19, 2026
Meta-Analysis Identifies Graft Elongation as Distinct Mode of ACL Reconstruction Failure

Meta-Analysis Identifies Graft Elongation as Distinct Mode of ACL Reconstruction Failure

November 19, 2026

Hospitalized Vasculitis Patients Face Heavy Early Mortality and Costs in Italy

November 19, 2026

Campylobacter jejuni in Ethiopian Beef and Raw Milk Shows High Multidrug Resistance

November 19, 2026

POPULAR NEWS

  • No Metabolic Crises Occurred With Peripartum Dextrose in Small Observational Study

    No Metabolic Crises Occurred With Peripartum Dextrose in Small Observational Study

    29 shares
    Share 12 Tweet 7
  • Mental Health Emergency Visits in Saudi Arabia Surge During Late Autumn and Winter

    29 shares
    Share 12 Tweet 7
  • Rare APBA1 Genetic Variants Identified in Children with Severe Monogenic Obesity

    29 shares
    Share 12 Tweet 7
  • RNA Thermometer Controls Recombinant Protein Expression in Engineered Escherichia Coli Cultures

    29 shares
    Share 12 Tweet 7

About

We bring you the latest biotechnology news from best research centers and universities around the world. Check our website.

Follow us

Recent News

Pune Cash Aid Beneficiaries Report High Perceived Independence but Limited Asset Ownership

Meta-Analysis Identifies Graft Elongation as Distinct Mode of ACL Reconstruction Failure

Acoustic Chaos and Entropy Metrics Track Laboratory Fatigue Damage in Asphalt Beams

Subscribe to Blog via Email

Enter your email address to subscribe to this blog and receive notifications of new posts by email.

Join 89 other subscribers
  • Contact Us

Bioengineer.org © Copyright 2023 All Rights Reserved.

Welcome Back!

Login to your account below

Forgotten Password?

Retrieve your password

Please enter your username or email address to reset your password.

Log In
No Result
View All Result
  • Homepages
    • Home Page 1
    • Home Page 2
  • News
  • National
  • Business
  • Health
  • Lifestyle
  • Science

Bioengineer.org © Copyright 2023 All Rights Reserved.