resector

实现了一个TorchIO转换，用于从T1加权脑MRI中模拟切除腔，以及相应的测地信息流（GIF）脑分区（版本3.0）。

MICCAI 2020上的相关演讲可在YouTube上找到

致谢

如果您在研究中使用了这个库，请引用以下出版物

Pérez-García, F., Dorent, R., Rizzi, M., Cardinale, F., Frazzini, V., Navarro, V., Essert, C., Ollivier, I., Vercauteren, T., Sparks, R., Duncan, J.S., Ourselin, S.: 一种用于模拟术后脑腔分割的自我监督学习策略. 国际计算机辅助放射学杂志 – IJCARS (Jun 2021)

Pérez-García, F., Rodionov, R., Alim-Marvasti, A., Sparks, R., Duncan, J.S., Ourselin, S.: 使用自我监督和半监督学习模拟脑部切除术进行腔分割. 在: 医学图像计算和计算机辅助介入 – MICCAI 2020. 第115-125页. 计算机科学讲座笔记，Springer国际出版社，Cham (2020)

Bibtex

@inproceedings{perez-garcia_simulation_2020,
    address = {Cham},
    series = {Lecture {Notes} in {Computer} {Science}},
    title = {Simulation of {Brain} {Resection} for {Cavity} {Segmentation} {Using} {Self}-supervised and {Semi}-supervised {Learning}},
    isbn = {978-3-030-59716-0},
    doi = {10.1007/978-3-030-59716-0\_12},
    language = {en},
    booktitle = {Medical {Image} {Computing} and {Computer} {Assisted} {Intervention} {\textendash} {MICCAI} 2020},
    publisher = {Springer International Publishing},
    author = {P{\'e}rez-Garc{\'i}a, Fernando and Rodionov, Roman and Alim-Marvasti, Ali and Sparks, Rachel and Duncan, John S. and Ourselin, S{\'e}bastien},
    year = {2020},
    keywords = {Segmentation, Self-supervised learning, Neurosurgery},
    pages = {115--125},
}

@article{perez-garcia_self-supervised_2021,
    title = {A self-supervised learning strategy for postoperative brain cavity segmentation simulating resections},
    issn = {1861-6429},
    url = {https://doi.org/10.1007/s11548-021-02420-2},
    doi = {10.1007/s11548-021-02420-2},
    language = {en},
    urldate = {2021-06-14},
    journal = {International Journal of Computer Assisted Radiology and Surgery},
    author = {P{\'e}rez-Garc{\'i}a, Fernando and Dorent, Reuben and Rizzi, Michele and Cardinale, Francesco and Frazzini, Valerio and Navarro, Vincent and Essert, Caroline and Ollivier, Ir{\`e}ne and Vercauteren, Tom and Sparks, Rachel and Duncan, John S. and Ourselin, S{\'e}bastien},
    month = jun,
    year = {2021},
    file = {Springer Full Text PDF:/Users/fernando/Zotero/storage/SM9WHUB7/P{\'e}rez-Garc{\'i}a et al. - 2021 - A self-supervised learning strategy for postoperat.pdf:application/pdf},
}

安装

建议使用conda

conda create --name resenv python=3.8 --yes && conda activate resenv
pip install light-the-torch
ltt install torch
pip install git+https://github.com/fepegar/resector
resect --help

使用

resect t1.nii.gz gif_parcellation.nii.gz t1_resected.nii.gz t1_resection_label.nii.gz

使用与resector一起安装的TorchIO可以下载一些样本图像

T1=`python -c "import torchio as tio; print(tio.datasets.FPG().t1.path)"`
GIF=`python -c "import torchio as tio; print(tio.datasets.FPG().seg.path)"`
resect $T1 $GIF t1_resected.nii.gz t1_resection_label.nii.gz

运行resect --help获取更多选项。

资助

这项工作得到了工程与物理科学研究委员会（EPSRC）和惠康基金会的资助。

此外，还得到了EPSRC资助的伦敦大学学院智能集成成像与健康（i4health）博士研究生培训中心以及惠康/ EPSRC介入与外科科学中心（WEISS）的支持。