University of Groningen
Unsupervised brain anomaly detection in MR images Botter Martins, Samuel
DOI:
10.33612/diss.144368886
IMPORTANT NOTE: You are advised to consult the publisher's version (publisher's PDF) if you wish to cite from it. Please check the document version below.
Document Version
Publisher's PDF, also known as Version of record
Publication date: 2020
Link to publication in University of Groningen/UMCG research database
Citation for published version (APA):
Botter Martins, S. (2020). Unsupervised brain anomaly detection in MR images. University of Groningen. https://doi.org/10.33612/diss.144368886
Copyright
Other than for strictly personal use, it is not permitted to download or to forward/distribute the text or part of it without the consent of the author(s) and/or copyright holder(s), unless the work is under an open content license (like Creative Commons).
Take-down policy
If you believe that this document breaches copyright please contact us providing details, and we will remove access to the work immediately and investigate your claim.
Downloaded from the University of Groningen/UMCG research database (Pure): http://www.rug.nl/research/portal. For technical reasons the number of authors shown on this cover page is limited to 10 maximum.
Propositions
• Deviations from the normal pattern of structural brain asymmetries are useful insights of neurological pathologies.
• Automatic detection of abnormal brain asymmetries supports neurologists during medical diagnosis, surgical planning, and treatment assessment.
• Unsupervised brain asymmetry detection methods are generic in detecting any lesions, e.g., coming from multiple diseases, as long as these notably differ from healthy training samples.
• Brain image segmentation supports automatic asymmetry detection by removing non-brain tissues (e.g., skull, eyes, and neck) during analysis.
• Convolutional Autoencoders can model normal hippocampal asymmetries from 3D patches of healthy subjects to detect abnormal asymmetries.
• Supervoxels provide meaningful regions of interest that fit lesions and tissues, with minimum heterogeneous information.
• Using specialized one-class per-supervoxel classifiers for each patient image, trained from texture features (asymmetries), can detect abnormal asymmetries accurately.
• Modeling normal patterns of image registration errors from healthy subjects can be useful to detect outliers associated with symmetric and asymmetric brain lesions.
