2026-10-12T13:48:18Z https://www.scidb.cn/oai

10.57760/sciencedb.45751 2026-08-12T09:37:52Z

2026-08-12 A probabilistic atlas of human hypothalamic pathways based on connectome imaging data doi:10.57760/sciencedb.45751 en This dataset contains a probabilistic atlas of 22 hypothalamus-centered fiber bundles (11 per hemisphere) in MNI152 standard space, together with the ROI protocols used to reconstruct them.The bundles were reconstructed from multi-shell diffusion MRI of the Human Connectome Project in a quality-controlled pool of 100 unrelated subjects, with the population map of each bundle computed from 40 subjects. They span three connectivity systems: hypothalamo-limbic (precommissural and postcommissural fornix, stria terminalis, ventral amygdalofugal pathway), hypothalamo-diencephalic (mammillothalamic tract, stria medullaris, accumbens-VTA segment of the medial forebrain bundle), and hypothalamo-brainstem (dorsal longitudinal fasciculus, mammillotegmental tract, hypothalamo-mesencephalic reticular fibers, hypothalamo-brainstem segment of the medial forebrain bundle).File contents. Probabilistic maps are provided as NIfTI (.nii.gz) volumes in MNI152 space, named by bundle abbreviation and hemisphere (e.g. MTT_L). Voxel values are the subject-averaged normalized tract density, ranging from 0 to 1; they reflect both how many subjects have streamlines passing through a voxel and how densely those streamlines run, and are not a per-voxel proportion of subjects. A threshold of 0.2 was used for the renderings and analyses in the associated paper. ROI protocols for reconstructing each bundle de novo in new datasets are provided as separate files. hypothalamus; diffusion MRI; tractography; probabilistic atlas; white matter; connectome; MNI152 乔豫川 李贞仪 PUBLIC https://creativecommons.org/licenses/by/4.0/ dataset Science Data Bank