sct_deepseg_gm

Spinal Cord Gray Matter (GM) Segmentation using deep dilated convolutions. The contrast of the input image must be similar to a T2*-weighted image: WM dark, GM bright and CSF bright. Reference: Perone CS, Calabrese E, Cohen-Adad J. Spinal cord gray matter segmentation using deep dilated convolutions. Sci Rep 2018;8(1):5966.

usage: sct_deepseg_gm -i <file> [-o <file>] [-m {large,challenge}]
                      [-thr <float>] [-t] [-qc <str>] [-qc-dataset <str>]
                      [-qc-subject <str>] [-h] [-v <int>]

MANDATORY ARGUMENTS

-i

Image filename to segment (3D volume). Example: t2s.nii.gz.

OPTIONAL ARGUMENTS

-o

Output segmentation file name. Example: sc_gm_seg.nii.gz

-m

Possible choices: large, challenge

Model to use (large or challenge). The model ‘large’ will be slower but will yield better results. The model ‘challenge’ was built using data from the following challenge: goo.gl/h4AVar.

Default: 'large'

-thr

Threshold to apply in the segmentation predictions, use 0 (zero) to disable it.

Default: 0.999

-t

Enable TTA (test-time augmentation). Better results, but takes more time and provides non-deterministic results.

MISC ARGUMENTS

-qc

The path where the quality control generated content will be saved.

-qc-dataset

If provided, this string will be mentioned in the QC report as the dataset the process was run on

-qc-subject

If provided, this string will be mentioned in the QC report as the subject the process was run on

-v

Possible choices: 0, 1, 2

Verbosity. 0: Display only errors/warnings, 1: Errors/warnings + info messages, 2: Debug mode.

Default: 1