Bloodstream

Add a comment

A PET BIDS blood-processing tool

Bloodstream is a simplified and automated pipeline designed for PET studies to process blood data [1,2] in BIDS (Brain Imaging Data Structure) format [3]. The bloodstream package is based on functions found in kinfitr, but strings them together into a blood processing pipeline, producing a parameterized report as well as processed blood derivatives.

Usage Features

Bloodstream expects the user to specify as input a studypath and a configpath.

  • The studypath is the location of the BIDS data, e.g. ../ds004230 (relative or full paths are allowed).
  • The configpath is the path to the bloodstream configuration file, which specifies the modelling choices which you will make as a user. To create a configuration file, go to the bloodstream configuration web app, fill in the fields as required, and download the JSON configuration file. The configpath specifies the location of the downloaded config file, e.g. ../config_test_analysis.json. If left blank, then the blood data will simply be combined using linear interpolation.

Bloodstream will then generate the following outputs:

  • A report showing all the code and functions used, as well as plots before and after modelling.

… and for all individual PET measurements, the following

  • Tabular tsv output (*_inputfunction.tsv) containing the estimated interpolated data which can be used for modelling.
  • JSON sidecar accompanying the tabular tsv data (*_inputfunction.tsv).
  • Model configuration JSON files, containing the models used and the arterial input function (AIF) fit parameters if applicable (*_config.json).

GitHub Page

Installation

Docker Container

Tutorial Video

Citation recommendation: Until there is a preprint or publication about bloodstream, the developers ask authors to just specify that “bloodstream was used for blood analysis, which is a blood processing pipeline built around kinfitr [1,2]”.

[1] Matheson, G. J. (2019). Kinfitr: Reproducible PET Pharmacokinetic Modelling in R. bioRxiv: 755751 DOI: 10.1101/755751

[2] Tjerkaski, J., Cervenka, S., Farde, L., & Matheson, G. J. (2020). Kinfitr – an open source tool for reproducible PET modelling: Validation and evaluation of test-retest reliability. EJNMMI Res 10, 77 (2020). DOI: 10.1186/s13550-020-00664-8

[3] Gorgolewski, K.J., Auer, T., Calhoun, V.D., Craddock, R.C., Das, S., Duff, E.P., Flandin, G., Ghosh, S.S., Glatard, T., Halchenko, Y.O. and Handwerker, D.A. (2016). The brain imaging data structure, a format for organizing and describing outputs of neuroimaging experiments. Scientific data 3(1) p.1-9 DOI:10.1038/sdata.2016.44

Leave a comment