Machine Learning-Based Perivascular Space Volumetry in Alzheimer Disease
Katerina Deike1, Andreas Decker, Paul Scheyhing
1From the German Center for Neurodegenerative Diseases (DZNE), Bonn, Germany (K.D., A.D., K.F., O.K., F.J., Annika Spottke, N.R., M.W., S.R., M.T.H., F.B., Alfredo Ramirez, S.W., L.K., M.S., M.C.S., G.C.P., Anja Schneider, Alexander Radbruch); Department of Neuroradiology, University Hospital, Bonn, Germany (K.D., P.S., D.P., Alexander Radbruch); Department of Neurodegenerative Disease and Geriatric Psychiatry/Psychiatry, University Hospital Bonn, Bonn, Germany (J.H., N.Z., K.F., M.W., Alfredo Ramirez, S.W., L.K., Anja Schneider); Department of Radiology, German Cancer Research Center (DKFZ), Heidelberg, Germany (D.P.); German Center for Neurodegenerative Diseases (DZNE), Berlin, Germany (O.P., S.D.F., J.P., E.S., S.A.); Institute of Psychiatry and Psychotherapy, Charité-Corporate Member of Freie Universität Berlin and Humboldt-Universität zu Berlin, Berlin, Germany (O.P., S.D.F., L.-S.S., L.P.); Department of Psychiatry and Psychotherapy, Charité, Berlin, Germany (J.P., E.S., S.A., A.L.); Department of Psychiatry and Psychotherapy, School of Medicine, Munich, Germany (J.P.); University of Edinburgh and UK DRI, Edinburgh, United Kingdom (J.P.); German Center for Neurodegenerative Diseases (DZNE), Goettingen, Germany (J.W.); Department of Psychiatry and Psychotherapy, University Medical Center, Goettingen, Germany (J.W., C.B., N.H.); Neurosciences and Signaling Group, Institute of Biomedicine (iBiMED), University of Aveiro, Aveiro, Portugal (J.W.); Department of Psychiatry, University of Cologne, Cologne, Germany (F.J., Ayda Rostamzadeh); Excellence Cluster on Cellular Stress Responses in Aging-Associated Diseases, University of Cologne, Cologne, Germany (F.J., Alfredo Ramirez); German Center for Neurodegenerative Diseases (DZNE), Magdeburg, Germany (E.D., W.G., E.I.I., Michaela Butryn, L.D., R.Y.); Institute of Cognitive Neurology and Dementia Research, Otto-von-Guericke University, Magdeburg, Germany (E.D., W.G., E.I.I., Michaela Butryn); Department for Psychiatry and Psychotherapy, University Clinic Magdeburg, Magdeburg, Germany (E.I.I.); German Center for Neurodegenerative Diseases (DZNE), Munich, Germany (K.B., M.E., R.P.); Institute for Stroke and Dementia Research, LMU Munich, Germany (K.B., D.J., M.E.); Department of Psychiatry and Psychotherapy, LMU Munich, Germany (R.P., B.-S.R.); Munich Cluster for Systems Neurology (SyNergy), Munich, Germany (R.P.); Ageing Epidemiology Research Unit, School of Public Health, Imperial College London, London, United Kingdom (R.P.); Sheffield Institute for Translational Neuroscience, University of Sheffield, Sheffield, United Kingdom (R.P., B.-S.R.); Department of Neuroradiology, University Hospital Munich, Munich, Germany (B.-S.R.); German Center for Neurodegenerative Diseases (DZNE), Rostock, Germany (S.T., I.K., D.G.); Department of Psychosomatic Medicine, Rostock University Medical Center, Rostock, Germany (S.T., I.K., D.G.); German Center for Neurodegenerative Diseases (DZNE), Tübingen, Germany (C.L., M.H.M.); Section for Dementia Research, Hertie Institute for Clinical Brain Research and Department of Psychiatry and Psychotherapy, Tübingen, Germany (C.L.); Department of Psychiatry and Psychotherapy, University of Tübingen, Tübingen Germany (M.H.M.); Department of Neurology, University of Bonn, Bonn, Germany (Annika Spottke); Division of Neurogenetics and Molecular Psychiatry, Department of Psychiatry and Psychotherapy, Cologne, Germany (Alfredo Ramirez); Department of Psychiatry and Glenn Biggs Institute for Alzheimer's and Neurodegenerative Diseases, San Antonio, TX (Alfredo Ramirez); Institute for Medical Biometry, Informatics, and Epidemiology, University Hospital Bonn, Bonn, Germany (M.C.S., Moritz Berger); Berlin Center for Advanced Neuroimaging, Charité-Universitätsmedizin, Berlin, Germany (S.H.); MR-Research in Neurosciences, Department of Cognitive Neurology, Göttingen, Germany (P.D.); Department for Biomedical Magnetic Resonance, University of Tübingen, Tübingen, Germany (K.S.); Division of Vascular Neurology, Department of Neurology, University Hospital Bonn, Bonn, Germany (G.C.P.); and Institute for Applied Mathematics, University of Bonn, Bonn, Germany (A.E.).
Objectives:
Impaired perivascular clearance has been suggested as a contributing factor to the pathogenesis of Alzheimer disease (AD). However, it remains unresolved when the anatomy of the perivascular space (PVS) is altered during AD progression. Therefore, this study investigates the association between PVS volume and AD progression in cognitively unimpaired (CU) individuals, both with and without subjective cognitive decline (SCD), and in those clinically diagnosed with mild cognitive impairment (MCI) or mild AD.
Materials And Methods:
A convolutional neural network was trained using manually corrected, filter-based segmentations (n = 1000) to automatically segment the PVS in the centrum semiovale from interpolated, coronal T2-weighted magnetic resonance imaging scans (n = 894). These scans were sourced from the national German Center for Neurodegenerative Diseases Longitudinal Cognitive Impairment and Dementia Study. Convolutional neural network-based segmentations and those performed by a human rater were compared in terms of segmentation volume, identified PVS clusters, as well as Dice score. The comparison revealed good segmentation quality (Pearson correlation coefficient r = 0.70 with P < 0.0001 for PVS volume, detection rate in cluster analysis = 84.3%, and Dice score = 59.0%). Subsequent multivariate linear regression analysis, adjusted for participants' age, was performed to correlate PVS volume with clinical diagnoses, disease progression, cerebrospinal fluid biomarkers, lifestyle factors, and cognitive function. Cognitive function was assessed using the Mini-Mental State Examination, the Comprehensive Neuropsychological Test Battery, and the Cognitive Subscale of the 13-Item Alzheimer's Disease Assessment Scale.
Results:
Multivariate analysis, adjusted for age, revealed that participants with AD and MCI, but not those with SCD, had significantly higher PVS volumes compared with CU participants without SCD ( P = 0.001 for each group). Furthermore, CU participants who developed incident MCI within 4.5 years after the baseline assessment showed significantly higher PVS volumes at baseline compared with those who did not progress to MCI ( P = 0.03). Cognitive function was negatively correlated with PVS volume across all participant groups ( P ≤ 0.005 for each). No significant correlation was found between PVS volume and any of the following parameters: cerebrospinal fluid biomarkers, sleep quality, body mass index, nicotine consumption, or alcohol abuse.
Conclusions:
The very early changes of PVS volume may suggest that alterations in PVS function are involved in the pathophysiology of AD. Overall, the volumetric assessment of centrum semiovale PVS represents a very early imaging biomarker for AD.
