A 200-micron structural connectivity atlas of the post-mortem human brain using HPC global tractography on the Chenonceau dataset

Motivation: Questions about the complete structural connectivity of the human brain remain unresolved. There is a need for high-resolution mapping to unveil finer-scale connections that conventional techniques may miss. Goal(s): This study aims to combine a high-resolution post-mortem human brain dataset with an advanced global tractography algorithm to achieve unprecedented levels of detail. Approach: We used ExaTract, a massively parallelized global tractography approach dedicated to supercomputer architecture, and fiber clustering methods to process the Chenonceau dataset at the mesoscale. Results: This work produced the first mesoscale atlas of human brain connectivity using global tractography, mapping structural connections at resolutions finer than previously achieved. Impact: This deep phenotyping approach represents a significant advance in the human brain structural connectivity mapping. Including subcortical and intracortical connections at very high level of details, it provides a valuable reference dataset for comparison with conventional MRI tractography in vivo.