Preventive medicine reports · 2026
Temporal epidemiology and multi-source forecasting of hemorrhagic fever with renal syndrome and leptospirosis in mainland China: An interpretable machine learning study.
СтатьяНазвание и аннотация: Перевод пока не подготовлен
Обработка статьи · этапы, ошибки и управление
Загрузка состояния…
Ошибки и ограничения
История · было → стало
Состояние обновляется каждые 5 секунд, пока блок открыт. Процент показывается только там, где известен объём работы. Пауза останавливает продолжение этапа; текущий запрос к источнику может завершиться.
Gong Meiqian, Zhou Yinzhu, Chen Shuilian, Zhang Chi
Аннотация
OBJECTIVE: Hemorrhagic fever with renal syndrome (HFRS) and leptospirosis are environmentally associated zoonoses in China, yet their long-term temporal dynamics and predictive determinants have not been systematically compared.
METHODS: We analyzed monthly surveillance data on reported cases of HFRS and leptospirosis in mainland China from January 2010 to December 2025 using a multi-source data-driven framework. Five statistical, machine-learning, and deep-learning models were evaluated using historical, socioeconomic, meteorological, ecological, and behavioral predictors.
RESULTS: During the study period, 152,183 HFRS cases and 6288 leptospirosis cases were reported. HFRS declined significantly after 2019, whereas leptospirosis declined until 2018 and subsequently rebounded. Both diseases exhibited distinct seasonality, with HFRS peaking in late autumn and winter and leptospirosis in late summer and autumn. Extreme gradient boosting (XGBoost) with the combined predictor set achieved the best out-of-sample performance for both diseases. Shapley additive explanations identified 12-month-lagged incidence as the dominant predictor for both diseases, whereas precipitation and relative humidity contributed more strongly to leptospirosis than to HFRS.
CONCLUSIONS: HFRS and leptospirosis showed distinct epidemiological patterns and predictor contributions. Integrating multi-source predictors improved forecasting performance, while interpretable machine learning provided disease-specific insights to support early warning and targeted prevention of environmentally associated zoonoses.
METHODS: We analyzed monthly surveillance data on reported cases of HFRS and leptospirosis in mainland China from January 2010 to December 2025 using a multi-source data-driven framework. Five statistical, machine-learning, and deep-learning models were evaluated using historical, socioeconomic, meteorological, ecological, and behavioral predictors.
RESULTS: During the study period, 152,183 HFRS cases and 6288 leptospirosis cases were reported. HFRS declined significantly after 2019, whereas leptospirosis declined until 2018 and subsequently rebounded. Both diseases exhibited distinct seasonality, with HFRS peaking in late autumn and winter and leptospirosis in late summer and autumn. Extreme gradient boosting (XGBoost) with the combined predictor set achieved the best out-of-sample performance for both diseases. Shapley additive explanations identified 12-month-lagged incidence as the dominant predictor for both diseases, whereas precipitation and relative humidity contributed more strongly to leptospirosis than to HFRS.
CONCLUSIONS: HFRS and leptospirosis showed distinct epidemiological patterns and predictor contributions. Integrating multi-source predictors improved forecasting performance, while interpretable machine learning provided disease-specific insights to support early warning and targeted prevention of environmentally associated zoonoses.