Data for: Use of artificial intelligence and environmental data to estimate respiratory hospitalizations in support of SUS management

This dataset contains weekly aggregated data used in the study “Use of artificial intelligence and environmental data to estimate respiratory hospitalizations in support of SUS management”. It includes hospital admissions due to respiratory diseases (ICD-10 J00–J99) in the state of São Paulo, Brazil, from 2017 to 2022, combined with air pollution indicators (PM2.5, PM10, O3, NO2, SO2, CO) and meteorological variables (temperature, relative humidity, precipitation, among others). Data were obtained from official public sources (DATASUS, CETESB, INMET) and processed to support the development, training, and evaluation of CatBoost-based predictive models, using feature engineering such as temporal lags, moving averages, and seasonal components. All information is provided in aggregated, de-identified form. (2025-12-10)