Detection of coronary lesions in Kawasaki disease by Scaled-YOLOv4 with HarDNet backbone

IntroductionKawasaki disease (KD) may increase the risk of myocardial infarction or sudden death. In children, delayed KD diagnosis and treatment can increase coronary lesions (CLs) incidence by 25% and mortality by approximately 1%. This study focuses on the use of deep learning algorithm-based KD...

Full description

Bibliographic Details
Main Authors: Ho-Chang Kuo, Shih-Hsin Chen, Yi-Hui Chen, Yu-Chi Lin, Chih-Yung Chang, Yun-Cheng Wu, Tzai-Der Wang, Ling-Sai Chang, I-Hsin Tai, Kai-Sheng Hsieh
Format: Article
Language:English
Published: Frontiers Media S.A. 2023-01-01
Series:Frontiers in Cardiovascular Medicine
Subjects:
Online Access:https://www.frontiersin.org/articles/10.3389/fcvm.2022.1000374/full
_version_ 1797946997533048832
author Ho-Chang Kuo
Shih-Hsin Chen
Yi-Hui Chen
Yi-Hui Chen
Yu-Chi Lin
Chih-Yung Chang
Yun-Cheng Wu
Tzai-Der Wang
Ling-Sai Chang
I-Hsin Tai
I-Hsin Tai
Kai-Sheng Hsieh
author_facet Ho-Chang Kuo
Shih-Hsin Chen
Yi-Hui Chen
Yi-Hui Chen
Yu-Chi Lin
Chih-Yung Chang
Yun-Cheng Wu
Tzai-Der Wang
Ling-Sai Chang
I-Hsin Tai
I-Hsin Tai
Kai-Sheng Hsieh
author_sort Ho-Chang Kuo
collection DOAJ
description IntroductionKawasaki disease (KD) may increase the risk of myocardial infarction or sudden death. In children, delayed KD diagnosis and treatment can increase coronary lesions (CLs) incidence by 25% and mortality by approximately 1%. This study focuses on the use of deep learning algorithm-based KD detection from cardiac ultrasound images.MethodsSpecifically, object detection for the identification of coronary artery dilatation and brightness of left and right coronary artery is proposed and different AI algorithms were compared. In infants and young children, a dilated coronary artery is only 1-2 mm in diameter than a normal one, and its ultrasound images demonstrate a large amount of noise background-this can be a considerable challenge for image recognition. This study proposes a framework, named Scaled-YOLOv4-HarDNet, integrating the recent Scaled-YOLOv4 but with the CSPDarkNet backbone replaced by the CSPHarDNet framework.ResultsThe experimental result demonstrated that the mean average precision (mAP) of Scaled-YOLOv4-HarDNet was 72.63%, higher than that of Scaled YOLOv4 and YOLOv5 (70.05% and 69.79% respectively). In addition, it could detect small objects significantly better than Scaled-YOLOv4 and YOLOv5.ConclusionsScaled-YOLOv4-HarDNet may aid physicians in detecting KD and determining the treatment approach. Because relatively few artificial intelligence solutions about images for KD detection have been reported thus far, this paper is expected to make a substantial academic and clinical contribution.
first_indexed 2024-04-10T21:20:55Z
format Article
id doaj.art-9b4b894d1cb94442920c0040744629cf
institution Directory Open Access Journal
issn 2297-055X
language English
last_indexed 2024-04-10T21:20:55Z
publishDate 2023-01-01
publisher Frontiers Media S.A.
record_format Article
series Frontiers in Cardiovascular Medicine
spelling doaj.art-9b4b894d1cb94442920c0040744629cf2023-01-20T05:56:33ZengFrontiers Media S.A.Frontiers in Cardiovascular Medicine2297-055X2023-01-01910.3389/fcvm.2022.10003741000374Detection of coronary lesions in Kawasaki disease by Scaled-YOLOv4 with HarDNet backboneHo-Chang Kuo0Shih-Hsin Chen1Yi-Hui Chen2Yi-Hui Chen3Yu-Chi Lin4Chih-Yung Chang5Yun-Cheng Wu6Tzai-Der Wang7Ling-Sai Chang8I-Hsin Tai9I-Hsin Tai10Kai-Sheng Hsieh11Department of Pediatrics, Kawasaki Disease Center, Kaohsiung Chang Gung Memorial Hospital, College of Medicine, Chang Gung University, Kaohsiung, TaiwanDepartment of Computer Science and Information Engineering, Tamkang University, New Taipei City, TaiwanDepartment of Pediatrics, Kawasaki Disease Center, Kaohsiung Chang Gung Memorial Hospital, College of Medicine, Chang Gung University, Kaohsiung, TaiwanDepartment of Information Management, Chang Gung University, Kaohsiung, TaiwanDepartment of Pediatrics, Kawasaki Disease Center, Kaohsiung Chang Gung Memorial Hospital, College of Medicine, Chang Gung University, Kaohsiung, TaiwanDepartment of Computer Science and Information Engineering, Tamkang University, New Taipei City, TaiwanDepartment of Computer Science and Information Engineering, Tamkang University, New Taipei City, TaiwanDepartment of E-Sport Technology Management, Cheng Shiu University, Kaohsiung, TaiwanDepartment of Pediatrics, Kawasaki Disease Center, Kaohsiung Chang Gung Memorial Hospital, College of Medicine, Chang Gung University, Kaohsiung, TaiwanDepartment of Medicine, College of Medicine, China Medical University, Taichung, TaiwanDepartment of Pediatric Cardiology, China Medical University Children's Hospital, China Medical University, Taichung, TaiwanCenter of Structure and Congenital Heart Disease/Ultrasound and Department of Cardiology, Children's Hospital, China Medical University, Taichung, TaiwanIntroductionKawasaki disease (KD) may increase the risk of myocardial infarction or sudden death. In children, delayed KD diagnosis and treatment can increase coronary lesions (CLs) incidence by 25% and mortality by approximately 1%. This study focuses on the use of deep learning algorithm-based KD detection from cardiac ultrasound images.MethodsSpecifically, object detection for the identification of coronary artery dilatation and brightness of left and right coronary artery is proposed and different AI algorithms were compared. In infants and young children, a dilated coronary artery is only 1-2 mm in diameter than a normal one, and its ultrasound images demonstrate a large amount of noise background-this can be a considerable challenge for image recognition. This study proposes a framework, named Scaled-YOLOv4-HarDNet, integrating the recent Scaled-YOLOv4 but with the CSPDarkNet backbone replaced by the CSPHarDNet framework.ResultsThe experimental result demonstrated that the mean average precision (mAP) of Scaled-YOLOv4-HarDNet was 72.63%, higher than that of Scaled YOLOv4 and YOLOv5 (70.05% and 69.79% respectively). In addition, it could detect small objects significantly better than Scaled-YOLOv4 and YOLOv5.ConclusionsScaled-YOLOv4-HarDNet may aid physicians in detecting KD and determining the treatment approach. Because relatively few artificial intelligence solutions about images for KD detection have been reported thus far, this paper is expected to make a substantial academic and clinical contribution.https://www.frontiersin.org/articles/10.3389/fcvm.2022.1000374/fullKawasaki diseaseechocardiographydeep learningobject detectionScaled-YOLOv4HarDNet
spellingShingle Ho-Chang Kuo
Shih-Hsin Chen
Yi-Hui Chen
Yi-Hui Chen
Yu-Chi Lin
Chih-Yung Chang
Yun-Cheng Wu
Tzai-Der Wang
Ling-Sai Chang
I-Hsin Tai
I-Hsin Tai
Kai-Sheng Hsieh
Detection of coronary lesions in Kawasaki disease by Scaled-YOLOv4 with HarDNet backbone
Frontiers in Cardiovascular Medicine
Kawasaki disease
echocardiography
deep learning
object detection
Scaled-YOLOv4
HarDNet
title Detection of coronary lesions in Kawasaki disease by Scaled-YOLOv4 with HarDNet backbone
title_full Detection of coronary lesions in Kawasaki disease by Scaled-YOLOv4 with HarDNet backbone
title_fullStr Detection of coronary lesions in Kawasaki disease by Scaled-YOLOv4 with HarDNet backbone
title_full_unstemmed Detection of coronary lesions in Kawasaki disease by Scaled-YOLOv4 with HarDNet backbone
title_short Detection of coronary lesions in Kawasaki disease by Scaled-YOLOv4 with HarDNet backbone
title_sort detection of coronary lesions in kawasaki disease by scaled yolov4 with hardnet backbone
topic Kawasaki disease
echocardiography
deep learning
object detection
Scaled-YOLOv4
HarDNet
url https://www.frontiersin.org/articles/10.3389/fcvm.2022.1000374/full
work_keys_str_mv AT hochangkuo detectionofcoronarylesionsinkawasakidiseasebyscaledyolov4withhardnetbackbone
AT shihhsinchen detectionofcoronarylesionsinkawasakidiseasebyscaledyolov4withhardnetbackbone
AT yihuichen detectionofcoronarylesionsinkawasakidiseasebyscaledyolov4withhardnetbackbone
AT yihuichen detectionofcoronarylesionsinkawasakidiseasebyscaledyolov4withhardnetbackbone
AT yuchilin detectionofcoronarylesionsinkawasakidiseasebyscaledyolov4withhardnetbackbone
AT chihyungchang detectionofcoronarylesionsinkawasakidiseasebyscaledyolov4withhardnetbackbone
AT yunchengwu detectionofcoronarylesionsinkawasakidiseasebyscaledyolov4withhardnetbackbone
AT tzaiderwang detectionofcoronarylesionsinkawasakidiseasebyscaledyolov4withhardnetbackbone
AT lingsaichang detectionofcoronarylesionsinkawasakidiseasebyscaledyolov4withhardnetbackbone
AT ihsintai detectionofcoronarylesionsinkawasakidiseasebyscaledyolov4withhardnetbackbone
AT ihsintai detectionofcoronarylesionsinkawasakidiseasebyscaledyolov4withhardnetbackbone
AT kaishenghsieh detectionofcoronarylesionsinkawasakidiseasebyscaledyolov4withhardnetbackbone