Introducing rrAD420, an anatomical template and multi‐modal atlas for older adults

Abstract

AbstractBackgroundRisk Reduction for Alzheimer’s Disease (rrAD) is a recently completed randomized controlled trial assessing the effects of aerobic exercise training and pharmacological cardio‐vascular interventions on neurocognitive function in hypertensive older adults with a family history of dementia or subjective cognitive decline. The comprehensive neuroimaging protocol included anatomical, functional, and physiological MRI scans, obtained at baseline and after two years of intervention. 420 older subjects (68.8±5.9 years) were scanned on five different 3T MRI scanners (two Siemens, two GE, and one Philips). Brain atrophy and ventricular enlargement are common in this elderly study population, leading to substantial deviations from the standard MNI152 template (25.02±4.9 years). Severe artifacts might occur when transforming non‐normative brains into MNI standard space. Using the high‐quality baseline data of this study, we aim to integrate images from multiple modalities to create a reference that is better suited to investigate older populations.MethodsFor each subject, we acquired T1 MPRAGE, T2 FLAIR, ASL, DTI, and resting‐state fMRI. Each scan underwent rigorous manual quality control and linear sequence‐specific high precision alignment to the T1 image. All T1 images then underwent non‐linear DARTEL registration into a common space to create cohort‐specific transformations. We then transformed all images into the new common space, namely rrAD420 (Figure 1). Averages across subjects make up cohort‐ and modality‐specific templates (Figure 2). Seed‐based and data‐driven decompositions of resting‐state fMRI and DTI data created functional and structural connectivity atlases (Figure 3). Anatomical dual regression transforms common brain atlases from MNI into rrAD420 space, facilitating cross‐referencing.ResultsThe rrAD420 templates and atlases integrate multimodal neuroimaging data of older populations into a common space, accounting for cohort‐specific distinctions, such as cortical atrophy and enlarged ventricles. It provides references to commonly used brain parcellations which are integrated with functional and structural atlases created from the rrAD cohort. Thus, rrAD420 provides a reference framework for comprehensive multimodal MRI analyses of elderly cohorts.ConclusionsThrough rrAD420, longitudinal multi‐modal image analyses can now be carried out using templates unique to this population. Furthermore, rrAD420 should also apply to older populations in general, facilitating data integration, and biomarker detection.