Application of Artificial Neural Networks for Analysis of Ice Recrystallization Process for Cryopreservation

Publikation: Beitrag in Buch/Bericht/Sammelwerk/KonferenzbandAufsatz in KonferenzbandForschungPeer-Review

Autoren

  • Maksym Tymkovych
  • Oleksandr Gryshkov
  • Karina Selivanova
  • Vitalii Mutsenko
  • Oleg Avrunin
  • Birgit Glasmacher

Organisationseinheiten

Externe Organisationen

  • Kharkov National University of Radio Electronics
Forschungs-netzwerk anzeigen

Details

OriginalspracheEnglisch
Titel des Sammelwerks8th European Medical and Biological Engineering Conference
UntertitelProceedings of the EMBEC 2020
Herausgeber/-innenTomaz Jarm, Aleksandra Cvetkoska, Samo Mahnič-Kalamiza, Damijan Miklavcic
Herausgeber (Verlag)Springer Science and Business Media Deutschland GmbH
Seiten102-111
Seitenumfang10
ISBN (elektronisch)978-3-030-64610-3
ISBN (Print)9783030646097
PublikationsstatusVeröffentlicht - 30 Nov. 2020
Veranstaltung8th European Medical and Biological Engineering Conference, EMBEC 2020 - Portorož, Slowenien
Dauer: 29 Nov. 20203 Dez. 2020

Publikationsreihe

NameIFMBE Proceedings
Band80
ISSN (Print)1680-0737
ISSN (elektronisch)1433-9277

Abstract

Cryomicroscopy is one of the main techniques to visualize freezing and thawing events taking place during cryopreservation of cells, native and artificial tissues with the ultimate goal to provide cell- and tissue-specific cryogenic preservation. However, there is currently no unified software tool for the automated analysis of ice recrystallization kinetics for a variety of cryoprotective agents used in the cryobiological practice. In this regard, we focused on the particular aspect of image analysis in the course of ice recrystallization, i.e. the possibility of using a neural network for the segmentation of ice crystals during isothermal annealing. In the work, the U-Net deep neural network was used for segmentation of ice crystals on cryomicroscopic images. Using 100 images as training set, the resulting accuracy of ice crystal segmentation was about 74% on the test sample (30 images). The obtained results show the possibility of segmentation of ice crystals in cryomicroscopic images taking into account the overlapping of intensity levels of an object and background. Further improvement of the model through the use of an additional training data as well as augmentation techniques is required to more efficiently validate this approach.

ASJC Scopus Sachgebiete

Zitieren

Application of Artificial Neural Networks for Analysis of Ice Recrystallization Process for Cryopreservation. / Tymkovych, Maksym; Gryshkov, Oleksandr; Selivanova, Karina et al.
8th European Medical and Biological Engineering Conference: Proceedings of the EMBEC 2020. Hrsg. / Tomaz Jarm; Aleksandra Cvetkoska; Samo Mahnič-Kalamiza; Damijan Miklavcic. Springer Science and Business Media Deutschland GmbH, 2020. S. 102-111 (IFMBE Proceedings; Band 80).

Publikation: Beitrag in Buch/Bericht/Sammelwerk/KonferenzbandAufsatz in KonferenzbandForschungPeer-Review

Tymkovych, M, Gryshkov, O, Selivanova, K, Mutsenko, V, Avrunin, O & Glasmacher, B 2020, Application of Artificial Neural Networks for Analysis of Ice Recrystallization Process for Cryopreservation. in T Jarm, A Cvetkoska, S Mahnič-Kalamiza & D Miklavcic (Hrsg.), 8th European Medical and Biological Engineering Conference: Proceedings of the EMBEC 2020. IFMBE Proceedings, Bd. 80, Springer Science and Business Media Deutschland GmbH, S. 102-111, 8th European Medical and Biological Engineering Conference, EMBEC 2020, Portorož, Slowenien, 29 Nov. 2020. https://doi.org/10.1007/978-3-030-64610-3_13
Tymkovych, M., Gryshkov, O., Selivanova, K., Mutsenko, V., Avrunin, O., & Glasmacher, B. (2020). Application of Artificial Neural Networks for Analysis of Ice Recrystallization Process for Cryopreservation. In T. Jarm, A. Cvetkoska, S. Mahnič-Kalamiza, & D. Miklavcic (Hrsg.), 8th European Medical and Biological Engineering Conference: Proceedings of the EMBEC 2020 (S. 102-111). (IFMBE Proceedings; Band 80). Springer Science and Business Media Deutschland GmbH. https://doi.org/10.1007/978-3-030-64610-3_13
Tymkovych M, Gryshkov O, Selivanova K, Mutsenko V, Avrunin O, Glasmacher B. Application of Artificial Neural Networks for Analysis of Ice Recrystallization Process for Cryopreservation. in Jarm T, Cvetkoska A, Mahnič-Kalamiza S, Miklavcic D, Hrsg., 8th European Medical and Biological Engineering Conference: Proceedings of the EMBEC 2020. Springer Science and Business Media Deutschland GmbH. 2020. S. 102-111. (IFMBE Proceedings). doi: 10.1007/978-3-030-64610-3_13
Tymkovych, Maksym ; Gryshkov, Oleksandr ; Selivanova, Karina et al. / Application of Artificial Neural Networks for Analysis of Ice Recrystallization Process for Cryopreservation. 8th European Medical and Biological Engineering Conference: Proceedings of the EMBEC 2020. Hrsg. / Tomaz Jarm ; Aleksandra Cvetkoska ; Samo Mahnič-Kalamiza ; Damijan Miklavcic. Springer Science and Business Media Deutschland GmbH, 2020. S. 102-111 (IFMBE Proceedings).
Download
@inproceedings{41ae762a74444c6c823c57febfd491aa,
title = "Application of Artificial Neural Networks for Analysis of Ice Recrystallization Process for Cryopreservation",
abstract = "Cryomicroscopy is one of the main techniques to visualize freezing and thawing events taking place during cryopreservation of cells, native and artificial tissues with the ultimate goal to provide cell- and tissue-specific cryogenic preservation. However, there is currently no unified software tool for the automated analysis of ice recrystallization kinetics for a variety of cryoprotective agents used in the cryobiological practice. In this regard, we focused on the particular aspect of image analysis in the course of ice recrystallization, i.e. the possibility of using a neural network for the segmentation of ice crystals during isothermal annealing. In the work, the U-Net deep neural network was used for segmentation of ice crystals on cryomicroscopic images. Using 100 images as training set, the resulting accuracy of ice crystal segmentation was about 74% on the test sample (30 images). The obtained results show the possibility of segmentation of ice crystals in cryomicroscopic images taking into account the overlapping of intensity levels of an object and background. Further improvement of the model through the use of an additional training data as well as augmentation techniques is required to more efficiently validate this approach.",
keywords = "Artificial neural network, Cryomicroscopy, Cryopreservation, Ice recrystallization, Image processing, Segmentation, U-Net",
author = "Maksym Tymkovych and Oleksandr Gryshkov and Karina Selivanova and Vitalii Mutsenko and Oleg Avrunin and Birgit Glasmacher",
note = "Funding Information: Acknowledgments. The authors would like to thank the exchange program with East European Countries funded by the German Academic Exchange Service (DAAD, project number 54364768) and the joint German-Ukrainian Grant funded by Federal Ministry of Education and Research of Germany (BMBF, 01DK20017) and Ministry of Education and Science of Ukraine (project ID 100445538).; 8th European Medical and Biological Engineering Conference, EMBEC 2020 ; Conference date: 29-11-2020 Through 03-12-2020",
year = "2020",
month = nov,
day = "30",
doi = "10.1007/978-3-030-64610-3_13",
language = "English",
isbn = "9783030646097",
series = "IFMBE Proceedings",
publisher = "Springer Science and Business Media Deutschland GmbH",
pages = "102--111",
editor = "Tomaz Jarm and Aleksandra Cvetkoska and Samo Mahni{\v c}-Kalamiza and Damijan Miklavcic",
booktitle = "8th European Medical and Biological Engineering Conference",
address = "Germany",

}

Download

TY - GEN

T1 - Application of Artificial Neural Networks for Analysis of Ice Recrystallization Process for Cryopreservation

AU - Tymkovych, Maksym

AU - Gryshkov, Oleksandr

AU - Selivanova, Karina

AU - Mutsenko, Vitalii

AU - Avrunin, Oleg

AU - Glasmacher, Birgit

N1 - Funding Information: Acknowledgments. The authors would like to thank the exchange program with East European Countries funded by the German Academic Exchange Service (DAAD, project number 54364768) and the joint German-Ukrainian Grant funded by Federal Ministry of Education and Research of Germany (BMBF, 01DK20017) and Ministry of Education and Science of Ukraine (project ID 100445538).

PY - 2020/11/30

Y1 - 2020/11/30

N2 - Cryomicroscopy is one of the main techniques to visualize freezing and thawing events taking place during cryopreservation of cells, native and artificial tissues with the ultimate goal to provide cell- and tissue-specific cryogenic preservation. However, there is currently no unified software tool for the automated analysis of ice recrystallization kinetics for a variety of cryoprotective agents used in the cryobiological practice. In this regard, we focused on the particular aspect of image analysis in the course of ice recrystallization, i.e. the possibility of using a neural network for the segmentation of ice crystals during isothermal annealing. In the work, the U-Net deep neural network was used for segmentation of ice crystals on cryomicroscopic images. Using 100 images as training set, the resulting accuracy of ice crystal segmentation was about 74% on the test sample (30 images). The obtained results show the possibility of segmentation of ice crystals in cryomicroscopic images taking into account the overlapping of intensity levels of an object and background. Further improvement of the model through the use of an additional training data as well as augmentation techniques is required to more efficiently validate this approach.

AB - Cryomicroscopy is one of the main techniques to visualize freezing and thawing events taking place during cryopreservation of cells, native and artificial tissues with the ultimate goal to provide cell- and tissue-specific cryogenic preservation. However, there is currently no unified software tool for the automated analysis of ice recrystallization kinetics for a variety of cryoprotective agents used in the cryobiological practice. In this regard, we focused on the particular aspect of image analysis in the course of ice recrystallization, i.e. the possibility of using a neural network for the segmentation of ice crystals during isothermal annealing. In the work, the U-Net deep neural network was used for segmentation of ice crystals on cryomicroscopic images. Using 100 images as training set, the resulting accuracy of ice crystal segmentation was about 74% on the test sample (30 images). The obtained results show the possibility of segmentation of ice crystals in cryomicroscopic images taking into account the overlapping of intensity levels of an object and background. Further improvement of the model through the use of an additional training data as well as augmentation techniques is required to more efficiently validate this approach.

KW - Artificial neural network

KW - Cryomicroscopy

KW - Cryopreservation

KW - Ice recrystallization

KW - Image processing

KW - Segmentation

KW - U-Net

UR - http://www.scopus.com/inward/record.url?scp=85097625206&partnerID=8YFLogxK

U2 - 10.1007/978-3-030-64610-3_13

DO - 10.1007/978-3-030-64610-3_13

M3 - Conference contribution

AN - SCOPUS:85097625206

SN - 9783030646097

T3 - IFMBE Proceedings

SP - 102

EP - 111

BT - 8th European Medical and Biological Engineering Conference

A2 - Jarm, Tomaz

A2 - Cvetkoska, Aleksandra

A2 - Mahnič-Kalamiza, Samo

A2 - Miklavcic, Damijan

PB - Springer Science and Business Media Deutschland GmbH

T2 - 8th European Medical and Biological Engineering Conference, EMBEC 2020

Y2 - 29 November 2020 through 3 December 2020

ER -