Print Close

A Computational Ontology Framework and Diagnostic Reporting of Brain Atrophy Profiles in Dementia

Poster No:

1875

Submission Type:

Abstract Submission

Authors:

Devesh Singh¹, Alice Grazia¹, Achim Reiz², Andreas Hermann¹^,3, Alexander Bernhardt⁴^,5, Katharina Buerger⁴^,6, Emrah Düzel⁷^,8, Klaus Fließbach⁹^,10, Christoph Laske¹¹^,12, Robert Perneczky⁴^,13^,14^,15, Oliver Peters¹⁶^,17, Josef Priller¹⁶^,18^,19^,20, Johannes Prudlo¹^,21, Anja Schneider⁹^,10, Annika Spottke⁹^,22, Matthis Synofzik¹¹^,23, Jens Wiltfang²⁴^,25^,26, Frank Jessen⁹^,27^,28, Stefan Teipel¹^,29, Martin Dyrba¹

Institutions:

¹German Center for Neurodegenerative Diseases (DZNE), Rostock, Germany, ²Rostock University, Rostock, Germany, ³Translational Neurodegeneration Section “Albrecht Kossel”, Department of Neurology, University Medical Center Rostock, Rostock, Germany, ⁴German Center for Neurodegenerative Diseases (DZNE), Munich, Germany, ⁵Department of Neurology, University Hospital of Munich, LMU Munich, Munich, Germany, ⁶Institute for Stroke and Dementia Research (ISD), University Hospital, LMU Munich, Munich, Germany, ⁷Institute of Cognitive Neurology and Dementia Research, Magdeburg, Germany, ⁸German Center for Neurodegenerative Diseases (DZNE), Magdeburg, Germany, ⁹German Center for Neurodegenerative Diseases (DZNE), Bonn, Germany, ¹⁰University of Bonn Medical Center, Dept. of Neurodegenerative Disease and Geriatric Psychiatry/Psychiatry, Bonn, Germany, ¹¹German Centre for Neurodegenerative Diseases (DZNE), Tübingen, Germany, ¹²Section for Dementia Research, Hertie Institute for Clinical Brain Research and Department of Psychiatry and Psychotherapy, University of Tübingen, Tübingen, Germany, ¹³Department of Psychiatry and Psychotherapy, University Hospital, LMU Munich, Munich, Germany, ¹⁴Munich Cluster for Systems Neurology (SyNergy) Munich, Munich, Germany, ¹⁵Ageing Epidemiology Research Unit (AGE), School of Public Health, Imperial College London, London, United Kingdom, ¹⁶German Center for Neurodegenerative Diseases (DZNE), Berlin, Germany, ¹⁷Charité – Universitätsmedizin Berlin, corporate member of Freie Universität Berlin and Humboldt-Universität zu Berlin-Institute of Psychiatry and Psychotherapy, Berlin, Germany, ¹⁸Department of Psychiatry and Psychotherapy, Charité, Berlin, Germany, ¹⁹School of Medicine, Technical University of Munich; Department of Psychiatry and Psychotherapy, Munich, Germany, ²⁰University of Edinburgh and UK DRI, Edinburgh, United Kingdom, ²¹Department of Neurology, University Medical Centre, Rostock, Germany, ²²Department of Neurology, University of Bonn, Venusberg-Campus 1, Bonn, Germany, ²³Division Translational Genomics of Neurodegenerative Diseases, Hertie Institute for Clinical Brain Research and Center of Neurology, University of Tübingen, Tübingen, Germany, ²⁴Department of Psychiatry and Psychotherapy, Medical University Göttingen, Göttingen, Lower Saxony, ²⁵German Center for Neurodegenerative Diseases (DZNE), Göttingen, Germany, ²⁶Neurosciences and Signaling Group, Institute of Biomedicine (iBiMED), Department of Medical Sciences, University of Aveiro, Aveiro, Portugal, ²⁷Department of Psychiatry, University of Cologne, Medical Faculty, Cologne, Germany, ²⁸Excellence Cluster on Cellular Stress Responses in Aging-Associated Diseases (CECAD), University of Cologne, Cologne, Germany, ²⁹Department of Psychosomatic Medicine, University of Rostock, Rostock, Germany

First Author:

Devesh Singh
German Center for Neurodegenerative Diseases (DZNE)
Rostock, Germany

Co-Author(s):

Alice Grazia
German Center for Neurodegenerative Diseases (DZNE)
Rostock, Germany

Achim Reiz
Rostock University
Rostock, Germany

Andreas Hermann
German Center for Neurodegenerative Diseases (DZNE)|Translational Neurodegeneration Section “Albrecht Kossel”, Department of Neurology, University Medical Center Rostock
Rostock, Germany|Rostock, Germany

Alexander Bernhardt
German Center for Neurodegenerative Diseases (DZNE)|Department of Neurology, University Hospital of Munich, LMU Munich
Munich, Germany|Munich, Germany

Katharina Buerger
German Center for Neurodegenerative Diseases (DZNE)|Institute for Stroke and Dementia Research (ISD), University Hospital, LMU Munich
Munich, Germany|Munich, Germany

Emrah Düzel
Institute of Cognitive Neurology and Dementia Research|German Center for Neurodegenerative Diseases (DZNE)
Magdeburg, Germany|Magdeburg, Germany

Klaus Fließbach
German Center for Neurodegenerative Diseases (DZNE)|University of Bonn Medical Center, Dept. of Neurodegenerative Disease and Geriatric Psychiatry/Psychiatry
Bonn, Germany|Bonn, Germany

Christoph Laske
German Centre for Neurodegenerative Diseases (DZNE)|Section for Dementia Research, Hertie Institute for Clinical Brain Research and Department of Psychiatry and Psychotherapy, University of Tübingen
Tübingen, Germany|Tübingen, Germany

Robert Perneczky
German Center for Neurodegenerative Diseases (DZNE)|Department of Psychiatry and Psychotherapy, University Hospital, LMU Munich|Munich Cluster for Systems Neurology (SyNergy) Munich|Ageing Epidemiology Research Unit (AGE), School of Public Health, Imperial College London
Munich, Germany|Munich, Germany|Munich, Germany|London, United Kingdom

Oliver Peters
German Center for Neurodegenerative Diseases (DZNE)|Charité – Universitätsmedizin Berlin, corporate member of Freie Universität Berlin and Humboldt-Universität zu Berlin-Institute of Psychiatry and Psychotherapy
Berlin, Germany|Berlin, Germany

Josef Priller
German Center for Neurodegenerative Diseases (DZNE)|Department of Psychiatry and Psychotherapy, Charité|School of Medicine, Technical University of Munich; Department of Psychiatry and Psychotherapy|University of Edinburgh and UK DRI
Berlin, Germany|Berlin, Germany|Munich, Germany|Edinburgh, United Kingdom

Johannes Prudlo
German Center for Neurodegenerative Diseases (DZNE)|Department of Neurology, University Medical Centre
Rostock, Germany|Rostock, Germany

Anja Schneider
German Center for Neurodegenerative Diseases (DZNE)|University of Bonn Medical Center, Dept. of Neurodegenerative Disease and Geriatric Psychiatry/Psychiatry
Bonn, Germany|Bonn, Germany

Annika Spottke
German Center for Neurodegenerative Diseases (DZNE)|Department of Neurology, University of Bonn, Venusberg-Campus 1
Bonn, Germany|Bonn, Germany

Matthis Synofzik
German Centre for Neurodegenerative Diseases (DZNE)|Division Translational Genomics of Neurodegenerative Diseases, Hertie Institute for Clinical Brain Research and Center of Neurology, University of Tübingen
Tübingen, Germany|Tübingen, Germany

Jens Wiltfang
Department of Psychiatry and Psychotherapy, Medical University Göttingen|German Center for Neurodegenerative Diseases (DZNE)|Neurosciences and Signaling Group, Institute of Biomedicine (iBiMED), Department of Medical Sciences, University of Aveiro
Göttingen, Lower Saxony|Göttingen, Germany|Aveiro, Portugal

Frank Jessen
German Center for Neurodegenerative Diseases (DZNE)|Department of Psychiatry, University of Cologne, Medical Faculty|Excellence Cluster on Cellular Stress Responses in Aging-Associated Diseases (CECAD), University of Cologne
Bonn, Germany|Cologne, Germany|Cologne, Germany

Stefan Teipel
German Center for Neurodegenerative Diseases (DZNE)|Department of Psychosomatic Medicine, University of Rostock
Rostock, Germany|Rostock, Germany

Martin Dyrba
German Center for Neurodegenerative Diseases (DZNE)
Rostock, Germany

E-Poster

Introduction:

Estimates suggest that the global number of people with dementia could increase to 153 million cases in 2050, with Alzheimer's disease (AD) and frontotemporal lobar degeneration (FTLD) being the most common cause of dementia. The behavioral-variant of frontotemporal dementia (bvFTD) is at times hard to distinguish from AD. Both are marked with regional volumetric loss of grey matter, as visible in T1-weighted MRI scans. Software tools for MRI volumetry can reliably detect and summarize these volumetric findings, however, existing tools often report findings limited to fixed anatomical regions. These tools lack a generalizable framework for aggregating disease pathology at different levels of brain abstractions such as the lobe or hemisphere. They also lack intuitive visualizations for inspecting distinct atrophy profiles in exploratory research. We propose a computational pipeline for quantifying hierarchical volumetric deviations and generating interactive summary visualizations, for the reporting of brain atrophy findings. We also propose a prototypical method of mapping individual MRI scans to possible dementia types.

Methods:

The brain segmentation tool FastSurferCNN [1] was utilized as a preprocessing step. We derive anatomical brain region segmentations and associated raw volumes. We used 3433 MRI scans from seven cohorts - two ADNI datasets, AIBL, DELCODE, DESCRIBE, EDSD and NIFD, including healthy control cases, as well as cases with prodromal and clinical AD, and bvFTD. We created a two-step pipeline for disease exploration (Fig. 1). First, we created a semantic model, encoding hierarchical anatomical membership relationships in a Web Ontology Language (OWL) model [2]. The OWL model also includes a computational framework for estimating and aggregating the regional volumetric deviations from normal levels (i.e. w-scores) [3]. Second, we developed a visualization framework, providing visual summary plots implemented with Plotly Sunburst charts. We developed a prototypical framework of similarity quantification, between sample to group averages, to map individual MRI scans to possible dementia types.

·Proposed disease exploration pipeline

Results:

The semantic modeling framework implements computationally efficient calculation of volumetric deviations based on additive linear regression models. The summary plots visualize volumetric deviations at every hierarchical brain abstraction level at once. The summary plots can be used to highlight both mean group characteristics as well as single-subject atrophy profiles, enhancing visual comparison of atrophy profiles with different disease types and phases.

Conclusions:

Overall, our pipeline enables an automated and efficient alternative for explorative research and reporting of pathological brain atrophy findings. Our pipeline was able to capture, both visually and numerically, the volumetric changes generally associated with AD progression. The initial results from the prototypical framework for mapping samples to dementia type were also promising. Our pipeline could likely assist clinicians in discovering brain pathologies in an interpretable and reliable manner.

Disorders of the Nervous System:

Neurodegenerative/ Late Life (eg. Parkinson’s, Alzheimer’s) ²

Modeling and Analysis Methods:

Classification and Predictive Modeling

Methods Development ¹

Other Methods

Keywords:

Data analysis

Modeling

MRI

Open-Source Code

Segmentation

Statistical Methods

Other - Alzheimer's Disease

^1|2Indicates the priority used for review

Provide references using author date format

[1] Henschel, L., Conjeti, S., Estrada, S., Diers, K., Fischl, B., & Reuter, M. (2020), ‘FastSurfer - A fast and accurate deep learning based neuroimaging pipeline’, NeuroImage, vol. 219.
[2] Rosse, C., & Mejino, J. L., Jr (2003), ‘A reference ontology for biomedical informatics: the Foundational Model of Anatomy’, Journal of biomedical informatics, vol. 36, no. 6, pp. 478–500.
[3] Jack, C. R., Jr, Petersen, R. C., Xu, Y. C., Waring, S. C., O'Brien, P. C., Tangalos, E. G., Smith, G. E., Ivnik, R. J., & Kokmen, E. (1997), ‘Medial temporal atrophy on MRI in normal aging and very mild Alzheimer's disease’, Neurology, vol. 49, no. 3, pp. 786–794.