使用H&E组织全切片图像的深度学习分类器识别临床心力衰竭患者

数据集获取（Figshare）

• 源站标题：A deep-learning classifier identifies patients with clinical heart failure using whole-slide images of H&E tissue
• DOI：10.1371/journal.pone.0192726
• 发布日期：2018-04-03
• 文件数：6 个
• 文件总体积：883.23 KB
• 源站页面：https://plos.figshare.com/articles/dataset/A_deep-learning_classifier_identifies_patients_with_clinical_heart_failure_using_whole-slide_images_of_H_E_tissue/6081491

命令行下载

curl -L -o "S1 Fig.pdf" "https://ndownloader.figshare.com/files/10966886"
curl -L -o "S2 Fig.pdf" "https://ndownloader.figshare.com/files/10966892"
curl -L -o "S3 Fig.pdf" "https://ndownloader.figshare.com/files/10966898"
curl -L -o "S1 Table.pdf" "https://ndownloader.figshare.com/files/10966904"
curl -L -o "S2 Table.pdf" "https://ndownloader.figshare.com/files/10966907"
curl -L -o "S3 Table.pdf" "https://ndownloader.figshare.com/files/10966910"

文件清单

• S1 Fig.pdf（287.42 KB）
• S2 Fig.pdf（103.69 KB）
• S3 Fig.pdf（148.06 KB）
• S1 Table.pdf（115.54 KB）
• S2 Table.pdf（116.46 KB）
• S3 Table.pdf（112.05 KB）

文件总体积 883.23 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

Over 26 million people worldwide suffer from heart failure annually. When the cause of heart failure cannot be identified, endomyocardial biopsy (EMB) represents the gold-standard for the evaluation of disease. However, manual EMB interpretation has high inter-rater variability. Deep convolutional neural networks (CNNs) have been successfully applied to detect cancer, diabetic retinopathy, and dermatologic lesions from images. In this study, we develop a CNN classifier to detect clinical heart failure from H&E stained whole-slide images from a total of 209 patients, 104 patients were used for training and the remaining 105 patients for independent testing. The CNN was able to identify patients with heart failure or severe pathology with a 99% sensitivity and 94% specificity on the test set, outperforming conventional feature-engineering approaches. Importantly, the CNN outperformed two expert pathologists by nearly 20%. Our results suggest that deep learning analytics of EMB can be used to predict cardiac outcome.

许可

CC BY 4.0

引用

学术使用请引用 DOI 10.1371/journal.pone.0192726（source: https://plos.figshare.com/articles/dataset/A_deep-learning_classifier_identifies_patients_with_clinical_heart_failure_using_whole-slide_images_of_H_E_tissue/6081491）。