Deep learning-based detection of motion artifacts in probe-based confocal laser endomicroscopy images.

Journal article
(Original article)


Publication Details

Author(s): Aubreville M, Stöve M, Oetter N, Goncalves M, Knipfer C, Neumann H, Bohr C, Stelzle F, Maier A
Journal: International Journal of Computer Assisted Radiology and Surgery
Publication year: 2019
Volume: 14
Journal issue: 1
Pages range: 31-42
ISSN: 1861-6410
Language: English


Abstract

Probe-based confocal laser endomicroscopy (pCLE) is a subcellular in vivo imaging technique capable of producing images that enable diagnosis of malign structural modifications in epithelial tissue. Images acquired with pCLE are, however, often tainted by significant artifacts that impair diagnosis. This is especially detrimental for automated image analysis, which is why said images are often excluded from recognition pipelines.\nWe present an approach for the automatic detection of motion artifacts in pCLE images and apply this methodology to a data set of 15 thousand images of epithelial tissue acquired in the oral cavity and the vocal folds. The approach is based on transfer learning from intermediate endpoints within a pre-trained Inception v3 network with tailored preprocessing. For detection within the non-rectangular pCLE images, we perform pooling within the activation maps of the network and evaluate this at different network depths.\nWe achieved area under the ROC curve values of 0.92 with the proposed method, compared to 0.80 for the best feature-based machine learning approach. Our overall accuracy with the presented approach is 94.8%.\nOver traditional machine learning approaches with state-of-the-art features, we achieved significantly improved overall performance.\nPURPOSE\nMETHODS\nRESULTS\nCONCLUSION


FAU Authors / FAU Editors

Aubreville, Marc
Lehrstuhl für Informatik 5 (Mustererkennung)
Maier, Andreas Prof. Dr.-Ing.
Lehrstuhl für Informatik 5 (Mustererkennung)
Oetter, Nicolai
Mund-, Kiefer- und Gesichtschirurgische Klinik
Stelzle, Florian Prof. Dr. Dr.
Medizinische Fakultät
Stöve, Maike
Lehrstuhl für Informatik 14 (Maschinelles Lernen und Datenanalytik)


External institutions with authors

Johannes Gutenberg-Universität Mainz
Universitätsklinikum Hamburg-Eppendorf (UKE)
Universitätsklinikum Regensburg


How to cite

APA:
Aubreville, M., Stöve, M., Oetter, N., Goncalves, M., Knipfer, C., Neumann, H.,... Maier, A. (2019). Deep learning-based detection of motion artifacts in probe-based confocal laser endomicroscopy images. International Journal of Computer Assisted Radiology and Surgery, 14(1), 31-42. https://dx.doi.org/10.1007/s11548-018-1836-1

MLA:
Aubreville, Marc, et al. "Deep learning-based detection of motion artifacts in probe-based confocal laser endomicroscopy images." International Journal of Computer Assisted Radiology and Surgery 14.1 (2019): 31-42.

BibTeX: 

Last updated on 2019-15-03 at 13:38

Share link