Data Sheet 1_急性主动脉夹层术后再插管多中心队列研究的多种自动化机器学习预测模型

数据集获取（Figshare）

• 源站标题：Data Sheet 1_Multiple automated machine-learning prediction models for postoperative reintubation in patients with acute aortic dissection: a multicenter cohort study.docx
• DOI：10.3389/fmed.2025.1531094.s003
• 发布日期：2025-04-11
• 文件数：1 个
• 文件总体积：1.38 MB
• 源站页面：https://frontiersin.figshare.com/articles/dataset/Data_Sheet_1_Multiple_automated_machine-learning_prediction_models_for_postoperative_reintubation_in_patients_with_acute_aortic_dissection_a_multicenter_cohort_study_docx/28775753

命令行下载

curl -L -o "Data Sheet 1_Multiple automated machine-learning prediction models for postoperative reintubation in patients with acute aortic dissection: a multicen.docx" "https://ndownloader.figshare.com/files/53596664"

文件清单

• Data Sheet 1_Multiple automated machine-learning prediction models for postoperative reintubation in patients with acute aortic dissection: a multicen.docx（1.38 MB）

文件总体积 1.38 MB（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

IntroductionReintubation is an adverse postoperative complication in patients with Type A aortic dissection (AAD) that correlates to poor outcomes. This study aims to employ machine learning algorithms to establish a practical platform for the prediction of reintubation.
MethodsA total of 861 patients diagnosed with AAD and undergoing surgical procedures, 688 patients as training and testing cohort from a single center, and 173 patients as validation cohort from four centers were enrolled. The least absolute shrinkage and selection operator (LASSO) was used for screening risk variables associated with reintubation for subsequent model construction. Subsequently, seven machine-learning models were built. The model with the best discrimination and calibration performance was used to predict reintubation. Finally, the SHapley Additive exPlanation (SHAP) was employed to explain the prediction model.
ResultsReintubation was performed in 107 patients (12.43%). The LASSO analysis identified re-admission to the intensive care unit (ICU), continuous renal replacement therapy, length of stay in the ICU, and duration of invasive mechanical ventilation as significant risk factors for reintubation. The XGBoost model was selected as the final prediction model due to its better performance than other models, with the AUC, sensitivity, and specificity of 0.969, 0.8889, and 0.8611 in the testing cohort. SHAP values demonstrated the effects of individual features on the overall model. Finally, a web calculator was developed based on XGBoost model for the clinical use.
ConclusionWe have developed and validated a high-performing risk prediction model for postoperative reintubation in patients with AAD. It can provide valuable guidance to clinicians in predicting reintubation and in developing timely preventative measures.

许可

CC BY 4.0

引用

学术使用请引用 DOI 10.3389/fmed.2025.1531094.s003（source: https://frontiersin.figshare.com/articles/dataset/Data_Sheet_1_Multiple_automated_machine-learning_prediction_models_for_postoperative_reintubation_in_patients_with_acute_aortic_dissection_a_multicenter_cohort_study_docx/28775753）。