Blood · 2026
The transposable element-PARP axis underpins synthetic lethality and immunogenic vulnerability in blood cancer.
СтатьяОбзорНазвание 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.
Li Ye, Yu Sihao, Yu Xiaojuan, Tian Bei, Tang Jiayan, Qu Haihong, Zhang Yongfang
Аннотация
PURPOSE: This study aimed to identify key risk factors and develop an explainable machine learning (ML) model for predicting early dysphagia in patients with acute ischemic stroke (AIS).
PATIENTS AND METHODS: In this cross-sectional study, 1041 patients with AIS were recruited from two tertiary hospitals. Participants were classified into a non-dysphagia group (n = 736) and a dysphagia group (n = 305). Feature selection was carried out using the Boruta algorithm and logistic regression. The dataset was randomly partitioned into a training set (n = 728) and a test set (n = 313) in a 7:3 ratio. Six ML models were trained with 10-fold cross-validation. Model performance was evaluated based on the area under the receiver operating characteristic curve (AUC-ROC), sensitivity, specificity, accuracy,positive predictive value (PPV), negative predictive value (NPV), F1-score and Youden's index. Key predictors were interpreted using SHapley Additive exPlanations (SHAP) analysis.
RESULTS: The incidence of early dysphagia with AIS was 29.3%. The Random Forest (RF) model demonstrated the best overall performance, with an AUC-ROC of 0.952 (95% CI: 0.927-0.976). The significant risk factors identified were Activities of Daily Living (ADL) grade, National Institutes of Health Stroke Scale (NIHSS) score, multifocal lesions, hypoalbuminemia, coronary heart disease, and lesion hemisphere.
CONCLUSION: ML models may serve as reliable assessment tools for predicting dysphagia in patients with AIS. The RF model demonstrated the best predictive performance. This predictive model could assist clinical healthcare providers in delivering early warnings and developing individualized treatment plans for high-risk patients.
PATIENTS AND METHODS: In this cross-sectional study, 1041 patients with AIS were recruited from two tertiary hospitals. Participants were classified into a non-dysphagia group (n = 736) and a dysphagia group (n = 305). Feature selection was carried out using the Boruta algorithm and logistic regression. The dataset was randomly partitioned into a training set (n = 728) and a test set (n = 313) in a 7:3 ratio. Six ML models were trained with 10-fold cross-validation. Model performance was evaluated based on the area under the receiver operating characteristic curve (AUC-ROC), sensitivity, specificity, accuracy,positive predictive value (PPV), negative predictive value (NPV), F1-score and Youden's index. Key predictors were interpreted using SHapley Additive exPlanations (SHAP) analysis.
RESULTS: The incidence of early dysphagia with AIS was 29.3%. The Random Forest (RF) model demonstrated the best overall performance, with an AUC-ROC of 0.952 (95% CI: 0.927-0.976). The significant risk factors identified were Activities of Daily Living (ADL) grade, National Institutes of Health Stroke Scale (NIHSS) score, multifocal lesions, hypoalbuminemia, coronary heart disease, and lesion hemisphere.
CONCLUSION: ML models may serve as reliable assessment tools for predicting dysphagia in patients with AIS. The RF model demonstrated the best predictive performance. This predictive model could assist clinical healthcare providers in delivering early warnings and developing individualized treatment plans for high-risk patients.