Xia Yuhan, Huang Yuezhong, Gong Min, Liu Weirong, Meng Yuanhui, Wu Huiyang, Zhang Hui, Zhang Hao, Weng Luyi, Chen Xiao-Li, Qiu Huixian, Rong Xing, Wu Rongzhou, Chu Maoping, Huang Xiu-Feng
Zhejiang Provincial Clinical Research Center for Pediatric Precision Medicine, The Second Affiliated Hospital and Yuying Children's Hospital of Wenzhou Medical University, Wenzhou, Zhejiang, China.
Pediatrics Discipline Group, The Second Affiliated Hospital and Yuying Children's Hospital of Wenzhou Medical University, Wenzhou, Zhejiang, China.
iScience. 2025 Feb 11;28(3):112004. doi: 10.1016/j.isci.2025.112004. eCollection 2025 Mar 21.
Accurate prediction of intravenous immunoglobulin (IVIG) resistance is crucial for the effective treatment of Kawasaki disease(KD). This study aimed to develop a predictive model for IVIG resistance in patients with Kawasaki disease and to identify the key predictors. The training set underwent cross-validation, and models were constructed using six machine learning algorithms. Model performance was validated through cross-validation, test set evaluation, and two external validation sets evaluation. The model constructed using the random forest algorithm demonstrated the best overall performance among six models. The areas under the receiver operating characteristic curve (AUCs) for 5-fold cross-validation, internal validation, and external validations from Shaoxing and Quzhou were 0.711, 0.751, 0.827, and 0.735, respectively. According to the Shapley additive explanation (SHAP) method, C-reactive protein-to-albumin ratio, prognostic nutritional index, and sex were identified as the most important predictors. Our model demonstrates strong predictive capability for assessing IVIG resistance in Kawasaki disease patients.
准确预测静脉注射免疫球蛋白（IVIG）抵抗对于川崎病（KD）的有效治疗至关重要。本研究旨在建立川崎病患者IVIG抵抗的预测模型，并确定关键预测因素。训练集进行交叉验证，使用六种机器学习算法构建模型。通过交叉验证、测试集评估和两个外部验证集评估来验证模型性能。使用随机森林算法构建的模型在六个模型中表现出最佳的整体性能。五折交叉验证、内部验证以及来自绍兴和衢州的外部验证的受试者操作特征曲线下面积（AUC）分别为0.711、0.751、0.827和0.735。根据Shapley加性解释（SHAP）方法，C反应蛋白与白蛋白比值、预后营养指数和性别被确定为最重要的预测因素。我们的模型在评估川崎病患者IVIG抵抗方面具有很强的预测能力。
