NeuroAI / Human cortical atlas explorer ===================================== This is a modified visualisation derivative of published reference anatomy, not a patient scan. No measured fMRI signals or clinical outcomes are included. Network colours identify atlas membership, not activation or treatment effects. Surface ------- FreeSurfer fsaverage5 pial surfaces, distributed with Nilearn 0.12.1: https://surfer.nmr.mgh.harvard.edu/fswiki/FsAverage https://nilearn.github.io/stable/modules/generated/nilearn.datasets.fetch_surf_fsaverage.html Fischl et al. (1999), High-resolution intersubject averaging and a coordinate system for the cortical surface. Human Brain Mapping 8(4):272-284. All or portions of this licensed product (such portions are the "Software") have been obtained under license from The General Hospital Corporation and are subject to the terms and conditions in FREESURFER-LICENSE.txt, included in this directory. Nilearn's distribution licence is also included as NILEARN-LICENSE.txt; it does not replace the upstream FreeSurfer terms. Parcellation ------------ Schaefer et al. (2018), Local-Global Parcellation of the Human Cerebral Cortex from Intrinsic Functional Connectivity MRI. Cerebral Cortex 28(9):3095-3114. https://doi.org/10.1093/cercor/bhx179 100 parcels / 7 networks, FreeSurfer5.3 fsaverage5 annotation files: https://github.com/ThomasYeoLab/CBIG/tree/master/stable_projects/brain_parcellation/Schaefer2018_LocalGlobal Copyright (c) 2016 Computational Brain Imaging Group (CBIG). See the included CBIG-LICENSE.txt for the complete MIT licence terms. The seven-network organisation follows Yeo et al. (2011), The organization of the human cerebral cortex estimated by intrinsic functional connectivity. https://doi.org/10.1152/jn.00338.2011 Derivation and interpretation ----------------------------- The original 20,484 vertices and 40,960 triangles are retained. Positions are centred, uniformly scaled, reoriented from RAS to display coordinates (X right, Y superior, negative Z anterior), and rounded to 5 decimal places. No hemisphere-specific recentering or label interpolation is used. Medial-wall/background vertices remain unassigned. Region markers are actual surface vertices nearest to each parcel centroid. The "research" transition rearranges these markers by network, preserving region IDs and colours. These are feature-space positions, not anatomical coordinates. Dots are parcels, not neurons; no edges or activation values are inferred from their proximity or movement. The figure 4,950 is 100 * 99 / 2: the number of possible undirected region pairs excluding the diagonal. It is not a number of statistically significant connections. Treatment-response content describes a proposed study design. The existing MaleCNS fruit-fly results are a separate evidence section and must not be interpreted as human anatomy or patient-level validation. Exact source filenames and SHA-256 checksums are in atlas.json. The reproducible conversion script is scripts/prepare-brain-atlas.py in the site source. The SVG poster is projected from the same released mesh. This display is for research communication, not clinical decision-making.