Deep learning-based cell segmentation for rapid optical cytopathology of thyroid cancer

Jermain, Peter R; Oswald, Martin; Langdun, Tenzin; Wright, Santana; Khan, Ashraf; Stadelmann, Thilo; Abdulkadir, Ahmed; Yaroslavsky, Anna N.

doi:10.1038/s41598-024-64855-2

Please use this identifier to cite or link to this item: https://doi.org/10.21256/zhaw-31147

Publication type:	Article in scientific journal
Type of review:	Peer review (publication)
Title:	Deep learning-based cell segmentation for rapid optical cytopathology of thyroid cancer
Authors:	Jermain, Peter R Oswald, Martin Langdun, Tenzin Wright, Santana Khan, Ashraf Stadelmann, Thilo Abdulkadir, Ahmed Yaroslavsky, Anna N.
et. al:	No
DOI:	10.1038/s41598-024-64855-2 10.21256/zhaw-31147
Published in:	Scientific Reports
Volume(Issue):	14
Issue:	1
Page(s):	16389
Issue Date:	16-Jul-2024
Publisher / Ed. Institution:	Nature
ISSN:	2045-2322
Language:	English
Subjects:	Automated cell segmentation; Cytopathology; Fluorescence polarization; Methylene blue; Semantic segmentation; Thyroid cancer; Humans; Methylene Blue; Image Processing, Computer-Assisted; Neural Networks, Computer; Thyroid Gland; Cytology; Deep Learning; Thyroid Neoplasms
Subject (DDC):	006: Special computer methods 616: Internal medicine and diseases
Abstract:	Fluorescence polarization (Fpol) imaging of methylene blue (MB) is a promising quantitative approach to thyroid cancer detection. Clinical translation of MB Fpol technology requires reduction of the data analysis time that can be achieved via deep learning-based automated cell segmentation with a 2D U-Net convolutional neural network. The model was trained and tested using images of pathologically diverse human thyroid cells and evaluated by comparing the number of cells selected, segmented areas, and Fpol values obtained using automated (AU) and manual (MA) data processing methods. Overall, the model segmented 15.8% more cells than the human operator. Differences in AU and MA segmented cell areas varied between - 55.2 and + 31.0%, whereas differences in Fpol values varied from - 20.7 and + 10.7%. No statistically significant differences between AU and MA derived Fpol data were observed. The largest differences in Fpol values correlated with greatest discrepancies in AU versus MA segmented cell areas. Time required for auto-processing was reduced to 10 s versus one hour required for MA data processing. Implementation of the automated cell analysis makes quantitative fluorescence polarization-based diagnosis clinically feasible.
URI:	https://digitalcollection.zhaw.ch/handle/11475/31147
Fulltext version:	Published version
License (according to publishing contract):	CC BY 4.0: Attribution 4.0 International
Departement:	School of Engineering
Organisational Unit:	Centre for Artificial Intelligence (CAI)
Appears in collections:	Publikationen School of Engineering

Files in This Item:

File	Description	Size	Format
2024_Jermain-etal_Deep-learning-cytopathology-thyroid_Nature.pdf		1.63 MB	Adobe PDF	View/Open

Show full item record

Jermain, P. R., Oswald, M., Langdun, T., Wright, S., Khan, A., Stadelmann, T., Abdulkadir, A., & Yaroslavsky, A. N. (2024). Deep learning-based cell segmentation for rapid optical cytopathology of thyroid cancer. Scientific Reports, 14(1), 16389. https://doi.org/10.1038/s41598-024-64855-2

Jermain, P.R. et al. (2024) ‘Deep learning-based cell segmentation for rapid optical cytopathology of thyroid cancer’, Scientific Reports, 14(1), p. 16389. Available at: https://doi.org/10.1038/s41598-024-64855-2.

P. R. Jermain et al., “Deep learning-based cell segmentation for rapid optical cytopathology of thyroid cancer,” Scientific Reports, vol. 14, no. 1, p. 16389, Jul. 2024, doi: 10.1038/s41598-024-64855-2.

JERMAIN, Peter R, Martin OSWALD, Tenzin LANGDUN, Santana WRIGHT, Ashraf KHAN, Thilo STADELMANN, Ahmed ABDULKADIR und Anna N. YAROSLAVSKY, 2024. Deep learning-based cell segmentation for rapid optical cytopathology of thyroid cancer. Scientific Reports. 16 Juli 2024. Bd. 14, Nr. 1, S. 16389. DOI 10.1038/s41598-024-64855-2

Jermain, Peter R, Martin Oswald, Tenzin Langdun, Santana Wright, Ashraf Khan, Thilo Stadelmann, Ahmed Abdulkadir, and Anna N. Yaroslavsky. 2024. “Deep Learning-Based Cell Segmentation for Rapid Optical Cytopathology of Thyroid Cancer.” Scientific Reports 14 (1): 16389. https://doi.org/10.1038/s41598-024-64855-2.

Jermain, Peter R., et al. “Deep Learning-Based Cell Segmentation for Rapid Optical Cytopathology of Thyroid Cancer.” Scientific Reports, vol. 14, no. 1, July 2024, p. 16389, https://doi.org/10.1038/s41598-024-64855-2.