智能手机耳镜临床实践评估数据集

数据集获取（Figshare）

• 源站标题：Smartphone-enabled otoscopy: method evaluation in clinical practice
• DOI：10.6084/m9.figshare.22121249.v1
• 发布日期：2023-02-18
• 文件数：12 个
• 文件总体积：2.39 MB
• 源站页面：https://scielo.figshare.com/articles/dataset/Smartphone-enabled_otoscopy_method_evaluation_in_clinical_practice/22121249

命令行下载

curl -L -o "ef89900ba3711ec56d06573fd7c0b2acad56622f.tif" "https://ndownloader.figshare.com/files/39312503"
curl -L -o "30749220960a1d33f9a9a1e60959749eb1381b72.png" "https://ndownloader.figshare.com/files/39312506"
curl -L -o "e28e1875131eee175dd42fee2fc197c366d217bf.jpg" "https://ndownloader.figshare.com/files/39312509"
curl -L -o "b88f6b9bc0ce53d26acd437e7ab2eb07e5fbeb18.tif" "https://ndownloader.figshare.com/files/39312512"
curl -L -o "df4c0efb0b213661f2d061b9cdeefbb2c51febe8.png" "https://ndownloader.figshare.com/files/39312515"
curl -L -o "747bcef8d0bc7a7d1724ed6c564729bff6ee1b4d.jpg" "https://ndownloader.figshare.com/files/39312518"
curl -L -o "c15338c90a588f9a0f75505003b1ab474591c478.tif" "https://ndownloader.figshare.com/files/39312521"
curl -L -o "d22064ba4f6a66f221aa22af4944654670bbbc96.png" "https://ndownloader.figshare.com/files/39312524"

文件清单

• ef89900ba3711ec56d06573fd7c0b2acad56622f.tif（33.81 KB）
• 30749220960a1d33f9a9a1e60959749eb1381b72.png（239.78 KB）
• e28e1875131eee175dd42fee2fc197c366d217bf.jpg（4.82 KB）
• b88f6b9bc0ce53d26acd437e7ab2eb07e5fbeb18.tif（86.76 KB）
• df4c0efb0b213661f2d061b9cdeefbb2c51febe8.png（60.19 KB）
• 747bcef8d0bc7a7d1724ed6c564729bff6ee1b4d.jpg（3.49 KB）
• c15338c90a588f9a0f75505003b1ab474591c478.tif（784.97 KB）
• d22064ba4f6a66f221aa22af4944654670bbbc96.png（65.52 KB）
• 37108d4874ac598cdca765581ed7175594379f46.jpg（5.89 KB）
• f3fdcf5627925377d54319001b59cfd1bdfc2c2e.tif（970.78 KB）
• 67b82e1bf938d9229ebb27745718021f596c44f5.png（126.12 KB）
• dae2f99c296fa5f0f616e0e0f29bdd46af661b85.jpg（7.64 KB）

文件总体积 2.39 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

Abstract Objective: To assess the diagnostic agreement between smartphone-enabled otoscopy and rigid otoendoscopy in tympanic membrane and middle ear diseases. Methods: A cross-sectional study was carried out to analyze otoscopies in patients seen at a general otorhinolaryngology (ORL) outpatient clinic, from June to December 2019. Eighty-three images of patients obtained from otoscopies performed through a smartphone device and a rigid endoscope were included, recorded, and stored for further analysis. The images were first analyzed by an experienced otologist, who assigned his diagnostic impression (defined as the gold standard) on each of the images. After this analysis, the images were displayed to a group of secondary raters (an experienced otorhinolaryngologist, a second-year resident in ORL, and a general practitioner). A questionnaire was applied related to each image. Results: There was high agreement between the smartphone device and the otoendoscopy images for all professionals, with a Kappa coefficient of 0.97 (p < 0.001). The smartphone device showed a diagnostic sensitivity of 81.1% and a specificity of 71.1%. As for the otoendoscopy, it showed a sensitivity of 84.7% and a specificity of 72.4%. The image classification as “2 = Good” was the most frequent one, with 34.9% for otoendoscopy and 31.6% for the smartphone device. Conclusion: There was a high diagnostic agreement between smartphone device-guided otoscopy and the rigid otoendoscopy, demonstrating the feasibility of using this device in clinical practice. Level of Evidence: 3

许可

CC BY 4.0

引用

学术使用请引用 DOI 10.6084/m9.figshare.22121249.v1（source: https://scielo.figshare.com/articles/dataset/Smartphone-enabled_otoscopy_method_evaluation_in_clinical_practice/22121249）。