如何打开训练测试后生成的Synapse分割图

[KYC890]

作者你好！感谢你的出色工作！能否提供一份可以打开gt.nii.gz、img.nii.gz和pred.nii.gz文件的代码，并像您一样进行颜色区别；同时我需要一份预处理后的ACDC数据集，十分感谢！邮箱是：1769838453@qq.com

[Dootmaan]

Hi @KYC890 you can directly open the .nii.gz files with ITK-Snap, which can be downloaded here. For the preprocessed ACDC dataset, it can be found here.

[KYC890]

Thank you very much, it has helped me a lot.

[KYC890]

Hi Author! Thanks for your reply. The three files have been opened. The predicted segmentation map is with multiple slices. May I ask if the final segmentation map effect is composed of multiple slices or a single slice? However, looking at a single slice does not achieve the predicted effect and the predicted segmentation map is very different from the labeling under the slice at the same location. Is the differentiation of multiple organs with different colors manually labeled or automatically labeled?

…

------------------ 原始邮件 ------------------ 发件人: "Dootmaan/MT-UNet" ***@***.***>; 发送时间: 2024年11月25日(星期一) 下午3:56 ***@***.***>; 抄送: "success and ***@***.******@***.***>; 主题: Re: [Dootmaan/MT-UNet] 如何打开训练测试后生成的Synapse分割图 (Issue #47) Hi @KYC890 you can directly open the .nii.gz files with ITK-Snap, which can be downloaded here. For the preprocessed ACDC dataset, it can be found here. — Reply to this email directly, view it on GitHub, or unsubscribe. You are receiving this because you were mentioned.Message ID: ***@***.***>

[penghujin]

作者你好！感谢你的出色工作！能够提供一个可以打开gt.nii.gz、img.nii.gz和pred.nii.gz文件的代码，并且像你一样进行颜色区别；同时我需要一个复印后的ACDC数据集，十分感谢！邮箱是：1769838453@qq.com

Hello, I would like to ask why I opened the specified gz file with ITK-Snap. Why is the image gray without color differentiation? Does it need to be divided manually?

[Dootmaan]

Hi there. To overlap the segmentation mask on the original 3d image, you can open the original 3d image instead and then import the predicted segmentation mask afterwards.