which Deep Learning network is available for 3D object such as MRI scan with .nii format?

1 次查看（过去 30 天）

Rahul Sharma 2020-2-10

0
链接

此问题的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/504666-which-deep-learning-network-is-available-for-3d-object-such-as-mri-scan-with-nii-format

评论： Ahmadi Anis 2020-3-28

Want to load Nifti image which is a MRI scan into either CNN or with Pretrained network. But none is working in MATLAB. Shall i switch to python or wait for mathwork to get the update or any news regarding Nifti Images handling. As across the world researches are goin on Nifti images but very surprised MATLAB doestnot have proper tools for the analysis.

1 个评论
显示 -1更早的评论隐藏 -1更早的评论

Rik 2020-2-11

What input does your specific network require? You should be able to read the files and convert them to jpeg, dicom or whichever format your network requires.

请先登录，再进行评论。

请先登录，再回答此问题。

回答（2 个）

Walter Roberson 2020-2-10

1
链接

此回答的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/504666-which-deep-learning-network-is-available-for-3d-object-such-as-mri-scan-with-nii-format#answer_414799

Questions about future MATLAB products must be directed to your sales representative, who can arrange a Nondisclosure Agreement with you.

0 个评论
显示 -2更早的评论隐藏 -2更早的评论

请先登录，再进行评论。

Rahul Sharma 2020-2-11

0
链接

此回答的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/504666-which-deep-learning-network-is-available-for-3d-object-such-as-mri-scan-with-nii-format#answer_415001

I am about to use a .nii image format which is a 3D image of dimension 240*256*170.

i can use 3d cnn but the fact is i cannot augment the data for training also no pretained network such as resnet50, vgg16 is available for 3d image such as mentioned above. Just want to know whether i am not able to do or MATLAB doesnot handle .nii format for augmentation and scaling or any other pretrained network which works on 3D image of such a dimension

4 个评论
显示 2更早的评论隐藏 2更早的评论

Walter Roberson 2020-2-12

If you refuse to contact your sales representative then the answer is clear, that you need to abandon MATLAB because it does not do what you want right now.