Machine learning–based surveillance and forecasting of infectious diseases: a systematic review of modeling approaches and multi-source data integration

Background Enteric infections, including rotavirus gastroenteritis, place a heavy burden on children and low-resource health systems. Surveillance still depends on laboratory-confirmed case reporting that moves slowly through hospital information systems (HIS), electronic medical records (EMR), and laboratory information systems (LIS), offering little predictive lead time. Digital health infrastructure and artificial intelligence (AI) now offer a way to rebuild surveillance around combined, near-real-time data. Methods Following the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) 2020, we searched PubMed and IEEE Xplore (through July 2026, with a supplementary PubMed search in September 2026) for studies using statistical, machine learning, deep learning, or multi-source AI-driven methods for infectious disease surveillance or forecasting, with attention to enteric infections and methodologically transferable work. Additional studies were identified through reference citation searching. Two reviewers screened records, extracted data, and assessed risk of bias (PROBAST for 23 prediction-model studies; narrative appraisal for 5 ecological association studies; contextual studies were not formally appraised). Results Thirty-three studies were included: 28 formal studies — statistical models ( n = 5), machine learning approaches ( n = 8), deep learning approaches ( n = 4), and multi-source digital surveillance ( n = 11) — and five contextual studies (environmental monitoring, virus–host association, simulation, surveillance-signal, and network studies) analyzed separately. Data sources broadened over time from meteorological and clinical records to EMR, telehealth, wastewater, internet-search, social-media, water-quality, and genomic streams. Machine learning and deep learning approaches were increasingly applied alongside conventional methods to model nonlinear and lagged exposure–response patterns and to integrate heterogeneous data streams, and multi-source systems—including wastewater-based and syndromic designs—reported earlier signal detection than single-source designs. Most evidence, however, is retrospective in design—spanning single-site to national and multinational datasets—and external validation and prospective deployment were largely absent. Conclusions The reviewed evidence suggests that AI-supported approaches may improve infectious disease surveillance by enabling earlier detection of epidemiological changes and supporting more timely public health responses. Operational use will depend less on retrospective accuracy than on external validation, interpretability, and integration with existing surveillance infrastructure.

Authors

Institutions

Publication Details

Journal
BMC Infectious Diseases
Published
2026-11-18
DOI
https://doi.org/10.1186/s12879-026-14824-z
Primary Topic
Data-Driven Disease Surveillance
Type
article
Field-Weighted Citation Impact
0.00
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
article

Machine learning–based surveillance and forecasting of infectious diseases: a systematic review of modeling approaches and multi-source data integration

Chunmin Pu, Yu Huang, Yuyan Tan, Shunjiang Ma et al.
2 links
BMC Infectious Diseases
Data-Driven Disease Surveillance
article

Machine learning–based surveillance and forecasting of infectious diseases: a systematic review of modeling approaches and multi-source data integration

Chunmin Pu, Yu Huang, Yuyan Tan, Shunjiang Ma, Yang Nian, Xiuzhu Li, Meng He
article en

Abstract

Background Enteric infections, including rotavirus gastroenteritis, place a heavy burden on children and low-resource health systems. Surveillance still depends on laboratory-confirmed case reporting that moves slowly through hospital information systems (HIS), electronic medical records (EMR), and laboratory information systems (LIS), offering little predictive lead time. Digital health infrastructure and artificial intelligence (AI) now offer a way to rebuild surveillance around combined, near-real-time data. Methods Following the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) 2020, we searched PubMed and IEEE Xplore (through July 2026, with a supplementary PubMed search in September 2026) for studies using statistical, machine learning, deep learning, or multi-source AI-driven methods for infectious disease surveillance or forecasting, with attention to enteric infections and methodologically transferable work. Additional studies were identified through reference citation searching. Two reviewers screened records, extracted data, and assessed risk of bias (PROBAST for 23 prediction-model studies; narrative appraisal for 5 ecological association studies; contextual studies were not formally appraised). Results Thirty-three studies were included: 28 formal studies — statistical models ( n = 5), machine learning approaches ( n = 8), deep learning approaches ( n = 4), and multi-source digital surveillance ( n = 11) — and five contextual studies (environmental monitoring, virus–host association, simulation, surveillance-signal, and network studies) analyzed separately. Data sources broadened over time from meteorological and clinical records to EMR, telehealth, wastewater, internet-search, social-media, water-quality, and genomic streams. Machine learning and deep learning approaches were increasingly applied alongside conventional methods to model nonlinear and lagged exposure–response patterns and to integrate heterogeneous data streams, and multi-source systems—including wastewater-based and syndromic designs—reported earlier signal detection than single-source designs. Most evidence, however, is retrospective in design—spanning single-site to national and multinational datasets—and external validation and prospective deployment were largely absent. Conclusions The reviewed evidence suggests that AI-supported approaches may improve infectious disease surveillance by enabling earlier detection of epidemiological changes and supporting more timely public health responses. Operational use will depend less on retrospective accuracy than on external validation, interpretability, and integration with existing surveillance infrastructure.

BMC Infectious Diseases
Kunming Medical University (CN)
Good health and well-being
Data-Driven Disease Surveillance
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Laika access coming soon