TU Darmstadt / ULB / TUbiblio

Fuzzy tissue detection for real-time focal control in corneal confocal microscopy

Bartschat, Andreas and Allgeier, Stephan and Scherr, Tim and Stegmaier, Johannes and Bohn, Sebastian and Reichert, Klaus-Martin and Kuijper, Arjan and Reischl, Markus and Stachs, Oliver and Köhler, Bernd and Mikut, Ralf (2019):
Fuzzy tissue detection for real-time focal control in corneal confocal microscopy.
In: at - Automatisierungstechnik, 67 (10), pp. 879-888. ISSN 0178-2312,
DOI: 10.1515/auto-2019-0034,
[Article]

Abstract

Corneal confocal laser scanning microscopy is a promising method for in vivo investigation of cellular structures, e. g., of nerve fibers in the sub-basal nerve plexus. During recording, even slight displacements of the focal plane lead to images of adjacent tissue layers. In this work, we propose a closed-loop control of the focal plane. To detect and evaluate the visible tissues, we utilize the Bag of Visual Words approach to implement a customizable image processing pipeline for real-time applications. Furthermore, we show that the proposed model can be trained with small classification datasets and can be applied as a segmentation method. The proposed control loop, including tissue detection, is implemented in a proof-of-concept setup and shows promising results in a first evaluation with a human subject.

Item Type: Article
Erschienen: 2019
Creators: Bartschat, Andreas and Allgeier, Stephan and Scherr, Tim and Stegmaier, Johannes and Bohn, Sebastian and Reichert, Klaus-Martin and Kuijper, Arjan and Reischl, Markus and Stachs, Oliver and Köhler, Bernd and Mikut, Ralf
Title: Fuzzy tissue detection for real-time focal control in corneal confocal microscopy
Language: English
Abstract:

Corneal confocal laser scanning microscopy is a promising method for in vivo investigation of cellular structures, e. g., of nerve fibers in the sub-basal nerve plexus. During recording, even slight displacements of the focal plane lead to images of adjacent tissue layers. In this work, we propose a closed-loop control of the focal plane. To detect and evaluate the visible tissues, we utilize the Bag of Visual Words approach to implement a customizable image processing pipeline for real-time applications. Furthermore, we show that the proposed model can be trained with small classification datasets and can be applied as a segmentation method. The proposed control loop, including tissue detection, is implemented in a proof-of-concept setup and shows promising results in a first evaluation with a human subject.

Journal or Publication Title: at - Automatisierungstechnik
Journal volume: 67
Number: 10
Uncontrolled Keywords: Image classification Bag-of-words Laser scanning confocal microscopy Pattern recognition Tissue classifications Realtime feedback Tissue segmentation
Divisions: 20 Department of Computer Science
20 Department of Computer Science > Mathematical and Applied Visual Computing
Date Deposited: 17 Apr 2020 09:59
DOI: 10.1515/auto-2019-0034
Official URL: https://doi.org/10.1515/auto-2019-0034
Alternative Abstract:
Alternative abstract Language

Die Konfokalmikroskopie der Kornea ist eine vielversprechende Methode zur in vivo Untersuchung von Zellstrukturen, z. B. von Nervenfasern im subbasalen Nervenplexus. Während der Aufnahme können schon geringe Verschiebungen der Fokusebene dazu führen, dass Bildern benachbarter Gewebeschichten aufgenommen werden. In dieser Arbeit schlagen wir eine Regelung der Fokusebene vor. Zum Erkennen und Auswerten der sichtbaren Gewebe verwenden wir die Bag of Visual Words Methode zur Implementierung einer anpassbaren Bildverarbeitungspipeline für Echtzeitanwendungen. Außerdem zeigen wir, dass das vorgeschlagene Modell mit kleinen Lerndatensätzen trainiert und als Segmentierungsmethode angewendet werden kann. Der vorgeschlagene Regelkreis, einschließlich der Gewebeerkennung, wurde im Rahmen einer Machbarkeitsstudie implementiert und zeigt vielversprechende Ergebnisse in einer ersten Auswertung mit einem menschlichen Probanden.

German
Export:
Suche nach Titel in: TUfind oder in Google
Send an inquiry Send an inquiry

Options (only for editors)
Show editorial Details Show editorial Details