Introduction to building processing pipelines using scripts¶
What are batch processing scripts?¶
Batch processing scripts are scripts written in shell language (i.e. same as your Terminal) containing a sequence of SCT commands. They are meant to establish a processing pipeline, i.e. a chain of commands arranged so that the output of each element is the input of the next.
Why use scripts?¶
Scripts are beneficial for a number of reasons:
They help with standardization of processing across subjects.
They save time through automation of processing over large datasets.
They promote transparency for scientific results (e.g., you can publish your article along with the batch processing scripts).
In the long-term, they help with reproducibility by making it possible to re-run entire processing pipelines to regenerate results if needed (e.g., if 6 months after a manuscript submission, the reviewer asks you to change one parameter in the processing).
Example scripts¶
The SCT team itself regularly creates customized pipelines for its research efforts. You can see examples of different projects and studies here: https://github.com/sct-pipeline
SCT also provides a large example script called batch_processing.sh that is meant to introduce all of the main functionality within SCT. You can read more about this on the Batch Processing Example page.
For the purposes of this tutorial, however, we will focus on a smaller script called process_data.sh that is included alongside the data for this tutorial. The sample script contains a basic processing workflow, demonstrating how to compute CSA for T2 data, and MTR for the white matter regions of MT data.