IST-3 CT Head Scans
收藏资源简介:
IST-3 CT头部扫描数据集由爱丁堡大学临床脑科学中心创建,包含10,659个CT系列,用于研究颅内动脉钙化的分割。数据集来源于第三届国际中风试验(IST-3),涉及3035名急性缺血性中风患者的非增强CT扫描。数据集创建过程中,通过与模板配准和质量控制,确保了数据的有效性和准确性。该数据集主要用于支持深度学习方法在中风风险评估中的应用,特别是在颅内动脉钙化的自动量化方面。
The IST-3 CT Head Scan Dataset was created by the Centre for Clinical Brain Sciences at the University of Edinburgh, comprising 10,659 CT series for research on the segmentation of intracranial arterial calcifications. The dataset is derived from the Third International Stroke Trial (IST-3), covering non-contrast CT scans of 3,035 patients with acute ischemic stroke. During the dataset creation process, data validity and accuracy were ensured through template registration and quality control. This dataset is primarily intended to support the application of deep learning methods in stroke risk assessment, especially for the automatic quantification of intracranial arterial calcifications.
CT扫描叠加工具
安装
-
克隆仓库: bash git clone https://github.com/bjin96/superimposition-tool.git
-
安装依赖: bash pip install -r requirements.txt
-
确保安装Qt5,按照Qt文档中的说明进行安装。
运行
在config.json中设置变量:
| 变量 | 描述 |
|---|---|
| batch_size | 一次叠加的扫描数量。 |
| template_path | 所有扫描共注册到的模板路径(NIfTI格式,.nii.gz)。 |
| blacklist_path | 存储黑名单路径的JSON文件路径。如果文件不存在,将会创建。 |
| input_file_list_path | 包含待分析CT扫描路径的JSON文件路径。 |
黑名单文件格式: json [ { "file": "/path/to/the/blacklisted/file1.nii.gz", "reason": "First comment" }, { "file": "/path/to/the/blacklisted/file2.nii.gz", "reason": "Second comment" }, ... ]
输入文件列表格式: json [ "/path/to/the/blacklisted/file1.nii.gz", "/path/to/the/blacklisted/file2.nii.gz", ... ]
启动工具: bash python run.py

- 1Pre-processing and quality control of large clinical CT head datasets for intracranial arterial calcification segmentation爱丁堡大学临床脑科学中心 · 2024年



