预测中国东部川崎病住院患儿静脉注射免疫球蛋白耐药风险的列线图

数据集获取（Figshare）

• 源站标题：Nomogram to predict risk of resistance to intravenous immunoglobulin in children hospitalized with Kawasaki disease in Eastern China
• DOI：10.6084/m9.figshare.19096455.v1
• 发布日期：2022-01-31
• 文件数：4 个
• 文件总体积：780.53 KB
• 源站页面：https://tandf.figshare.com/articles/dataset/Nomogram_to_predict_risk_of_resistance_to_intravenous_immunoglobulin_in_children_hospitalized_with_Kawasaki_disease_in_Eastern_China/19096455

命令行下载

curl -L -o "Supplement 3.jpg" "https://ndownloader.figshare.com/files/33936051"
curl -L -o "supplement 4.jpg" "https://ndownloader.figshare.com/files/33936054"
curl -L -o "supplement1.docx" "https://ndownloader.figshare.com/files/33936057"
curl -L -o "supplement2.docx" "https://ndownloader.figshare.com/files/33936060"

文件清单

• Supplement 3.jpg（434.50 KB）
• supplement 4.jpg（309.34 KB）
• supplement1.docx（17.51 KB）
• supplement2.docx（19.18 KB）

文件总体积 780.53 KB（源站 API 实测）。

数据加载示例（图像类）

from PIL import Image
import glob, os

files = (glob.glob(os.path.join(path, "**", "*.png"), recursive=True)
       + glob.glob(os.path.join(path, "**", "*.jpg"), recursive=True)
       + glob.glob(os.path.join(path, "**", "*.tif"), recursive=True))
print("图像文件数:", len(files))
img = Image.open(files[0]); print("尺寸/模式:", img.size, img.mode)

# torchvision Dataset 方式:
# from torchvision import datasets
# ds = datasets.ImageFolder(path)  # 要求 子目录=类别

目录组织与标注格式以源站说明和下载后实际文件为准。

数据说明

We aimed to develop a nomogram to predict risk of resistance to intravenous immunoglobulin (IVIG) in children with Kawasaki disease in eastern China.
We retrospectively analysed the data of children with Kawasaki disease who received IVIG during hospitalisation at Soochow University Affiliated Children’s Hospital. IVIG resistance was defined as recrudescent or persistent fever ≥36 h after the end of the IVIG infusion. Baseline variables were analysed using least absolute shrinkage and selection operator (LASSO) to identify the predictors of IVIG resistance, which were then used to construct a predictive nomogram. Calibration curve and area under the receiver operating characteristic curve (AUC) were used to evaluate the performance of the model. The predictive nomogram was validated on test sets of external data and prospective data.
Between January 2015 and December 2020, 1293 Kawasaki disease patients were hospitalized in Soochow University Affiliated Children’s Hospital. Among them, 72 (5.57%) showed IVIG resistance. LASSO identified haemoglobin, percentage of neutrophils, C-reactive protein level, platelet count, serum albumin, serum sodium, serum alkaline phosphatase, coronary artery damage, and complete Kawasaki disease as risk factors for IVIG resistance. The nomogram constructed using these factors showed satisfactory discriminatory power (AUC, 0.75), and sensitivity (0.74) and specificity (0.64). In the external data and prospective data, the AUC was 0.66 and 0.83, respectively, the sensitivity was 0.86 and 1, respectively, and the specificity was 0.49 and 0.60, respectively.
The predictive nomogram constructed using nine factors associated with IVIG resistance in children with Kawasaki disease could be a useful tool for identifying patients likely to show IVIG resistance. This nomogram may help reduce the risk of coronary artery lesions.Key Messages
None of the IVIG resistance scoring systems has shown consistently good performance in previous studies. Tools to predict the risk of IVIG resistance in eastern China are lacking.
In our series, haemoglobin level, percentage of neutrophils, platelet count, coronary artery damage, incomplete Kawasaki dise

许可

CC BY 4.0

引用

学术使用请引用 DOI 10.6084/m9.figshare.19096455.v1（source: https://tandf.figshare.com/articles/dataset/Nomogram_to_predict_risk_of_resistance_to_intravenous_immunoglobulin_in_children_hospitalized_with_Kawasaki_disease_in_Eastern_China/19096455）。