4D datasets

[fotaras-github]

Hi Amir,

Can the tool also deal with 4D datasets (e.g. fMRI)? I find that matplotlib returns an error ('Third dimension must be 3 or 4') when the input is 4D.

Am I missing something here?

Thanks!

[satishev]

Looping in @Amir-Reza-Sadri

[Amir-Reza-Sadri]

Hello,

It should be working, can you share some of your data with me: ars329@case.edu

Thanks.

[fotaras-github]

Hello,

It should be working, can you share some of your data with me: ars329@case.edu

Thanks.

@Amir-Reza-Sadri Done!

[cammu00]

Hi Amir,
I got a same error for 4D dataset (fMRI, DTI). How to input my dataset or where could I change in codes?

Thanks!

[Amir-Reza-Sadri]

Please use the code QC4d.py instead of QC.py with the same settings for 4D data.
Here are the considerations for this version:

1. Assuming the volumetric 4D data is in the form of x × y × z × w, w is the fourth dimension (let's say the time). Right now, you can run the MRQy for a fixed time volume of the data. Depending on the time sequence of your data, you can specify the desired time at line #147 of the QC4d.py script (default value: time_dim=5).
2. If you want to compare other dimensions of your data in terms of quality metrics, feel free to replace the arguments in line #150.
   Some important things:
3. Make sure your data files have different names or id.
4. Make sure your volumetric inputs have the same size.
5. Make sure you have at least 6 input data.
   Let me know if this works.
   Thanks.

[fotaras-github]

Please use the code QC4d.py instead of QC.py with the same settings for 4D data. Here are the considerations for this version:

1. Assuming the volumetric 4D data is in the form of x × y × z × w, w is the fourth dimension (let's say the time). Right now, you can run the MRQy for a fixed time volume of the data. Depending on the time sequence of your data, you can specify the desired time at line #147 of the QC4d.py script (default value: time_dim=5).
2. If you want to compare other dimensions of your data in terms of quality metrics, feel free to replace the arguments in line #150.
   Some important things:
3. Make sure your data files have different names or id.
4. Make sure your volumetric inputs have the same size.
5. Make sure you have at least 6 input data.
   Let me know if this works.
   Thanks.

Thank you for the update Amir.

With the default time_dim=5, QC4d.py works on datasets with the following dimensions: [64x64x36x200]. But what is time_dim exactly? Time, in seconds/milliseconds, required for each volume? That's the part unclear to me.

[Amir-Reza-Sadri]

It is the desired time that you want to run the QC over. This way, the tool focuses/compares all scans in a single time-origin.