AI tool outperforms existing methods for identifying severe childhood pneumonia
· Medical Xpressedited by Lisa Lock, reviewed by Andrew Zinin
Lisa Lock
Scientific Editor
Meet our editorial team
Behind our editorial process
Andrew Zinin
Lead Editor
Meet our editorial team
Behind our editorial process Editors' notes
This article has been reviewed according to Science X's editorial process and policies. Editors have highlighted the following attributes while ensuring the content's credibility:
fact-checked
peer-reviewed publication
trusted source
proofread
The GIST Add as preferred source
Pneumonia remains the leading infectious cause of death among children under 5 worldwide, claiming almost 1 million lives each year. Researchers at University College Dublin have developed an artificial intelligence tool that can more accurately identify young children with pneumonia at serious risk of needing hospital treatment. Found to significantly outperform existing risk assessment methods used to identify children who require urgent referral to the hospital, the BIOTOPE algorithm has the potential to save lives in low-resource health care settings.
In a study published in the journal PLOS Medicine, the international team behind BIOTOPE described how the machine-learning algorithm was trained and validated using data from more than 2,500 children attending primary care clinics in Malawi.
Importantly, it has been designed to work within Malawi's existing Integrated Community Health Information System (iCHIS), allowing it to be used without placing additional administrative burdens on health care workers. The need for improved diagnostic support is particularly acute in Malawi, where there is approximately one doctor for every 28,000 people, compared with about one for every 250 people in Ireland.
Dr. Joe Gallagher, UCD School of Medicine, who led the study, said BIOTOPE demonstrated the potential for AI to support frontline health care workers making critical decisions. "What this research shows is that we can do much better for children who are severely ill with pneumonia," he said. "A child's life can depend on whether a health worker in a clinic correctly identifies how sick they are. We now have a tool that could help make that decision easier in the real world."
The research was carried out through the BIOTOPE (BIOmarkers TO diagnose PnEumonia) project, an international collaboration led by UCD involving researchers from Mzuzu University in Malawi, the University of Galway, Queen's University Belfast, the World Health Organization, the Malawi Ministry of Health and Luke International Norway.
The BIOTOPE algorithm uses a machine-learning technique known as a "random forest" model to analyze a wide range of factors simultaneously, including breathing rate, temperature, heart rate, oxygen levels, nutritional status and household conditions.
Researchers noted that current international referral guidelines can miss seriously ill children. Previous studies have found that many children who died from severe pneumonia did not display the standard warning signs typically used to trigger hospital referral.
"Machine learning gives us the ability to create something that improves over time rather than becoming obsolete," said Professor Cathal Seoighe, University of Galway. "This algorithm can be continuously retrained as new data accumulates, keeping it relevant across changing disease patterns and health system contexts."
"Pneumonia continues to kill far too many children who could be saved with timely referral and treatment," added Dr. Chris Watson, Queen's University Belfast, who said this project demonstrated "what can be achieved when researchers, clinicians and communities work together across borders.
"Embedding this tool into existing health information systems is exactly the kind of practical, scalable solution that can make a real difference on the ground."
The research also involved extensive public participation, with parents and caregivers helping shape the study's priorities and local artists creating a sculpture reflecting community hopes for the project and its potential impact on health care.
"This is science that starts and ends with the communities it is designed to serve," said Professor Balwani Mbakaya, Mzuzu University. "Building this in genuine partnership, and ensuring it works for health workers in the field without adding burden, has been at the heart of everything we have done."
Publication details
Patrick Staunton et al, Prediction of hospitalisation in young children with pneumonia in Malawi: A machine learning-based approach, PLOS Medicine (2026). DOI: 10.1371/journal.pmed.1005122
Journal information: PLoS Medicine
Key medical concepts
PneumoniaArtificial IntelligenceRandom Forest
Clinical categories
PediatricsChildren's healthPulmonary medicineCommon illnesses & PreventionInfectious diseases Provided by University College Dublin Who's behind this story?
Lisa Lock
BA art history, MA material culture. Former museum editor, paramedic, and transplant coordinator. Editing for Science X since 2021. Full profile →
Andrew Zinin
Master's in physics with research experience. Long-time science news enthusiast. Plays key role in Science X's editorial success. Full profile →
Citation: AI tool outperforms existing methods for identifying severe childhood pneumonia (2026, June 17) retrieved 17 June 2026 from https://medicalxpress.com/news/2026-06-ai-tool-outperforms-methods-severe.html This document is subject to copyright. Apart from any fair dealing for the purpose of private study or research, no part may be reproduced without the written permission. The content is provided for information purposes only.