The use of artificial intelligence in assessing fever in the returning traveller: A scoping review
- 1Michael G. DeGroote School of Medicine, McMaster University, Hamilton, Ontario, Canada
- 2Division of General Internal Medicine, Michael G. DeGroote School of Medicine, McMaster University Medical Center, Hamilton, Ontario, Canada
- 3Division of Infectious Diseases, Michael G. DeGroote School of Medicine, McMaster University Medical Center, Hamilton, Ontario, Canada
Abstract
Fever in the returning traveller is a challenging presentation that requires a wide array of clinical expertise and draws on complex medical decision-making. Recent advancements in artificial intelligence have shown it can be a helpful clinical support tool in this context. This scoping review, conducted in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses extension for Scoping Reviews guidelines and registered on the Open Science Framework (https://doi.org/10.17605/OSF.IO/BSPKX), aimed to identify the current state of artificial intelligence use in assessing fever in the returning traveller. Here we describe the three eligible studies evaluating three distinct AI approaches identified in the literature. These heterogenous studies reported promising performance on study-specific endpoints in controlled settings. Given the lack of external validation, narrow populations studied, and absence of real-world implementation or user acceptance testing, further research is needed to explore this technology in a real-world clinical setting. The limited available evidence suggests that AI may have future potential to support assessment of fever in the returning traveller; however, clinical utility, safety, and real-world effectiveness remain unestablished. Large language models represent an emerging area of interest but were not evaluated in the studies included in this review. Author summary: We wanted to understand how artificial intelligence has been used to help clinicians assess fever in people who become ill after international travel. This clinical problem is difficult because the list of possible infections is broad and depends on travel destination, exposure history, symptoms, and test results. We reviewed the published literature and found three eligible studies. These studies described different approaches: a rule-based expert system to generate a ranked list of possible diagnoses, a machine-learning model to predict malaria risk, and a model using immune markers to help distinguish between infections. All three studies suggested that artificial intelligence could support clinical assessment in controlled settings. We also found that all studies were conducted in hospital-based populations. None of the included studies reported real-world clinical implementation or user acceptability testing, and external validation was limited. We believe this review shows that artificial intelligence is a promising but early area in travel-related fever assessment. More research, including external validation and real-world implementation studies, is needed before these tools can be confidently adopted in routine clinical care.
Related
- Notification: Artificial Intelligence in Autism Spectrum Disorder Diagnosis: A Scoping Review of Face, Voice, and Text Analysis Methods.
- Assessing spoken discourse in aphasia using multimodal artificial intelligence
- The applications of artificial intelligence in liver transplantation: A comprehensive scoping review
- Artificial Intelligence Versus Conventional Scoring for Assessing Small Bowel Capsule Endoscopy Cleanliness in Crohn's Disease
- Mapping Artificial Intelligence Research in Aphasia: An AI-Assisted Scoping Review and Semantic Network Analysis
BibTeX
@article{gandhi2026artificial,
title = {{The use of artificial intelligence in assessing fever in the returning traveller: A scoping review}},
author = {Bhavya Gandhi and Leo Morjaria and Imeth Illamperuma and Meha Bhatt and Andrew Kapoor},
journal = {PLOS Digital Health},
year = {2026},
month = oct,
doi = {10.1371/journal.pdig.0001742},
url = {https://journals.plos.org/digitalhealth/article?id=10.1371/journal.pdig.0001742},
note = {CC BY 4.0},
}
RIS
TY - JOUR
TI - The use of artificial intelligence in assessing fever in the returning traveller: A scoping review
AU - Bhavya Gandhi
AU - Leo Morjaria
AU - Imeth Illamperuma
AU - Meha Bhatt
AU - Andrew Kapoor
PY - 2026
DA - 2026/10/01
JO - PLOS Digital Health
DO - 10.1371/journal.pdig.0001742
UR - https://journals.plos.org/digitalhealth/article?id=10.1371/journal.pdig.0001742
AB - Fever in the returning traveller is a challenging presentation that requires a wide array of clinical expertise and draws on complex medical decision-making. Recent advancements in artificial intelligence have shown it can be a helpful clinical support tool in this context. This scoping review, conducted in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses extension for Scoping Reviews guidelines and registered on the Open Science Framework (https://doi.org/10.17605/OSF.IO/BSPKX), aimed to identify the current state of artificial intelligence use in assessing fever in the returning traveller. Here we describe the three eligible studies evaluating three distinct AI approaches identified in the literature. These heterogenous studies reported promising performance on study-specific endpoints in controlled settings. Given the lack of external validation, narrow populations studied, and absence of real-world implementation or user acceptance testing, further research is needed to explore this technology in a real-world clinical setting. The limited available evidence suggests that AI may have future potential to support assessment of fever in the returning traveller; however, clinical utility, safety, and real-world effectiveness remain unestablished. Large language models represent an emerging area of interest but were not evaluated in the studies included in this review. Author summary: We wanted to understand how artificial intelligence has been used to help clinicians assess fever in people who become ill after international travel. This clinical problem is difficult because the list of possible infections is broad and depends on travel destination, exposure history, symptoms, and test results. We reviewed the published literature and found three eligible studies. These studies described different approaches: a rule-based expert system to generate a ranked list of possible diagnoses, a machine-learning model to predict malaria risk, and a model using immune markers to help distinguish between infections. All three studies suggested that artificial intelligence could support clinical assessment in controlled settings. We also found that all studies were conducted in hospital-based populations. None of the included studies reported real-world clinical implementation or user acceptability testing, and external validation was limited. We believe this review shows that artificial intelligence is a promising but early area in travel-related fever assessment. More research, including external validation and real-world implementation studies, is needed before these tools can be confidently adopted in routine clinical care.
SN - 2767-3170
ER -