Data-Driven Disease Surveillance

Pilot study of SIMG, a web-based pregnancy monitoring system in Brazil, reveals marginal usability and suboptimal uptake despite high willingness to use the tool in future pregnancies.

County-level analysis reveals disparities in cancer mortality decline across geography, income, and urbanization in the United States.

Study comparing machine learning and rule-based algorithms for identifying smokers in administrative health data found ML models doubled sensitivity for detecting current smokers.

Qualitative study examines acceptability of sharing internet browsing history for cancer research among diverse populations, identifying trust and transparency as key factors.

Ecological niche modeling reveals shifted environmental predictors and expanded geographic risk zones for highly pathogenic avian influenza H5 after 2020.

Analysis of dengue's economic burden in Puerto Rico from 2010–2023 reveals substantial costs during epidemics, underscoring need for enhanced prevention and resource allocation strategies.

US immigration enforcement agents use Palantir-developed software to access health records of millions of Americans for locating and detaining undocumented immigrants.

Examine how Finland's COVID-19 health monitoring systems shaped pandemic response and health equity through data governance decisions, revealing how aggregate data masked vulnerable populations.

XGBoost machine learning model demonstrates superior accuracy for monthly tuberculosis forecasting in coastal urban environments compared to traditional ARIMA and Prophet approaches.

Study evaluates quality of online osteoporosis information across 146 German and English websites using evidence-based criteria, finding significant gaps in diagnostic, treatment, and prevention.

Review examining social dimensions of HIV public health data systems, conceptual frameworks for ethical practice, and governance concerns for marginalized populations.

Hybrid explainable AI and Bayesian deep learning system predicts dengue outbreaks at district level in Bangladesh using climate, socioeconomic, and healthcare data from 2017-2024.

Protocol for evaluating code-based versus questionnaire-based influenza-like illness surveillance systems in Belgian general practices using CDC guidelines and multi-criteria decision analysis.

Standardized statistical framework for National Health Interview Survey analyses to improve methodological consistency, enable cross-study comparisons, and strengthen population health.

Machine learning analysis of environmental and socioeconomic determinants of tuberculosis incidence in Taiwan, identifying key drivers and nonlinear relationships for disease forecasting.

Review synthesizes big data and machine learning integration in U.S. infectious disease surveillance, examining data sources and computational methods for detection, forecasting, and control.

Mathematical modeling frameworks for infectious disease dynamics integrate computational methods, network analysis, and machine learning to forecast epidemics and optimize public health interventions.

Machine learning analysis of one million canine electronic health records identifies disease phenotypes, breed predispositions, and emerging health patterns using unsupervised topic modeling.

geeLite R package simplifies building local SQLite databases from Google Earth Engine data, enabling longitudinal environmental monitoring and time series analysis with portable.

Validation study of Georgia's immunization registry found nearly complete COVID-19 vaccine reporting among hospitalized patients in metropolitan Atlanta, 2023-2024.