A Pipeline for the Diagnosis and Classification of Lung Lesions for Patients with COVID-19

Oleksandr Davydko; Olena Horodetska; Ievgen Nastenko; Yaroslav Hladkyi; Vladimir Pavlov; Luca Longo; Mykola Linnik; Oleksandr Galkin

doi:10.1109/CSIT56902.2022.10000435

A Pipeline for the Diagnosis and Classification of Lung Lesions for Patients with COVID-19

Oleksandr Davydko, Olena Horodetska, Ievgen Nastenko, Yaroslav Hladkyi, Vladimir Pavlov, Luca Longo, Mykola Linnik, Oleksandr Galkin

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

Abstract

The current study considers the development of a 5-layer pipeline for identifying and classifying COVID-19-induced lung lesions. Such system is multilayer, built upon convolutional and fully connected neural networks and logistic self-organised forest built using the group method of data handling (GMDH) principles. This pipeline includes a mechanism for finding lesions regions in lungs computer tomography images and for calculating related lung damage volume. The layer for finding images with lesions reached a Matthews Correlation Coefficient of 0.98. The layer for lesions segmentation reached a Dice similarity coefficient of 0.74, while the layer for lesions classification reached Fl-scores of 1, 0.95, 0.93 respectively for the ground-glass, opacity, crazy-paving and consolidation lesion type. Results demonstrate the effectiveness of the implemented multi-layer system in solving tasks of lesions identification and classification while being composed into a single pipeline.

Original language	English
Title of host publication	IEEE 17th International Conference on Computer Science and Information Technologies, CSIT 2022 - Proceedings
Publisher	Institute of Electrical and Electronics Engineers Inc.
Pages	551-554
Number of pages	4
ISBN (Electronic)	9798350334319
DOIs	https://doi.org/10.1109/CSIT56902.2022.10000435
Publication status	Published - 2022
Event	17th IEEE International Conference on Computer Science and Information Technologies, CSIT 2022 - Lviv, Ukraine Duration: 10 Nov 2022 → 12 Nov 2022

Publication series

Name	International Scientific and Technical Conference on Computer Sciences and Information Technologies
Volume	2022-November
ISSN (Print)	2766-3655
ISSN (Electronic)	2766-3639

Conference

Conference	17th IEEE International Conference on Computer Science and Information Technologies, CSIT 2022
Country/Territory	Ukraine
City	Lviv
Period	10/11/22 → 12/11/22

Keywords

classification
COVID-19
logistic self-organized forest
neural network
segmentation
texture analysis.

Access to Document

10.1109/CSIT56902.2022.10000435Licence: Unspecified

Cite this

Davydko, O., Horodetska, O., Nastenko, I., Hladkyi, Y., Pavlov, V., Longo, L., Linnik, M., & Galkin, O. (2022). A Pipeline for the Diagnosis and Classification of Lung Lesions for Patients with COVID-19. In IEEE 17th International Conference on Computer Science and Information Technologies, CSIT 2022 - Proceedings (pp. 551-554). (International Scientific and Technical Conference on Computer Sciences and Information Technologies; Vol. 2022-November). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/CSIT56902.2022.10000435

Davydko, Oleksandr ; Horodetska, Olena ; Nastenko, Ievgen et al. / A Pipeline for the Diagnosis and Classification of Lung Lesions for Patients with COVID-19. IEEE 17th International Conference on Computer Science and Information Technologies, CSIT 2022 - Proceedings. Institute of Electrical and Electronics Engineers Inc., 2022. pp. 551-554 (International Scientific and Technical Conference on Computer Sciences and Information Technologies).

@inproceedings{26b1526c3f7646d8a1c4eedd3c0d0f28,

title = "A Pipeline for the Diagnosis and Classification of Lung Lesions for Patients with COVID-19",

abstract = "The current study considers the development of a 5-layer pipeline for identifying and classifying COVID-19-induced lung lesions. Such system is multilayer, built upon convolutional and fully connected neural networks and logistic self-organised forest built using the group method of data handling (GMDH) principles. This pipeline includes a mechanism for finding lesions regions in lungs computer tomography images and for calculating related lung damage volume. The layer for finding images with lesions reached a Matthews Correlation Coefficient of 0.98. The layer for lesions segmentation reached a Dice similarity coefficient of 0.74, while the layer for lesions classification reached Fl-scores of 1, 0.95, 0.93 respectively for the ground-glass, opacity, crazy-paving and consolidation lesion type. Results demonstrate the effectiveness of the implemented multi-layer system in solving tasks of lesions identification and classification while being composed into a single pipeline.",

keywords = "classification, COVID-19, logistic self-organized forest, neural network, segmentation, texture analysis.",

author = "Oleksandr Davydko and Olena Horodetska and Ievgen Nastenko and Yaroslav Hladkyi and Vladimir Pavlov and Luca Longo and Mykola Linnik and Oleksandr Galkin",

note = "Publisher Copyright: {\textcopyright} 2022 IEEE.; 17th IEEE International Conference on Computer Science and Information Technologies, CSIT 2022 ; Conference date: 10-11-2022 Through 12-11-2022",

year = "2022",

doi = "10.1109/CSIT56902.2022.10000435",

language = "English",

series = "International Scientific and Technical Conference on Computer Sciences and Information Technologies",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

pages = "551--554",

booktitle = "IEEE 17th International Conference on Computer Science and Information Technologies, CSIT 2022 - Proceedings",

address = "United States",

}

Davydko, O, Horodetska, O, Nastenko, I, Hladkyi, Y, Pavlov, V, Longo, L, Linnik, M & Galkin, O 2022, A Pipeline for the Diagnosis and Classification of Lung Lesions for Patients with COVID-19. in IEEE 17th International Conference on Computer Science and Information Technologies, CSIT 2022 - Proceedings. International Scientific and Technical Conference on Computer Sciences and Information Technologies, vol. 2022-November, Institute of Electrical and Electronics Engineers Inc., pp. 551-554, 17th IEEE International Conference on Computer Science and Information Technologies, CSIT 2022, Lviv, Ukraine, 10/11/22. https://doi.org/10.1109/CSIT56902.2022.10000435

A Pipeline for the Diagnosis and Classification of Lung Lesions for Patients with COVID-19. / Davydko, Oleksandr; Horodetska, Olena; Nastenko, Ievgen et al.
IEEE 17th International Conference on Computer Science and Information Technologies, CSIT 2022 - Proceedings. Institute of Electrical and Electronics Engineers Inc., 2022. p. 551-554 (International Scientific and Technical Conference on Computer Sciences and Information Technologies; Vol. 2022-November).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - A Pipeline for the Diagnosis and Classification of Lung Lesions for Patients with COVID-19

AU - Davydko, Oleksandr

AU - Horodetska, Olena

AU - Nastenko, Ievgen

AU - Hladkyi, Yaroslav

AU - Pavlov, Vladimir

AU - Longo, Luca

AU - Linnik, Mykola

AU - Galkin, Oleksandr

PY - 2022

Y1 - 2022

N2 - The current study considers the development of a 5-layer pipeline for identifying and classifying COVID-19-induced lung lesions. Such system is multilayer, built upon convolutional and fully connected neural networks and logistic self-organised forest built using the group method of data handling (GMDH) principles. This pipeline includes a mechanism for finding lesions regions in lungs computer tomography images and for calculating related lung damage volume. The layer for finding images with lesions reached a Matthews Correlation Coefficient of 0.98. The layer for lesions segmentation reached a Dice similarity coefficient of 0.74, while the layer for lesions classification reached Fl-scores of 1, 0.95, 0.93 respectively for the ground-glass, opacity, crazy-paving and consolidation lesion type. Results demonstrate the effectiveness of the implemented multi-layer system in solving tasks of lesions identification and classification while being composed into a single pipeline.

AB - The current study considers the development of a 5-layer pipeline for identifying and classifying COVID-19-induced lung lesions. Such system is multilayer, built upon convolutional and fully connected neural networks and logistic self-organised forest built using the group method of data handling (GMDH) principles. This pipeline includes a mechanism for finding lesions regions in lungs computer tomography images and for calculating related lung damage volume. The layer for finding images with lesions reached a Matthews Correlation Coefficient of 0.98. The layer for lesions segmentation reached a Dice similarity coefficient of 0.74, while the layer for lesions classification reached Fl-scores of 1, 0.95, 0.93 respectively for the ground-glass, opacity, crazy-paving and consolidation lesion type. Results demonstrate the effectiveness of the implemented multi-layer system in solving tasks of lesions identification and classification while being composed into a single pipeline.

KW - classification

KW - COVID-19

KW - logistic self-organized forest

KW - neural network

KW - segmentation

KW - texture analysis.

UR - http://www.scopus.com/inward/record.url?scp=85146350103&partnerID=8YFLogxK

U2 - 10.1109/CSIT56902.2022.10000435

DO - 10.1109/CSIT56902.2022.10000435

M3 - Conference contribution

AN - SCOPUS:85146350103

T3 - International Scientific and Technical Conference on Computer Sciences and Information Technologies

SP - 551

EP - 554

BT - IEEE 17th International Conference on Computer Science and Information Technologies, CSIT 2022 - Proceedings

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - 17th IEEE International Conference on Computer Science and Information Technologies, CSIT 2022

Y2 - 10 November 2022 through 12 November 2022

ER -

Davydko O, Horodetska O, Nastenko I, Hladkyi Y, Pavlov V, Longo L et al. A Pipeline for the Diagnosis and Classification of Lung Lesions for Patients with COVID-19. In IEEE 17th International Conference on Computer Science and Information Technologies, CSIT 2022 - Proceedings. Institute of Electrical and Electronics Engineers Inc. 2022. p. 551-554. (International Scientific and Technical Conference on Computer Sciences and Information Technologies). doi: 10.1109/CSIT56902.2022.10000435

A Pipeline for the Diagnosis and Classification of Lung Lesions for Patients with COVID-19

Abstract

Publication series

Conference

Keywords

Access to Document

Fingerprint

Cite this