Evaluation of a New Paracingulate Sulcus Identification and Measurement Protocol

The folding pattern of the anterior cingulate cortex is variable among individuals. The paracingulate sulcus (PCS), in particular, can be difficult to identify because of this variability. In this paper, we assessed the benefits of a new protocol to identify the PCS and measure its length using 3D reconstructions of the brain obtained through the BrainVISA software. Both the previous state-of-the-art protocol and the new protocol were applied to identify the PCS in subjects from a UK cohort (n = 50), with the length of the PCS computed automatically by BrainVISA after PCS identification. We assessed inter-rater reliability for PCS length under both protocols and the correlation of PCS length values output by the two protocols. Inspection of the results indicated an advantage of the new protocol as there are 8 out of 100 hemispheres in which the old protocol detects a PCS due to the mislabelling of the intra-limbic sulcus as the cingulate sulcus and the new protocol does not. This advantage is conferred by the incorporation of identification of the intra-limbic sulcus (when present) in the new protocol. For instances where the protocols agreed on the presence of a PCS, the new protocol for PCS length obtained intra-class correlations of 0.85 and 0.86 on respectively untrained and trained experimenters, compared to 0.81 when trained experimenters used the previous protocol. The PCS length correlation between both protocols was 0.73 for the entire sample and 0.85 after excluding instances where the protocols disagreed on the presence of a PCS. These findings suggest that taking into account the intra-limbic sulcus and taking advantage of 3D sulcal visualisation may help improve reliable PCS identification, and that the new protocol is a reliable tool that is likely to prove useful in research into cingulate and paracingulate cortical folding.