ORCID: https://orcid.org/0009-0003-9931-0710; Grazia, Alice ORCID: https://orcid.org/0000-0001-8861-3183; Reiz, Achim ORCID: https://orcid.org/0000-0003-1446-9670; Hermann, Andreas ORCID: https://orcid.org/0000-0002-7364-7791; Altenstein, Slawek ORCID: https://orcid.org/0000-0003-2753-5999; Beichert, Lukas ORCID: https://orcid.org/0009-0000-9259-9230; Bernhardt, Alexander ORCID: https://orcid.org/0000-0002-2572-5062; Buerger, Katharina ORCID: https://orcid.org/0000-0002-5898-9953; Butryn, Michaela; Dechent, Peter ORCID: https://orcid.org/0009-0006-4005-3352; Duezel, Emrah; Ewers, Michael ORCID: https://orcid.org/0000-0001-5231-1714; Fliessbach, Klaus; Freiesleben, Silka D. ORCID: https://orcid.org/0000-0002-2141-8671; Glanz, Wenzel ORCID: https://orcid.org/0000-0002-5865-4176; Hetzer, Stefan ORCID: https://orcid.org/0000-0002-1773-1518; Janowitz, Daniel ORCID: https://orcid.org/0009-0003-4090-547X; Kilimann, Ingo ORCID: https://orcid.org/0000-0002-3269-4452; Kimmich, Okka ORCID: https://orcid.org/0009-0008-2119-7590; Laske, Christoph; Levin, Johannes ORCID: https://orcid.org/0000-0001-5092-4306; Lohse, Andrea; Luesebrink, Falk ORCID: https://orcid.org/0000-0001-5770-0727; Munk, Matthias ORCID: https://orcid.org/0000-0002-5339-4045; Perneczky, Robert ORCID: https://orcid.org/0000-0003-1981-7435; Peters, Oliver ORCID: https://orcid.org/0000-0003-0568-2998; Preis, Lukas ORCID: https://orcid.org/0000-0001-7601-6410; Priller, Josef ORCID: https://orcid.org/0000-0001-7596-0979; Prudlo, Johannes; Rauchmann, Boris S. ORCID: https://orcid.org/0000-0003-4547-6240; Rostamzadeh, Ayda ORCID: https://orcid.org/0000-0001-5189-134X; Roy-Kluth, Nina; Scheffler, Klaus ORCID: https://orcid.org/0000-0001-6316-8773; Schneider, Anja; Schneider, Luisa S. ORCID: https://orcid.org/0000-0001-5822-1744; Schott, Björn H. ORCID: https://orcid.org/0000-0002-8237-4481; Spottke, Annika; Spruth, Eike J. ORCID: https://orcid.org/0000-0002-8976-7309; Synofzik, Matthis ORCID: https://orcid.org/0000-0002-2280-7273; Wiltfang, Jens ORCID: https://orcid.org/0000-0003-1492-5330; Jessen, Frank; Teipel, Stefan J. ORCID: https://orcid.org/0000-0002-3586-3194 und Dyrba, Martin ORCID: https://orcid.org/0000-0002-3353-3167
(2025):
A computational ontology framework for the synthesis of multi-level pathology reports from brain MRI scans.
In: Journal of Alzheimer’s Disease [Forthcoming]
Abstract
Background: Convolutional neural network (CNN) based volumetry of MRI data can help differentiate Alzheimer's disease (AD) and the behavioral variant of frontotemporal dementia (bvFTD) as causes of cognitive decline and dementia. However, existing CNN-based MRI volumetry tools lack a structured hierarchical representation of brain anatomy, which would allow for aggregating regional pathological information and automated computational inference.
Objective: Develop a computational ontology pipeline for quantifying hierarchical pathological abnormalities and visualize summary charts for brain atrophy findings, aiding differential diagnosis.
Methods: Using FastSurfer, we segmented brain regions and measured volume and cortical thickness from MRI scans pooled across multiple cohorts (N = 3433; ADNI, AIBL, DELCODE, DESCRIBE, EDSD, and NIFD), including healthy controls, prodromal and clinical AD cases, and bvFTD cases. Employing the Web Ontology Language (OWL), we built a semantic model encoding hierarchical anatomical information. Additionally, we created summary visualizations based on sunburst plots for visual inspection of the information stored in the ontology.
Results: Our computational framework dynamically estimated and aggregated regional pathological deviations across different levels of neuroanatomy abstraction. The disease similarity index derived from the volumetric and cortical thickness deviations achieved an AUC of 0.88 for separating AD and bvFTD, which was also reflected by distinct atrophy profile visualizations.
Conclusions: The proposed automated pipeline facilitates visual comparison of atrophy profiles across various disease types and stages. It provides a generalizable computational framework for summarizing pathologic findings, potentially enhancing the physicians’ ability to evaluate brain pathologies robustly and interpretably.
| Dokumententyp: | Zeitschriftenartikel |
|---|---|
| Fakultät: | Medizin > Munich Cluster for Systems Neurology (SyNergy)
Medizin > Institut für Schlaganfall- und Demenzforschung (ISD) Medizin > Klinikum der LMU München > Neurologische Klinik und Poliklinik mit Friedrich-Baur-Institut Medizin > Klinikum der LMU München > Klinik und Poliklink für Psychiatrie und Psychotherapie Medizin > Klinikum der LMU München > Klinik und Poliklinik für Radiologie |
| Themengebiete: | 600 Technik, Medizin, angewandte Wissenschaften > 610 Medizin und Gesundheit |
| ISSN: | 1387-2877 |
| Bemerkung: | with ADNI, AIBL, and FTLDNI study groups |
| Sprache: | Englisch |
| Dokumenten ID: | 126048 |
| Datum der Veröffentlichung auf Open Access LMU: | 26. Mai 2025 16:57 |
| Letzte Änderungen: | 26. Mai 2025 16:57 |
| DFG: | Gefördert durch die Deutsche Forschungsgemeinschaft (DFG) - 390857198 |
- Dokument bearbeiten
