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

Kuo, Ho-Chang; Chen, Shih-Hsin; Chen, Yihui; Lin, Yu Chi; Chang, Chih‐Yung; Wu, Yuncheng; Wang, Tzai-Der; Chang, Ling‐Sai; Tai, I-Hsin; Hsieh, Kai‐Sheng

doi:10.3389/fcvm.2022.1000374

Cited by 5 publications

(1 citation statement)

References 21 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…From above proposed model, the result calculated from the formula of the mean average precision (mAP) of YOLO version, and also predict the result of HarDNet of 72.63%, which is larger than Scaled YOLOv. YOLOv are 70.05% and 67.79% respectively [29].…”

mentioning

confidence: 94%

Blood Vessels Disease Detection of Coronary Angiography Images using Deep learning Model

Osama,

Kumar,

Shahid

2024

Preprint

View full text Add to dashboard Cite

Presently Coronary artery disease, often caused by the narrowing of the coronary artery lumen due to atherosclerosis, is a leading cause of death. Coronary angiography also known as cardiac catheterization or X-ray angiography, is a medical procedure that uses X-ray imaging to visualize the coronary arteries, which supply blood to the heart muscle. X-ray angiography is procedure to assess the blood flow through these arteries and to identify any blockages or abnormalities. The accuracy of X-ray angiography depends on the quality of the imaging equipment as well as experience and expertise of the radiologist. Poor image quality could affect the accurate diagnosis of coronary arteries. Manual interpretation of angiography images is subjective and time consuming. In some cases, small or diffuse blockages may not be easily visible, and additional imaging techniques may be required. Therefore, early automated detection of blockage of heart vessels became necessary for detection and diagnosis. The artificial intelligence algorithms could play a vital role in this area. In this paper, a deep-learning based algorithm has been used for recognition of blockage in coronary angiographic visuals. Here, we proposed deep learning (YOLOv8) models for the detection of blockage into blood vessels coronary angiography images. In this experiment about 1934 labelled X-ray angiography images has been used from Mendeley. For Experimentation purpose, images are preprocessed and augmented. Total 80% images have been used for training and 20% images has been used for testing. The experimental results show that the measuring metrices of proposed model for detection of blood vessels blockage area in rectangular box. The performance of model represented by predicted value of Precision, recall, mean average precision (mAP) and F1 score are, 99.4%, 100%, 99.5% and 99.7% respectively.

show abstract

mentioning

confidence: 94%

Blood Vessels Disease Detection of Coronary Angiography Images using Deep learning Model

Osama,

Kumar,

Shahid

2024

Preprint

View full text Add to dashboard Cite

show abstract

YOLO Based Segmentation and CNN Based Classification Framework for Epithelial and Pus Cell Detection

Shwetha

Prasad

Mukhopadhyay

et al. 2023

Communications in Computer and Information Science

View full text Add to dashboard Cite

Progression prediction of coronary artery lesions by echocardiography-based ultrasomics analysis in Kawasaki disease

Xu,

Feng,

Cao

et al. 2024

Ital J Pediatr

View full text Add to dashboard Cite

Background Echocardiography-based ultrasomics analysis aids Kawasaki disease (KD) diagnosis but its role in predicting coronary artery lesions (CALs) progression remains unknown. We aimed to develop and validate a predictive model combining echocardiogram-based ultrasomics with clinical parameters for CALs progression in KD. Methods Total 371 KD patients with CALs at baseline were enrolled from a retrospective cohort (cohort 1, n = 316) and a prospective cohort (cohort 2, n = 55). CALs progression was defined by increased Z scores in any coronary artery branch at the 1-month follow-up. Patients in cohort 1 were split randomly into training and validation set 1 at the ratio of 6:4, while cohort 2 comprised validation set 2. Clinical parameters and ultrasomics features at baseline were analyzed and selected for models construction. Model performance was evaluated by area under the receiver operating characteristic curve (AUROC), area under the precision-recall curve (AUPRC) and decision curve analysis (DCA) in the training and two validation sets. Results At the 1-month follow-ups, 65 patients presented with CALs progression. Three clinical parameters and six ultrasomics features were selected to construct the model. The clinical-ultrasomics model exhibited a good predictive capability in the training, validation set 1 and set 2, achieving AUROCs of 0.83 (95% CI, 0.75–0.90), 0.84 (95% CI, 0.74–0.94), and 0.73 (95% CI, 0.40–0.86), respectively. Moreover, the AUPRC values and DCA of three model demonstrated that the clinical-ultrasomics model consistently outperformed both the clinical model and the ultrasomics model across all three sets, including the training set and the two validation sets. Conclusions Our study demonstrated the effective predictive capacity of a prediction model combining echocardiogram-based ultrasomics features and clinical parameters in predicting CALs progression in KD.

show abstract

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

Cited by 5 publications

References 21 publications

Blood Vessels Disease Detection of Coronary Angiography Images using Deep learning Model

Blood Vessels Disease Detection of Coronary Angiography Images using Deep learning Model

YOLO Based Segmentation and CNN Based Classification Framework for Epithelial and Pus Cell Detection

Progression prediction of coronary artery lesions by echocardiography-based ultrasomics analysis in Kawasaki disease

Contact Info

Product

Resources

About