conn connectivity (conn) toolbox based on matlab 2022b  (MathWorks Inc)

Journal: Pain

Article Title: Identification of group differences in predictive anticipatory biasing of pain during uncertainty: preparing for the worst but hoping for the best

doi: 10.1097/j.pain.0000000000003207

Figure Lengend Snippet: Illustration of applied methods. (A) Data collection: pain anticipation paradigm; high pain (red, N = 7), low pain (green, N = 7), and uncertain pain (yellow, N = 14) visual cues, followed by pain stimulation. (B) Model verification: fMRI preprocessing with CONN toolbox ( www.nitrc.org/projects/conn , RRID:SCR_009550) and task-based regression (including least squares sum model) completed in AFNI. Activation maps extracted in 26 a priori–chosen ROIs depicted in glass brain on the right side only for each high and low pain anticipation event (see supplemental digital content 3, http://links.lww.com/PAIN/C20 for further details). Elastic net regression is used to train and test classifier to separate low and high pain anticipation neural patterns (c.f. Methods for more details) (C) Uncertainty prediction: Each uncertain anticipation trial is compared with the certain activation maps and a probabilistic prediction is determined by LASSO. Predictions ≥0.5 are classified as “high,” and predictions <0.5 as “low.” (D) Group classification: Predictions across all 14 uncertain trials for each subject are provided to mixAK cluster analysis in R, and each subject is clustered based on individual anticipation profile.

Article Snippet: All fMRI data were preprocessed using a MatLab-based functional connectivity toolbox, CONN, to denoise and align the images for analysis (Fig. B).

Techniques: Activation Assay