European journal of clinical investigation · 2025
Prediction of major outcomes in patients with malignant hypertension using machine learning: A report from the West Birmingham malignant hypertension registry.
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.
Argyris Antonios A, Ishiguchi Hironori, Chen Yang, Zheng Yalin, Shantsila Alena, Shantsila Eduard, Beevers D Gareth, Lip Gregory Y H
Аннотация
BACKGROUND: Malignant hypertension (MHT) is a rare, yet severe condition with high morbidity and mortality. We aimed to assess the potential of machine learning (ML) algorithms in forecasting prognostic outcomes in MHT patients.
METHODS: Data from the West Birmingham MHT Registry were used. We evaluated the efficacy of 9 ML algorithms, CatBoost, Decision Tree (DT), Light-Gradient Boosting Machine (LightGBM), K-Nearest Neighbours (KNN), Logistic Regression (LR), Multi-Layer Perceptron (MLP), Random Forest (RF), Support Vector Machine (SVM) and XGBoost in predicting a composite outcome of all-cause mortality/dialysis. Evaluation metrics included the area under the receiver operating characteristic curve (AUC) and F1 score. SHapley Additive exPlanations values were employed to quantify the importance of each feature.
RESULTS: The cohort comprised 385 individuals with MHT (mean age 48 ± 13 years, 66% male). During a median follow-up of 11 (interquartile range: 3-18) years, 282 patients (73%) experienced the composite outcome. Among 24 demographic and clinical variables, 16 were selected into the ML models. The SVM, LR, and MLP models exhibited robust predictive performance, achieving AUCs of .81 (95% CI: .70-.90), .82 (95% CI: .71-.92) and .81 (95% CI: .71-.90), respectively. Furthermore, these models demonstrated high F1 scores (SVM: .75, LR: .80. MLP: .75). Age, smoking, follow-up systolic blood pressure, and baseline creatinine were commonly identified as primary prognostic features in both SVM and LR models.
CONCLUSIONS: The application of ML algorithms facilitates effective prediction of prognostic outcomes in MHT patients, illustrating their potential utility in clinical decision-making through more targeted risk stratification and individualised patient care.
METHODS: Data from the West Birmingham MHT Registry were used. We evaluated the efficacy of 9 ML algorithms, CatBoost, Decision Tree (DT), Light-Gradient Boosting Machine (LightGBM), K-Nearest Neighbours (KNN), Logistic Regression (LR), Multi-Layer Perceptron (MLP), Random Forest (RF), Support Vector Machine (SVM) and XGBoost in predicting a composite outcome of all-cause mortality/dialysis. Evaluation metrics included the area under the receiver operating characteristic curve (AUC) and F1 score. SHapley Additive exPlanations values were employed to quantify the importance of each feature.
RESULTS: The cohort comprised 385 individuals with MHT (mean age 48 ± 13 years, 66% male). During a median follow-up of 11 (interquartile range: 3-18) years, 282 patients (73%) experienced the composite outcome. Among 24 demographic and clinical variables, 16 were selected into the ML models. The SVM, LR, and MLP models exhibited robust predictive performance, achieving AUCs of .81 (95% CI: .70-.90), .82 (95% CI: .71-.92) and .81 (95% CI: .71-.90), respectively. Furthermore, these models demonstrated high F1 scores (SVM: .75, LR: .80. MLP: .75). Age, smoking, follow-up systolic blood pressure, and baseline creatinine were commonly identified as primary prognostic features in both SVM and LR models.
CONCLUSIONS: The application of ML algorithms facilitates effective prediction of prognostic outcomes in MHT patients, illustrating their potential utility in clinical decision-making through more targeted risk stratification and individualised patient care.