Health science reports · 2026
Identifying the Most Important Influencing Factors Based on Demographic Information, Nutrition, Physical Activity, Leisure Time, and Underlying Diseases in the Severity of Disease in COVID-19 Patients Using Data Mining Techniques.
СтатьяНазвание и аннотация: Перевод в очереди
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.
Mohammadifard Noushin, Sarafzadegan Nizal, Doost Fahimeh Haghighat, Najafian Jamshid, Rouhani Mohammad Hossein, Askari Gholamreza, Sattari Mohammad
Аннотация
BACKGROUND AND AIMS: The COVID-19 pandemic has affected millions of individuals worldwide and resulted in substantial mortality. Data mining and machine learning techniques enable the analysis of comprehensive patient data, facilitating the identification of key patterns and determinants that support clinical and preventive decision-making.
METHODS: This study evaluates the performance of three machine learning approaches-Deep Learning, Gradient Boosted Decision Trees (GBDT), and Support Vector Machine (SVM)-in identifying factors associated with COVID-19 severity. A comparative analysis based on ROC curves and quantitative performance metrics demonstrates that Deep Learning and GBDT outperform SVM.
RESULTS: All models consistently identify hypertension as a major risk factor. However, differences emerge in other influential variables: GBDT ranks hypertension as the most significant factor, followed by stomach ulcers, whereas Deep Learning highlights vitamin intake as the primary determinant. The deep learning tree variant further supports the importance of vitamin consumption.
CONCLUSION: The superior predictive performance of Deep Learning and GBDT provides a reliable benchmark and yields novel insights, including the potential roles of stomach ulcers and vitamin intake in disease severity. These findings emphasize the value of employing multiple high-performing models and robust validation strategies for accurate identification of clinical risk factors in COVID-19.
METHODS: This study evaluates the performance of three machine learning approaches-Deep Learning, Gradient Boosted Decision Trees (GBDT), and Support Vector Machine (SVM)-in identifying factors associated with COVID-19 severity. A comparative analysis based on ROC curves and quantitative performance metrics demonstrates that Deep Learning and GBDT outperform SVM.
RESULTS: All models consistently identify hypertension as a major risk factor. However, differences emerge in other influential variables: GBDT ranks hypertension as the most significant factor, followed by stomach ulcers, whereas Deep Learning highlights vitamin intake as the primary determinant. The deep learning tree variant further supports the importance of vitamin consumption.
CONCLUSION: The superior predictive performance of Deep Learning and GBDT provides a reliable benchmark and yields novel insights, including the potential roles of stomach ulcers and vitamin intake in disease severity. These findings emphasize the value of employing multiple high-performing models and robust validation strategies for accurate identification of clinical risk factors in COVID-19.