单基因病无创产前诊断的开发

数据集获取

源站：https://clinicaltrials.gov/study/NCT06147414

标准获取流程：

1. 访问源站链接，注册/登录账号；
2. 进入数据集详情页，按页面指引申请权限或直接下载；
3. 部分平台提供 SDK，例如：
   • OpenDataLab: pip install openxlab 后 openxlab.dataset.get(...)
   • ModelScope: pip install modelscope 后 modelscope download
   • 天池: 登录后页面提供直链下载
     以源站「数据下载/使用文档」页面为准。

数据说明

Conditions: Invasive PreNatal Diagnosis in a Context of Family History of Single-gene Disorders, Including, Sickle Cell Disease, Cystic Fibrosis, Fragile X Syndrome, Proximal Spinal Muscular Atrophy, Myotonic Dystrophy, Muscular Dystrophy, Duchenne, Muscular Dystrophy, Becker. Since the identification of cffDNA in maternal plasma in 1997, there have been rapid developments in exploiting its presence for prenatal diagnosis and screening. The first proof of principle studies using cfDNA from maternal plasma to detect fetal aneuploidy were published in 2008 following which there was rapid commercialization. Nowadays, non-invasive prenatal testing (NIPT) for aneuploidies is widely used across the world as a screening test for the most frequent fetal trisomies. Unlike non-invasive prenatal testing, where a positive result requires confirmation following an invasive test, non-invasive prenatal diagnosis (NIPD) offers the advantage of a definitive diagnosis without an invasive procedure - and its associated miscarriage risk - because confined placental mosaicism does not occur with NIPD for single-gene disorders (SGD). NIPD can be offered earlier in pregnancy than invasive testing, from 7 weeks of gestation. This can reduce parental anxiety and allows more time for decision-making and planning. Indeed, there are substantial challenges to overcome for SGD-NIPD. i/ Circulating cffDNA, which is released from the placenta from about 4 weeks gestation, makes up only 5%-20% of total circulating cfDNA in maternal plasma. This percentage increases with gestation and is influenced by factors such as maternal weight, smoking, and pregnancy complications such as preeclampsia. Consequently, optimized techniques and highly sensitive detection approaches are required to detect variants in the fetal DNA. ii/ fetal fraction must be calculated to confirm that there are sufficient levels of cffDNA present and to avoid false negative results. iii/ Another issue is the short fragment length of cffDNA, which makes detection of triplet repeats and large deletions or duplications challenging. Besides fetal sex determination and fetal Rhesus D status, principal investigator’s team was the fir

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。