Related Experiment Video
Updated: Aug 11, 2025
Author Spotlight: Advancing Cardiovascular Imaging - Introducing the Spatially Weighted Calcium Score for Early Disease Detection
Published on: September 22, 2023
Detection of coronary lesions in Kawasaki disease by Scaled-YOLOv4 with HarDNet backbone
Ho-Chang Kuo1, Shih-Hsin Chen2, Yi-Hui Chen1,3
1Department of Pediatrics, Kawasaki Disease Center, Kaohsiung Chang Gung Memorial Hospital, College of Medicine, Chang Gung University, Kaohsiung, Taiwan.
Insights
A new deep learning algorithm, Scaled-YOLOv4-HarDNet, shows promise for detecting Kawasaki disease (KD) in cardiac ultrasound images. This AI tool can identify coronary artery issues, potentially improving early diagnosis and treatment for children at risk of heart problems.
Area of Science:
- Cardiology
- Artificial Intelligence
- Medical Imaging
Background:
- Kawasaki disease (KD) poses risks of myocardial infarction and sudden death.
- Delayed diagnosis and treatment of KD in children increase coronary lesions (CLs) incidence by 25% and mortality by 1%.
- Accurate and timely KD detection is crucial for preventing severe cardiac complications.
Purpose of the Study:
- To develop and evaluate a deep learning algorithm for detecting Kawasaki disease (KD) from cardiac ultrasound images.
- To improve the accuracy of identifying coronary artery dilatation and brightness in pediatric patients.
- To address the challenges of image noise and small object detection in pediatric cardiac ultrasounds.
Main Methods:
- Proposed a novel framework, Scaled-YOLOv4-HarDNet, integrating Scaled-YOLOv4 with a CSPHarDNet backbone.
- Utilized object detection for identifying coronary artery dilatation and brightness.
- Compared the performance of Scaled-YOLOv4-HarDNet against Scaled YOLOv4 and YOLOv5 algorithms.
Main Results:
- Scaled-YOLOv4-HarDNet achieved a mean average precision (mAP) of 72.63%, outperforming Scaled YOLOv4 (70.05%) and YOLOv5 (69.79%).
- The proposed algorithm demonstrated superior performance in detecting small objects compared to existing methods.
- The framework effectively addressed challenges posed by noisy ultrasound images in young children.
Conclusions:
- Scaled-YOLOv4-HarDNet shows significant potential to aid physicians in KD detection and treatment planning.
- This AI-driven approach offers a novel solution for KD diagnosis using cardiac imaging.
- The study is expected to make a substantial academic and clinical contribution to the field of pediatric cardiology and AI in medicine.
Introduction:
Kawasaki 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.
Methods:
Specifically, 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.
Results:
The 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.
Conclusions:
Scaled-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.
Related Concept Videos
01:30Coronary Artery Disease I: Introduction
01:30Acute Coronary Syndrome III: Diagnostic Studies
01:25Imaging Studies for Cardiovascular System VI: Calcium -Scoring CT
01:26Coronary Artery Disease II: Pathophysiology
01:26Coronary Artery Disease IV: Preventive Measures
01:30Coronary Artery Disease III: Clinical Manifestations