gms | German Medical Science

67. Jahrestagung der Deutschen Gesellschaft für Medizinische Informatik, Biometrie und Epidemiologie e. V. (GMDS), 13. Jahreskongress der Technologie- und Methodenplattform für die vernetzte medizinische Forschung e. V. (TMF)

21.08. - 25.08.2022, online

Two routes to Alzheimer’s disease based on differential structural changes in key brain regions

Meeting Abstract

  • Yasmin Hollenbenders - Zentrum für Maschinelles Lernen, Hochschule Heilbronn, Heilbronn, Germany; Medizinische Fakultät Heidelberg, Universität Heidelberg, Heidelberg, Germany
  • Monika Pobiruchin - GECKO-Institut, Hochschule Heilbronn, Heilbronn, Germany
  • Alexandra Reichenbach - Zentrum für Maschinelles Lernen, Hochschule Heilbronn, Heilbronn, Germany; Medizinische Fakultät Heidelberg, Universität Heidelberg, Heidelberg, Germany

Deutsche Gesellschaft für Medizinische Informatik, Biometrie und Epidemiologie. 67. Jahrestagung der Deutschen Gesellschaft für Medizinische Informatik, Biometrie und Epidemiologie e. V. (GMDS), 13. Jahreskongress der Technologie- und Methodenplattform für die vernetzte medizinische Forschung e.V. (TMF). sine loco [digital], 21.-25.08.2022. Düsseldorf: German Medical Science GMS Publishing House; 2022. DocAbstr. 113

doi: 10.3205/22gmds046, urn:nbn:de:0183-22gmds0468

Published: August 19, 2022

© 2022 Hollenbenders et al.
This is an Open Access article distributed under the terms of the Creative Commons Attribution 4.0 License. See license information at



Introduction: Alzheimer's disease (AD) is a neurodegenerative disorder and the most common type of dementia [1]. Neuropathological changes precede noticeable symptoms by up to two decades [2], [3], [4], [5]. Usually, AD is diagnosed in patients with signs of dementia but absence of other dementias [6], which results in a heterogeneous disease pattern. To deal with the heterogeneity, patients are stratified based on cognitive function [7] or brain atrophy [8]. To bypass the necessity of defining the onset of the disease for modeling its progress, we are using Hidden Markov Models (HMMs) [9] with markers of brain structure.

Methods: Data used for this study were obtained from the Alzheimer's Disease Neuroimaging Initiative (ADNI) database [10]. We focused on structural MRI images that were preprocessed, and quality checked [11] because cortical thickness is a marker rather robust against different types of scanners [12]. Subcortical volume (SV) and cortical thickness (CT) of 14 brain regions well known for their involvement in AD [13], [14], [15], [16] were extracted to train a Gaussian HMM [17]. For a complete picture of the progression, we included subjects irrespective of their diagnosis. To tie the progression in brain atrophy to a rich behavioral marker, we analyzed the ADAS-cog 11 [18] subscores of the subjects in the different model states.

Results: The optimal model consists of eight states with differentiable neuroanatomical features (Figure 1 [Fig. 1]). It forms two parallel routes crossing once at a very early point and merging at the final state. The cortical route is characterized by early and sustained atrophy in cortical regions (PreC, IT, MT). The limbic route is characterized by an early decrease in limbic regions (AM, HC, PHC, EC). All regions discriminate the two routes significantly throughout the progression. Cognitive differences between the subjects traversing the two routes are most noticeable in the memory domain with subjects from the limbic route experiencing stronger memory impairments and a similar albeit weaker impairment for language and praxis sub scores.

Discussion: The hippocampus plays a prominent role in AD research [19] and is associated with early atrophy and memory deficiencies. We corroborate these findings with the limbic route of our model. The second path with early atrophy of mainly cortical regions is supported by Goyal [9], who also found two paths with differentiable involvement of the hippocampus. Contrary to Goyal [9], our model is more parsimonious with only one imaging modality, which is better suited for diagnostic purposes. Furthermore, we found specific cognitive decline tied to the different routes.

Conclusion: This study corroborates that HMMs constitute a promising approach to differentiate several disease progressions for a heterogeneous disorder like AD.

The authors declare that they have no competing interests.

The authors declare that an ethics committee vote is not required.


Alzheimer's Association. 2019 Alzheimer's disease facts and figures. Alzheimers Dement. 2019;15(3):321-87.
Bateman RJ, Xiong C, Benzinger TLS, Fagan AM, Goate A, Fox NC, et al. Clinical and Biomarker Changes in Dominantly Inherited Alzheimer's Disease. New Engl J Med. 2012;367(9):795-804.
Braak H, Thal DR, Ghebremedhin E, Del Tredici K. Stages of the Pathologic Process in Alzheimer Disease: Age Categories From 1 to 100 Years. J Neuropath Exp Neur. 2011;70(11):960-9.
Dubois B, Padovani A, Scheltens P, Rossi A, Dell'Agnello G. Timely Diagnosis for Alzheimer's Disease: A Literature Review on Benefits and Challenges. J Alzheimers Dis. 2016;49(3):617-31.
Villemagne VL, Burnham S, Bourgeat P, Brown B, Ellis KA, Salvado O, et al. Amyloid \u946 ? deposition, neurodegeneration, and cognitive decline in sporadic Alzheimer's disease: a prospective cohort study. Lancet Neurol. 2013;12(4):357-67. DOI: 10.1016/s1474-4422(13)70044-9 External link
Dubois B, Feldman HH, Jacova C, Hampel H, Molinuevo JL, Blennow K, et al. Advancing research diagnostic criteria for Alzheimer's disease: the IWG-2 criteria. Lancet Neurol. 2014;13(6):614-29.
Scheltens NME, Tijms BM, Koene T, Barkhof F, Teunissen CE, Wolfsgruber S, et al. Cognitive subtypes of probable Alzheimer's disease robustly identified in four cohorts. Alzheimers Dement. 2017;13(11):1226-36.
Dong A, Honnorat N, Gaonkar B, Davatzikos C. CHIMERA: Clustering of Heterogeneous Disease Effects via Distribution Matching of Imaging Patterns. IEEE T Med Imaging. 2016;35(2):612-21.
Goyal D, Tjandra D, Migrino RQ, Giordani B, Syed Z, Wiens J. Characterizing heterogeneity in the progression of Alzheimer's disease using longitudinal clinical and neuroimaging biomarkers. Alzheimers Dement: Diagnosis, Assessment & Disease Monitoring. 2018;10:629-37.
Alzheimer’s Disease Neuroimaging Initiative (ADNI) [Internet]. 2017 [cited 2022 Mar 31]. Available from: External link
Jack CR, Bernstein MA, Fox NC, Thompson P, Alexander G, Harvey D, et al. The Alzheimer's Disease Neuroimaging Initiative (ADNI): MRI methods. J Magn Reson Imaging. 2008;27(4):685-91.
Ecker C, Marquand A, Mourao-Miranda J, Johnston P, Daly EM, Brammer MJ, et al. Describing the Brain in Autism in Five Dimensions – Magnetic Resonance Imaging-Assisted Diagnosis of Autism Spectrum Disorder Using a Multiparameter Classification Approach. J Neurosci. 2010;30(32):10612-23.
Fan Y, Batmanghelich N, Clark CM, Davatzikos C. Spatial patterns of brain atrophy in MCI patients, identified via high-dimensional pattern classification, predict subsequent cognitive decline. NeuroImage. 2008;39(4):1731-43.
Misra C, Fan Y, Davatzikos C. Baseline and longitudinal patterns of brain atrophy in MCI patients, and their use in prediction of short-term conversion to AD: Results from ADNI☆. NeuroImage. 2009;44(4):1415-22.
Young J, Modat M, Cardoso MJ, Mendelson A, Cash D, Ourselin S. Accurate multimodal probabilistic prediction of conversion to Alzheimer's disease in patients with mild cognitive impairment. NeuroImage Clin. 2013;2:735-45.
Zu C, Jie B, Liu M, Chen S, Shen D, Zhang D. Label-aligned multi-task feature learning for multimodal classification of Alzheimer’s disease and mild cognitive impairment. Brain Imaging Behav. 2016;10(4):1148-59.
Pedregosa F, Varoquaux G, Gramfort A, Michel V, Thirion B, Grisel O, et al. Scikit-learn: Machine learning in Python. J Mach Learn Res. 2011;12(Oct):2825-30.
Rosen WG, Mohs RC, Davis KL. A new rating scale for Alzheimer's disease. Am J Psychiat. 1984;141(11):1356-64.
Rao YL, Ganaraja B, Murlimanju BV, Joy T, Krishnamurthy A, Agrawal A. Hippocampus and its involvement in Alzheimer’s disease: a review. 3 Biotech. 2022;12(2).