AbstractBackgroundWith the advent of disease modifying treatments for dementia, it is essential to restructure memory clinic services to provide accurate and early diagnosis to enable precision therapies and recruitment to clinical trials. The Oxford Brain Health Clinic (BHC) offers a comprehensive clinical and embedded research assessment in the patient pathway, including a research‐quality brain MRI scan (using the UK Biobank protocol), well‐tolerated by patients (Griffanti et al., 2022, Neuroimage:Clinical). In this study we investigated whether the images obtained from BHC patients were of sufficient quality for clinical/diagnostic purposes (i.e. to provide a radiology report) and research purposes (i.e. suitable to extract quantitative measures with automated software) and whether the scan quality was related to patients’ characteristics (capacity, frailty, anxiety, cognitive impairment ‐ CI), to inform future triage to Brain Health Clinics.MethodT1 and FLAIR scans from 138 BHC patients were reported by a radiologist and independently rated as high, medium or low quality for research purposes (consensus between 2 raters). The following patients’ characteristics were evaluated: capacity (yes/no), Rockwood Clinical Frailty Scale (more frail if 5 or more), anxiety before/during the scan (low/not‐low), Addenbrooke’s Cognitive Examination III (ACE‐III, score above or below 82). A two‐sided Fisher‐Freeman‐Halton test was run to test the association between scan quality (3 classes) and patients characteristics (2 classes for each independent variable). Bonferroni‐adjusted post‐hoc comparisons were performed between scan quality classes using 2×2 Fisher’s Exact Tests.ResultThe radiologist was able to provide a clinical report for all scans. Over 90% of scans were rated high or medium quality for research purposes (Fig.1). No significant associations were observed between patients’ characteristics and FLAIR quality. T1 quality was significantly associated with patients’ capacity (p = 0.001, more low‐quality images from patients lacking capacity, Fig.2) and frailty (p = 0.014, more low‐quality images from frailer patients, Fig.3).ConclusionResearch‐quality MRI scans in memory clinic are useful for both clinical and research purposes. For research comprised of high‐quality MRI, relatively generous thresholds relating to frailty and capacity could be considered to balance the trade‐off between inclusivity (i.e. increasing representativeness) and adherence to protocols.