DAT-SPECT subregional SBR Processing Pipeline Now Available on GitHub

Analyzing and Reusing Data
, , , ,
1

We would like to share a new MATLAB-based pipeline for processing DAT-SPECT data is now available through the **MJFF GitHub Imaging Repo (“DAT_SPECT_subregional_pipeline”). This pipeline computes striatal binding ratios (SBRs) in striatal subregions, allowing for more sensitive detection of subtle focal changes in DAT activity. This can be especially valuable in the preclinical (prodromal) phases of Parkinson’s disease, where pathology typically progresses along a dorso-posterior to ventro-anterior gradient, and early focal changes may be missed by whole-striatum analyses.

This was developed in parallel with the ongoing DAT-SPECT re-analysis by PPMI, providing a complementary tool for more detailed assessment of striatal subregional DAT changes over time. The resulting subregional SBR measures may serve as promising candidate progression markers, possibly capturing both co-occurring dopaminergic damage and compensatory processes, and offering richer insight than global SBR metrics alone.

This pipeline has been validated on two independent datasets, demonstrating robustness, and outputs are provided as tabular CSV /excel files that can be used for clustering analyses or as a complement to the Parkinson’s Insight Engine (PIE) pipeline recently presented by @vcatterson @cameronreidhamilton to the community.

We highly encourage interested members to test the pipeline to their own data and provide feedback on any issues or bugs so that we can continue to refine and improve the tool.

:backhand_index_pointing_right: Access the pipeline here: Imaging/DAT_SPECT_subregional_pipeline at main · MJFF-ResearchCommunity/Imaging · GitHub

@braah @bmarebwa @hirotaka @ehutchins @apinstein

5 Likes
October 2025 Data Community Newsletter
2

This is really fantastic work! Thanks for such a valuable contribution to the community.

I wondered if the output values from the analysis were uploaded somewhere online? Or does the user have to run the pipeline themself? I can understand that the output may not be part of the LONI archive, but it would be awesome if this were coming in the future! My thinking is that it would avoid spinning up the infrastructure to rerun the pipeline, if a user simply wanted to make use of the data without tweaking the pipeline themself. Do you have thoughts on how likely this may be in the future?

1 Like
3

Thanks for the feedback @vcatterson ! At the moment, the outputs aren’t hosted online, so interested users would need to run the pipeline themselves. The pipeline was developed and tested on own data, but it’s designed so that users collecting DAT in other studies can apply it as well. This way, they can use it not only for their own analyses but also as a tool to compare against the other available derived values from other studies.

3 Likes
4

This is great. Thanks Amgad! I will definitely ping my colleagues to check this.

3 Likes
5

Always exciting to see more analyses related to prodromal work. Thanks for the update - this sounds great!

3 Likes