Data from: Clinical and pathologic correlations of machine learning quantification of Aβ deposits across three brain regions of decedents with Alzheimer's disease

The advent of machine learning enables scalable quantification of
neuropathology, offering deeper phenotyping of Alzheimer's disease
(AD). In this study, we quantified amyloid-beta (Aβ) deposits across
multiple brain regions and examined their associations with clinical,
demographic, and genetic factors in persons pathologically diagnosed with
AD. We analyzed densities (#/mm2) of cored plaques, diffuse plaques, and
cerebral amyloid angiopathy (CAA) in 273 individuals from three
Alzheimer’s Disease Research Centers. Formalin-fixed paraffin-embedded
sections of frontal, temporal, and parietal cortices were immunostained
and digitized, generating 799 whole slide images (WSIs). Following log
transformation, mixed-effects modeling revealed that the parietal cortex
had the highest cored plaque densities (p < 0.001), while the
temporal cortex had the highest diffuse plaque densities (p <
0.001); CAA showed no regional differences. Wilcoxon rank-sum tests and
covariate-adjusted linear models showed ApoE ε4− status was associated
with higher cored plaque densities in the temporal lobe (p = 0.04), while
ApoE ε4+ status was associated with higher diffuse plaque densities in the
temporal lobe (p = 0.001) and increased CAA in the frontal lobe (p =
0.004). These findings support deeper phenotyping to define generalizable
patterns of disease heterogeneity and neuroanatomical distribution in AD,
providing insights that may guide precision medicine–based approaches.