Abstract
INTRODUCTION: Magnetic resonance imaging (MRI) is crucial for dementia diagnosis and a pre-requisite for amyloid-lowering therapies in Alzheimer's disease. Despite guidelines, many patients never undergo MRI due to limited scanner availability. Shorter scan times would reduce costs and patient burden. We developed and tested a fast MRI protocol incorporating highly accelerated sequences. METHODS: We compared blinded neuroradiologist assessments of a fast protocol with the standard-of-care protocol in a prospective real-world study. We estimated agreement coefficients to evaluate reliability. RESULTS: The fast protocol cut scan times by 63% and showed non-inferior reliability measures for diagnosis, visual scale ratings, and disease-modifying therapy eligibility assessment. Between scan-type, intra-rater reliability for diagnosis was greater than inter-rater reliability on the standard-of-care protocol (ratio of 1.37, 95% confidence interval: 1.21-1.58). DISCUSSION: This study proposed and applied a way of showing non-inferiority of a highly accelerated dementia protocol. Ultra-fast protocols could improve MRI access and patient equity and support the implementation of disease-modifying therapies. HIGHLIGHTS: The fast dementia protocol with four core sequences reduced acquisition time by 63%. The fast scan showed non-inferior reliability for diagnosis and visual ratings. Assessment for disease-modifying therapy eligibility was similar between scan types. Fast protocols may improve access to magnetic resonance imaging and diagnosis in dementia.