volbrain/assemblynet

AssemblyNet: 3D Whole Brain MRI segmentation pipeline

assemblynet's Issues

Atlas for Visualization of results

After using volBrain to generate results, i would like to display areas for a single subject that were higher than average for GMV.
Which 3D Brain Atlas should I download for visualization purposes (i.e. what is the name of the atlas used by volBrain, and where can I download it from).
Roger

AssemblyNet vs vol2Brain

Which of the two has better segmentation performance? Have you calculated Dice score for both of them compared to a manually segmented ground truth brain?

NVIDIA cuda base image version

The instructions to test the NVIDIA container toolkit include 'nvidia/cuda:11.0-base', however, this image was removed, it is now nvidia/cuda:11.0.3-base (see, NVIDIA/nvidia-docker#1735)

Ranges with Scores

Is there a way to get the normative range for a person (shown in the PDF) as well as the actual volume values in the .csv file output?
This would be useful for identifying frequencies of people showing abnormally high or low volumes for their age/sex.

Docker image is excessively large

Seems the Dockerfile for this tool is not public, nor is the code, but whatever process is being done to generate the image, it is (probably) not following best practices for docker:

Statistics

This may be more of a statistics question, but what is the proper way to test they hypothesis that volumetry is different in a patient group than your control data?
There are a number of possibilities, including a chi-square to calculate frequency of deviants for a specific ROI, or submitting %DevianceFromTheMean value of an ROI to a t-test (comparing to the expected difference of 0).
Do you have any suggestions for how best to perform such a test.
In particular, I wish to compare M1 volume in athletes (N = 50) to the mean of the matched population (found in your bounds files).
Thank you!
Roger

-age-sex-csv

Our use of the parameter -age-sex-csv has been always failing and I don‘t know if there is something wrong with the csv file：
/T1/S0004.nii,66,Male
/T1/S0005.nii,58,Female
/T1/S0006.nii,69,Female
/T1/S0007.nii,75,Female
/T1/S0010.nii,69,Female
/T1/S0012.nii,64,Male
/T1/S0016.nii,79,Female
/T1/S0017.nii,60,Female

Age-sex csv file

The age-sex csv file only works without the '#' before the column names.

Volbrain other pipelines

Hello!

I'm not sure if this is the correct place to ask, but I was wondering if there are plans to release Docker images for other Volbrain pipelines in the near future?

Thank you!

problems for docker in ubuntu 20.04 on windows

Dear professor,

Thanks for your amazing tutorial! I was stuck in the following problems.

First, I've Install Windows Subsystem for Linux (WSL) 2 and also download the docker-desktop

Also, I've entered the command line ( sudo docker pull volbrain/assemblynet:1.0.0 ) in Ubuntu 20.04 on windows.

But, the problems I encountered is that I was confused about the command line (sudo docker run --rm -v /absolute/path/to/images:/data volbrain/assemblynet:1.0.0 /data/image.nii). What is the absolute path meaning?

If I have the nii file located in the path of my notebook such that "C:\Users\Downloads\mybrainimg.nii". What should I input my command line to get the AssemblyNet results. Thank you!

Sincerely,
Lonnie

Actual entry

Would it be possible to put the line of code to process one image (with actual path names and image names).
I am getting an error where it says arguments should be one filename or directory, so I am not sure the syntax to process all files in one directory or just one file.
Thanks!
Roger

Facing error: Error using load_nii_hdr (line 39)

Hello developers! I am currently facing issue when run any T1.nii image in the following form:

THE ERROR:
num t1_filenames= 1
./volBrain_v10_preprocessing_exe.sh /tmp/t1_0 /tmp
Error using load_nii_hdr (line 39)
Cannot find file "/tmp/t1_0.hdr".
Error in load_untouch_nii (line 154)
Error in read_nifti (line 9)
Error in sv_segment_preprocessing (line 88)

Can anyone please share their views/comments on how to resolve the issue? I am running on Ubuntu 20.04.6 with 200 GB RAM and 48 GB Threadripper Pro.