
Alzheimer s & Dementia, Journal Year: 2025, Volume and Issue: unknown
Published: Jan. 27, 2025
Abstract INTRODUCTION Traditional multivariate methods for neuroimaging studies overlook the interdependent relationship between brain features. This study addresses this gap by analyzing relative volumetric patterns to capture how Alzheimer's disease (AD) and genetics influence structure along continuum. METHODS analyzed data from participants across AD continuum Families (ALFA) Disease Neuroimaging Initiative (ADNI) studies. Compositional analysis (CoDA) was exploited examine variations that (1) were linked different stages compared cognitively unimpaired amyloid‐β–negative (CU A−) individuals (2) varied genetic risk. RESULTS stage–specific compositional scores identified, differentiating CU A− those in more advanced stages. Genetic risk–stratified models revealed a broader landscape affecting morphology AD, beyond well‐known apolipoprotein E ε4 allele. DISCUSSION CoDA emerges as an alternative framework deepen understanding of AD‐related structural changes support targeted interventions at higher Highlights variation region volumes, captured scores, capable discerning subjects within other disease‐stage groups also uncovered vulnerability specific regions each stage is integrating magnetic resonance imaging two cohorts without stringent requirements harmonization. translates advantage, traditional methods, strengthens reliability cross‐study comparisons standardizing despite labeling agreements, facilitating collaborative large‐scale research. The algorithm sensitive AD‐specific effects, main display little overlap with age‐specific score. provides accurate addressing its nature, which can development approaches, opening new avenues enhancing health.
Language: Английский