Automated brain region recognition based on elastic registration and atlas mapping
-
1
University of Antwerp, Department of Veterinary Sciences, Belgium
-
2
Janssen Pharmaceutical companies of Johnson & Johnson, Neuroscience Department, Belgium
-
3
Ghent University, Department of Molecular Biotechnology, Belgium
Brain tissue slices represent physiologically faithful models for analyzing pathological features in an anatomically defined manner. With throughput getting larger, fully automated analysis has become essential. A bottleneck in this context is the automated recognition and annotation of selected brain regions. To meet this demand, we are developing an automated image analysis pipeline (schematically shown in Fig. 1) for contextual quantification in the brain by mapping anatomical information onto microtome-cut brain slices. To this end, DAPI-stained brain slices are shape-normalized to a reference section by means of an elastic registration algorithm that is constrained by both intensity and shape information. The reference slices in turn, are mapped to manually or automatically assigned brain regions; the latter using information extracted from the openly available Allen Brain Atlas. A robust error metric is conceived to estimate registration performance, to optimize its parameters, and to automatically remove badly cut slices. The quality of our method is quantified by comparing the automatically defined brain tissue regions to the ground truth (manually segmented brain slices) of both registered and reference slices.
Résumé en Français:
Lors de l’étude des modèles animaux de maladies du système nerveux, l’analyse des modifications anatomiques induites par la pathologie est une étape importante mais fastidieuse car reposant sur l’observation de milliers de coupes et l’identification des structures anatomiques. Dans ce travail, nous avons mis au point un programme qui permet de déterminer de manière automatique toutes les structures anatomiques à partir de coupes histologiques. Grâce à un algorithme mathématique les coupes sont superposées à des coupes modèles annotées sur base d’un atlas référencé. Cette technique permettra d’analyser beaucoup plus rapidement les changements anatomiques induits par différentes pathologies du système nerveux.
Samenvatting in het Nederlands:
"In ziekten van het zenuwstelsel, zoals o.a. de ziekte van Alzheimer, treden “zieke” zenuwcellen veelal eerst op in bepaalde brein-regio’s. Om deze ziekten te onderzoeken maakt men veelal gebruik van scans van de hersenen of hersencoupes. Hierbij is het correct detecteren en onderscheiden van de verschillende regio’s van de hersenen erg belangrijk. Verder is, met het oog op een snelle en objectieve analyse van dergelijke scans, een geautomatiseerde regio-herkenning wenselijk. In dit onderzoek hebben we zulke methode ontwikkeld, toegespitst op beelden van fluorescente hersencoupes van muizen. Deze methode laat toe de beeldanalyse in onderzoek op muizen te versnellen door dit knelpunt te verwijderen."
Keywords:
Brain Mapping,
Elastic warping,
Brain Atlas,
brain tissue slices,
segmentation,
image analysis
Conference:
6th Belgian Brain Congress, MONS, Belgium, 8 Oct - 8 Oct, 2016.
Presentation Type:
Poster Presentation
Topic:
Brain and brain diseases: between heredity and environment
Citation:
Barbier
M,
Bottelbergs
A,
Verboven
R,
Nuydens
R,
Ebneth
A and
De Vos
WH
(2016). Automated brain region recognition based on elastic registration and atlas mapping.
Conference Abstract:
6th Belgian Brain Congress.
doi: 10.3389/conf.fnagi.2016.03.00064
Copyright:
The abstracts in this collection have not been subject to any Frontiers peer review or checks, and are not endorsed by Frontiers.
They are made available through the Frontiers publishing platform as a service to conference organizers and presenters.
The copyright in the individual abstracts is owned by the author of each abstract or his/her employer unless otherwise stated.
Each abstract, as well as the collection of abstracts, are published under a Creative Commons CC-BY 4.0 (attribution) licence (https://creativecommons.org/licenses/by/4.0/) and may thus be reproduced, translated, adapted and be the subject of derivative works provided the authors and Frontiers are attributed.
For Frontiers’ terms and conditions please see https://www.frontiersin.org/legal/terms-and-conditions.
Received:
01 Jul 2016;
Published Online:
13 Jul 2016.
*
Correspondence:
Dr. Michaël Barbier, University of Antwerp, Department of Veterinary Sciences, Antwerp, Antwerp, Belgium, michael.barbier@uantwerpen.be