社区获得性肺炎住院患者一年死亡率预测数据集

数据集获取（Figshare）

• 源站标题：Predicting 1-year mortality after hospitalization for community-acquired pneumonia
• DOI：10.1371/journal.pone.0192750
• 发布日期：2018-02-14
• 文件数：2 个
• 文件总体积：90.33 KB
• 源站页面：https://plos.figshare.com/articles/dataset/Predicting_1-year_mortality_after_hospitalization_for_community-acquired_pneumonia/5889574

命令行下载

curl -L -o "S1 Table.docx" "https://ndownloader.figshare.com/files/10488124"
curl -L -o "S2 Table.docx" "https://ndownloader.figshare.com/files/10488127"

文件清单

• S1 Table.docx（19.29 KB）
• S2 Table.docx（71.04 KB）

文件总体积 90.33 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

Background
Community-acquired pneumonia (CAP) is a major public health problem with high short- and long-term mortality. The main aim of this study was to develop and validate a specific prognostic index for one-year mortality in patients admitted for CAP.
Methods
This was an observational, prospective study of adults aged ≥18 years admitted to Galdakao-Usansolo Hospital (Bizkaia, Spain) from January 2001 to July 2009 with a diagnosis of CAP surviving the first 15 days. The entire cohort was divided into two parts, in order to develop a one-year mortality predictive model in the derivation cohort, before validation using the second cohort.
Results
A total of 2351 patients were included and divided into a derivation and a validation cohort. After deaths within 15 days were excluded, one-year mortality was 10.63%. A predictive model was created in order to predict one-year mortality, with a weighted score that included: aged over 80 years (4 points), congestive heart failure (2 points), dementia (6 points), respiratory rate ≥30 breaths per minute (2 points) and blood urea nitrogen >30 mg/dL (3 points) as predictors of higher risk with C-index of 0.76. This new model showed better predictive ability than current risk scores, PSI, CURB65 and SCAP with C-index of 0.73, 0.69 and 0.70, respectively.
Conclusions
An easy-to-use score, called the one-year CAPSI, may be useful for identifying patients with a high probability of dying after an episode of CAP.

许可

CC BY 4.0

引用

学术使用请引用 DOI 10.1371/journal.pone.0192750（source: https://plos.figshare.com/articles/dataset/Predicting_1-year_mortality_after_hospitalization_for_community-acquired_pneumonia/5889574）。