Mapping neurodegeneration with diffusion MRI: biomarkers, mechanisms, and clinical translation

Neurodegenerative diseases share convergent mechanisms involving microstructural degeneration, neuroinflammation, vascular dysfunction, and impaired brain fluid homeostasis. The neurovascular unit (NVU) represents a critical interface where these processes interact, integrating neuronal, glial, vascular, and perivascular components that regulate metabolism, immune surveillance, and waste clearance. This review examines advanced diffusion MRI as a noninvasive framework to investigate NVU-related pathology, with a specific focus on tissue microstructure, water dynamics, and perivascular spaces (PVS). We summarize diffusion MRI techniques ranging from conventional diffusion tensor imaging to multi-compartment and biophysical models that probe neurite architecture, extracellular free water, and perivascular transport. Across aging and major neurodegenerative disorders, diffusion-derived markers consistently reveal microstructural disorganization, extracellular fluid expansion, PVS enlargement, and glymphatic dysfunction. These alterations reflect coupled tissue–fluid pathology rather than isolated cellular damage. While advanced diffusion approaches provide increased sensitivity to early and subtle changes, they are influenced by acquisition quality, model assumptions, physiological confounders, and limited histopathological validation. Importantly, diffusion MRI metrics should be interpreted as complementary biomarkers that enhance, but do not replace, established diagnostic criteria and molecular biomarkers for specific neurodegenerative diseases. When integrated within multimodal and longitudinal frameworks, diffusion MRI offers valuable insights into NVU dysfunction, supporting early disease stratification, progression monitoring, and mechanistic understanding of neurodegeneration.