License
This toolbox is shared under Apache 2.0 License.

Link to zip
A zip file containing all the toolbox utilities can be downloaded at [COMING].

Description
TbCAPs is a toolbox enabling the user to apply co-activation pattern (CAP) analysis to functional magnetic resonance imaging (fMRI) data. CAP analysis revolves around the following key steps: (1) suitable preprocessing following standard fMRI norms, with a final z-scoring step for each voxel-wise time course, (2) selection of fMRI volumes of interest through thresholding of a seed's activity time course, (3) k-means clustering of the retained volumes into CAPs, and (4) extraction of relevant metrics reflective of CAP temporal dynamics.
CAP analysis is a versatile analytical approach that may be leveraged both on resting-state and on task-based data. It belongs to the broad family of dynamic functional connectivity approaches, as it enables to analyze the fMRI data at the single-frame temporal resolution.
TbCAPs enables all of the aforementioned CAP analysis steps to be performed, and also includes possible alternative analytical pipelines: for example, one can select up to three seeds that can be jointly considered to select volumes of interest, and clustering may also be run on all volumes, without resorting to a seed thresholding step. In addition, the toolbox also includes a consensus clustering-based determination of the optimal number of clusters to use for an analysis.

Installation and updates
For installation, clone this repository using git:

git clone --depth=1 https://c4science.ch/source/CAP_Toolbox

To get the latest update of the toolbox, use:

git pull

Some potentially useful references
Liu, Xiao, and Jeff H. Duyn. "Time-varying functional network information extracted from brief instances of spontaneous brain activity." Proceedings of the National Academy of Sciences 110.11 (2013): 4392-4397.
This work was the first to introduce CAP analysis. The authors showed that several hubs from the Default Mode Network, one of the brain's canonical functional networks, would sometimes (de)activate individually, challenging a static understanding of brain function. They further demonstrated that at different moments in time, a given region of interest actually co-(de)activates with distinct sets of brain areas, and proposed CAP analysis as a way to capture it.

Liu, Xiao, et al. "Co-activation patterns in resting-state fMRI signals." Neuroimage 180 (2018): 485-494.
This is a recent review on CAP analysis, written by one of its inventors.

Chen, Jingyuan E., et al. "Introducing co-activation pattern metrics to quantify spontaneous brain network dynamics." Neuroimage 111 (2015): 476-488.
In this work, metrics reflective of CAP temporal dynamics are introduced and applied on fMRI data.

Liu, Xiao, Catie Chang, and Jeff H. Duyn. "Decomposition of spontaneous brain activity into distinct fMRI co-activation patterns." Frontiers in systems neuroscience 7 (2013): 101.
This is the article that proposed to conduct clustering of volumes not on retained data following threshold, but on all the data skipping this previous step.

Bolton, Thomas AW, et al. "TbCAPs: A toolbox for co-activation pattern analysis." Neuroimage 211 (2020): 116621.
TbCAPs was introduced in this article, alongside an example application for the study of attentional performance in healthy subjects.

Rey, Gwladys, et al. "Dynamics of amygdala connectivity in bipolar disorders: a longitudinal study across mood states." Neuropsychopharmacology 46.9 (2021): 1693-1701.
An example application of the toolbox on fMRI data from bipolar patients longitudinally scanned across various mood stages.

Tuleasca, Constantin, et al. "Normalization of aberrant pretherapeutic dynamic functional connectivity of extrastriate visual system in patients who underwent thalamotomy with stereotactic radiosurgery for essential tremor: a resting-state functional MRI study." Journal of neurosurgery 132.6 (2019): 1792-1801.
Another example application of the toolbox on fMRI data from patients with essential tremor, tracking changes in functional brain dynamics after stereotactic radiosurgical thalamotomy.

Iraji, A., et al. "Moving beyond the ‘CAP’of the Iceberg: Intrinsic connectivity networks in fMRI are continuously engaging and overlapping." NeuroImage 251 (2022): 119013.
An article reminding us that while CAPs are useful to understand how the brain works, they do not encompass all the meaningful information.

Li, Meiling, et al. "Co-activation patterns across multiple tasks reveal robust anti-correlated functional networks." Neuroimage 227 (2021): 117680.
An example of work focusing on task-based, instead of resting-state, fMRI data.

Amico, Enrico, et al. "Posterior cingulate cortex-related co-activation patterns: a resting state FMRI study in propofol-induced loss of consciousness." PloS one 9.6 (2014): e100012.
A study in which spatial, as opposed to temporal, CAP features were analyzed across states of consciousness.

Contact
If you have any questions, error reports, or if you simply want to chat about CAP analysis and how it may be applied to your own work, feel free to contact me at thomas.bolton@epfl.ch!