DCM metric normalization

This repository contains the analysis scripts related to morphometric measures normalization.

The MRI data of healthy individuals from spine-generic project were used to construct a database of quantitative metrics in the PAM50 reference space; for details, see PAM50-normalized-metrics.

Dependencies

• Python 3 and packages listed in requirements.txt
• SCT

Usage

Each dataset has its own processing script containing preprocessing, spinal cord segmentation and vertebral labeling. The analysis pipeline slightly differs between datasets based on available images.

Scripts for processing of individual datasets are located under the scripts folder and can be run using the sct_run_batch wrapper script (part of the Spinal Cord Toolbox). For example:

sct_run_batch -path-data PATH_TO_DATA -path-output PATH_TO_OUTPUT -script scripts/process_data_spine-generic.sh

or using the JSON configuration file:

sct_run_batch -c etc/config_process_data.json

Note

The etc/config_process_data.json configuration file is just a template --> you need to edit it to match your data.

Note

There is an exclude.yml file associated with each dataset under the etc folder. This file lists subjects/images which should be excluded from analyses. You can exclude those subjects using the -exclude-list flag (when running sct_run_batch from CLI) or the exclude_list key (when running sct_run_batch using JSON configuration file).

---

Tip

For manual corrections of spinal cord segmentations, please refer to this repository: manual-correction and its wiki for examples.