Interactive resource

Brain Atlas

Where 13,926 brain metastases in 3,290 patients from 4 institutions were found, on the MNI152 brain at 1 mm. Population frequencies, not individual predictions. This update is not yet published; the published atlas and its findings are on the Brain Metastases page.

Update 2026.03 · unpublished

All primaries · Anatomical, coarseMNI152 · 1 mm
Twelve axial slices of the population brain metastasis frequency map, all primaries

Highest relative frequency

    Shown only where at least 5 lesions contribute; regions with fewer than 5 lesions are withheld. 1× is the frequency expected if metastases were spread evenly through the brain; for a region, its share of lesions divided by its share of brain volume, with 95% confidence intervals. Radiological convention: the patient's right is on the left. Full-resolution maps are available on request: Olivier.Morin@ucsf.edu.

    Compare

    Primary cancers on one scale

    The coarse anatomical atlas, each region colored by its share of lesions relative to its share of brain volume, from 0× to 2× and above. Gray regions have fewer than five lesions. Drag an image to turn it.

    All primaries3,290 patients
    Lung1,362 patients
    Breast577 patients
    Melanoma410 patients
    Other primaries941 patients

    Reference atlases

    Four ways to divide the brain

    28 regionsAnatomical, coarseLobes by side, deep nuclei, ventricles, cerebellum, brainstem and pituitary.Explore89 regionsAnatomical, fineCortical parcels (Desikan–Killiany) and subcortical structures, each side separately.Explore66 regionsFunctionalThalamic connectivity zones (Behrens 2003), basal ganglia and thalamic subregions (He 2020), white matter tracts (Tang 2017) and a cognitive decline map (Reich 2022).Explore10 regionsVascular territoriesTerritories of the anterior, middle and posterior cerebral arteries and the vertebrobasilar system, by side.Explore

    References

    1. Barrios J, Porter E, Capaldi DPI, et al. Multi-institutional atlas of brain metastases informs spatial modeling for precision imaging and personalized therapy. Nat Commun. 2025;16:4536. doi:10.1038/s41467-025-59584-7
    2. Morin O, Vallières M, Braunstein S, et al. An artificial intelligence framework integrating longitudinal electronic health records with real-world data enables continuous pan-cancer prognostication. Nat Cancer. 2021;2(7):709-722. PMID 35121948