Unraveling the deep learning gearbox in optical coherence tomography image segmentation towards explainable artificial intelligence

Maloca, Peter M.; Müller, Philipp L.; Lee, Aaron Y.; Tufail, Adnan; Balaskas, Konstantinos; Niklaus, Stephanie; Kaiser, Pascal; Suter, Susanne; Zarranz-Ventura, Javier; Egan, Catherine; Scholl, Hendrik P. N.; Schnitzer, Tobias K.; Singer, Thomas; Hasler, Pascal W.; Denk, Nora

doi:10.1038/s42003-021-01697-y

Please use this identifier to cite or link to this item: https://doi.org/10.21256/zhaw-21902

Publication type:	Article in scientific journal
Type of review:	Peer review (publication)
Title:	Unraveling the deep learning gearbox in optical coherence tomography image segmentation towards explainable artificial intelligence
Authors:	Maloca, Peter M. Müller, Philipp L. Lee, Aaron Y. Tufail, Adnan Balaskas, Konstantinos Niklaus, Stephanie Kaiser, Pascal Suter, Susanne Zarranz-Ventura, Javier Egan, Catherine Scholl, Hendrik P. N. Schnitzer, Tobias K. Singer, Thomas Hasler, Pascal W. Denk, Nora
et. al:	No
DOI:	10.1038/s42003-021-01697-y 10.21256/zhaw-21902
Published in:	Communications Biology
Volume(Issue):	4
Issue:	1
Page(s):	170
Issue Date:	5-Feb-2021
Publisher / Ed. Institution:	Nature Publishing Group
ISSN:	2399-3642
Language:	English
Subjects:	Medical research; Molecular medicine; Deep learning; Explainable AI
Subject (DDC):	006: Special computer methods
Abstract:	Machine learning has greatly facilitated the analysis of medical data, while the internal operations usually remain intransparent. To better comprehend these opaque procedures, a convolutional neural network for optical coherence tomography image segmentation was enhanced with a Traceable Relevance Explainability (T-REX) technique. The proposed application was based on three components: ground truth generation by multiple graders, calculation of Hamming distances among graders and the machine learning algorithm, as well as a smart data visualization ('neural recording'). An overall average variability of 1.75% between the human graders and the algorithm was found, slightly minor to 2.02% among human graders. The ambiguity in ground truth had noteworthy impact on machine learning results, which could be visualized. The convolutional neural network balanced between graders and allowed for modifiable predictions dependent on the compartment. Using the proposed T-REX setup, machine learning processes could be rendered more transparent and understandable, possibly leading to optimized applications.
URI:	https://digitalcollection.zhaw.ch/handle/11475/21902
Fulltext version:	Published version
License (according to publishing contract):	CC BY 4.0: Attribution 4.0 International
Departement:	Life Sciences and Facility Management
Organisational Unit:	Institute of Computational Life Sciences (ICLS)
Appears in collections:	Publikationen Life Sciences und Facility Management

Files in This Item:

File	Description	Size	Format
2021_Maloca-etal_Deep-learning-gearbox-optical-coherence-tomography.pdf		2.54 MB	Adobe PDF	View/Open

Show full item record

Maloca, P. M., Müller, P. L., Lee, A. Y., Tufail, A., Balaskas, K., Niklaus, S., Kaiser, P., Suter, S., Zarranz-Ventura, J., Egan, C., Scholl, H. P. N., Schnitzer, T. K., Singer, T., Hasler, P. W., & Denk, N. (2021). Unraveling the deep learning gearbox in optical coherence tomography image segmentation towards explainable artificial intelligence. Communications Biology, 4(1), 170. https://doi.org/10.1038/s42003-021-01697-y

Maloca, P.M. et al. (2021) ‘Unraveling the deep learning gearbox in optical coherence tomography image segmentation towards explainable artificial intelligence’, Communications Biology, 4(1), p. 170. Available at: https://doi.org/10.1038/s42003-021-01697-y.

P. M. Maloca et al., “Unraveling the deep learning gearbox in optical coherence tomography image segmentation towards explainable artificial intelligence,” Communications Biology, vol. 4, no. 1, p. 170, Feb. 2021, doi: 10.1038/s42003-021-01697-y.

MALOCA, Peter M., Philipp L. MÜLLER, Aaron Y. LEE, Adnan TUFAIL, Konstantinos BALASKAS, Stephanie NIKLAUS, Pascal KAISER, Susanne SUTER, Javier ZARRANZ-VENTURA, Catherine EGAN, Hendrik P. N. SCHOLL, Tobias K. SCHNITZER, Thomas SINGER, Pascal W. HASLER und Nora DENK, 2021. Unraveling the deep learning gearbox in optical coherence tomography image segmentation towards explainable artificial intelligence. Communications Biology. 5 Februar 2021. Bd. 4, Nr. 1, S. 170. DOI 10.1038/s42003-021-01697-y

Maloca, Peter M., Philipp L. Müller, Aaron Y. Lee, Adnan Tufail, Konstantinos Balaskas, Stephanie Niklaus, Pascal Kaiser, et al. 2021. “Unraveling the Deep Learning Gearbox in Optical Coherence Tomography Image Segmentation towards Explainable Artificial Intelligence.” Communications Biology 4 (1): 170. https://doi.org/10.1038/s42003-021-01697-y.

Maloca, Peter M., et al. “Unraveling the Deep Learning Gearbox in Optical Coherence Tomography Image Segmentation towards Explainable Artificial Intelligence.” Communications Biology, vol. 4, no. 1, Feb. 2021, p. 170, https://doi.org/10.1038/s42003-021-01697-y.