Imaging in Lifelines: MemoLife

Objective: Generally, this study aims to investigate how cerebrovascular function relates to aging, mild cognitive impairment (MCI) and AD. More specifically, the aim is to identify, based on cerebrovascular function, individuals at high-risk for disease (in the healthy population) and to assess risk for disease progression from MCI to AD. The first part of the study is community-driven and focuses on the relationship between coronary artery calcification (CAC) and cognition in healthy older individuals, as this relationship remains not fully elucidated. MRI-derived measures will be obtained to study the cerebrovascular system, where we predominantly focus on measures of cerebral blood flow. Two innovative microvascular imaging techniques that reflect cerebral blood flow, ASL-CBF and DSC-VAI, will be evaluated on their potential to be used as early biomarkers for cognitive impairment. To this end, identical MR data sets will also be obtained in groups of MCI and AD patients. Ultimately, the goal is to build an artificial intelligence model for automated analysis and integration of both imaging and non-imaging (clinical) information in order to make predictions about future cognitive change.
Study design: This study will be a combination of a cross-sectional study and a prospective cohort study. Part of the data for the healthy group has already been obtained by Lifelines (e.g. CAC scores and relevant medical and lifestyle information). All participants will undergo (longitudinal) a neuropsychological assessment (longitudinally for healthy individuals and MCI patients), blood sampling and MRI scanning, including anatomical scans, resting-state functional MRI, diffusion kurtosis imaging, quantitative susceptibility mapping, ASL-CBF and DSC-VAI. All data will be integrated in a so-called deep convolutional neural network (DCNN) model which can be used to predict changes in cognitive function.
Study population: The participants for the longitudinal, community-driven, part of the study will be selected from a subpopulation of Lifelines, where CAC scores have already been obtained in previous examinations. A total of 280 participants will be selected and dichotomized into “high CAC score” versus “low CAC score”. For the cross-sectional part of the study, 50 patients with MCI and 25 patients with AD will be recruited through the Memory Clinic UMCG and we will recruit patients from other memory clinics in the north of the Netherlands.
Main study parameters/endpoints: Main study parameters entail blood flow-related measures such as cerebral blood flow, cerebral blood volume, etc., and other imaging-derived parameters.
Non-imaging parameters include cognitive measures and other variables such as clinical information, sociodemographic factors, and lifestyle factors. Finally, the MR images will serve as input for the DCNN risk-detection model and will be augmented with the non-imaging parameters.