Diffusion MRI indices of brain microstructure: Evaluating the Ball‐and‐Stick model in distinguishing Alzheimer’s disease

Background Diffusion MRI is a valuable tool to assess white matter (WM) microstructure biomarkers of brain aging and Alzheimer’s disease (AD). The most common model used extract diffusion properties the tensor (DTI). However, DTI cannot resolve complex WM suffers from partial volume effects. Further, majority multi-site imaging clinical studies, such as ADNI, are limited single-shell acquisitions due time constraints. Therefore, multi-compartment models not feasible overcome these limitations except multi-tensor models, Ball-and-Stick (B&S) under some B&S offers novel approach derive for early detection AD because it splits diffusion-weighted MR signal into several anisotropic components (Stick) single isotropic component (Ball). This implemented in Anima software. Here, we propose classification using four two indices quantified various anatomical regions interest (ROIs) compare their ability distinguish AD. Method Cross-sectional data were available 413 ADNI-3 participants: 68 (age: 76.81±8.3; 50% women) 345 cognitively normal (CN; age:72.91±7.27; 62% women). We processed T1-weighted sequences following pipeline Fig.1. obtained maps: fractional anisotropy (FA), axial (AxD), mean (MD), radial (RD) diffusivity; free water weight (FWW) Stick-axial diffusivity (ST). maps within set ROIs. Then, performed high-dimensional pattern approach. For each metric, consider all ROIs jointly identify minimal whose values maximally differentiate between CN. Result with regional FWW ST measures shows greatest discrimination CN (FWW ROCAUC = 0.942; 0.938) compared (Table 1). Left PHG Parahippocampal Gyrus Right Hippocampus have highest contribution 2, Fig.2). Conclusion Our results suggest more sensitive discriminate than measures, widely research. has advantage identifying separating crossing fibers, therefore, potential offer greater insights underlying pathology