A pipeline for the Diagnosis and Classification of Lung Lesions for patients with COVID-19

被引：0

作者：

Davydko, Oleksandr ^{[1
]}

Hladkyi, Yaroslav ^{[1
]}

Linnik, Mykola ^{[2
]}

Horodetska, Olena ^{[1
]}

Pavlov, Vladimir ^{[1
]}

Galkin, Oleksandr ^{[1
]}

Nastenko, Ievgen ^{[3
]}

Longo, Luca ^{[4
]}

机构：

[1] Natl Tech Univ Ukraine, Igor Sikorsky Kyiv Polytech Inst, Dept Biomed Cybernet, Kiev, Ukraine

[2] NAMS Ukraine, FG Yanovsky Natl Inst Phthisiol & Pulmonol, Dept Epidemiol & Org Problems Phthisiol, Kiev, Ukraine

[3] Natl Tech Univ Ukraine, Igor Sikorsky Kyiv Polytech Inst, Dept Biomed Cybernet, Amosov Natl Inst Cardiovasc Surg, Kiev, Ukraine

[4] Technol Univ Dublin, Sch Comp Sci, AICL Lab, Dublin, Ireland

来源：

2022 IEEE 17TH INTERNATIONAL CONFERENCE ON COMPUTER SCIENCES AND INFORMATION TECHNOLOGIES (CSIT) | 2022年

关键词：

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

D O I：

10.1109/CSIT56902.2022.10000435

中图分类号：

TP18 [人工智能理论];

学科分类号：

081104 ; 0812 ; 0835 ; 1405 ;

摘要：

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 F1-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.

引用

页码：551 / 554

页数：4

共 15 条

[1] Diagnosis of COVID-19 in CT image using CNN and XGBoost [J].

Carvalho, Edelson Damasceno ;

Carvalho, Edson Damasceno ;

de Carvalho Filho, Antonio Oseas ;

Duarte de Araujo, Flavio Henrique ;

Lira Rabelo, Ricardo de Andrade .

2020 IEEE SYMPOSIUM ON COMPUTERS AND COMMUNICATIONS (ISCC), 2020, :811-816

[2]

Davydko O., 2021, 2021 IEEE 16 INT C C

[3] Automatic lung segmentation in routine imaging is primarily a data diversity problem, not a methodology problem [J].

Hofmanninger, Johannes ;

Prayer, Forian ;

Pan, Jeanny ;

Roehrich, Sebastian ;

Prosch, Helmut ;

Langs, Georg .

EUROPEAN RADIOLOGY EXPERIMENTAL, 2020, 4 (01)

[4]

Ma J., 2020, Covid-19 ct lung and infection segmentation dataset

[5]

Matviichuk O., 2021, 2021 IEEE 16 INT C C

[6]

Nastenko I., 2019, ADV INTELLIGENT SYST, V4, P535

[7] Classification of Coronavirus (COVID-19) fromX-rayandCTimages using shrunken features [J].

Ozturk, Saban ;

Ozkaya, Umut ;

Barstugan, Mucahid .

INTERNATIONAL JOURNAL OF IMAGING SYSTEMS AND TECHNOLOGY, 2021, 31 (01) :5-15

[8] U-Net: Convolutional Networks for Biomedical Image Segmentation [J].

Ronneberger, Olaf ;

Fischer, Philipp ;

Brox, Thomas .

MEDICAL IMAGE COMPUTING AND COMPUTER-ASSISTED INTERVENTION, PT III, 2015, 9351 :234-241

[9]

Shah V., 2020, PREPRINT MEDRXIV, DOI [10.1101/2020.07.11.20151332, DOI 10.1101/2020.07.11.20151332]

[10]

Silva Pedro, 2020, Inform Med Unlocked, V20, P100427, DOI 10.1016/j.imu.2020.100427

← 1 2 →