YOLOv9 for fracture detection in pediatric wrist trauma X-ray images

被引:20
作者
Chien, Chun-Tse [1 ]
Ju, Rui-Yang [2 ]
Chou, Kuang-Yi [3 ]
Chiang, Jen-Shiun [1 ]
机构
[1] Tamkang Univ, Dept Elect & Comp Engn, New Taipei City, Taiwan
[2] Natl Taiwan Univ, Grad Inst Networking & Multimedia, Taipei City, Taiwan
[3] Natl Taipei Univ Nursing & Hlth Sci, Sch Nursing, Taipei City, Taiwan
关键词
biomedical imaging; computer vision; object detection; X-ray detection; NETWORK;
D O I
10.1049/ell2.13248
中图分类号
TM [电工技术]; TN [电子技术、通信技术];
学科分类号
0808 ; 0809 ;
摘要
The introduction of YOLOv9, the latest version of the you only look once (YOLO) series, has led to its widespread adoption across various scenarios. This paper is the first to apply the YOLOv9 algorithm model to the fracture detection task as computer-assisted diagnosis to help radiologists and surgeons to interpret X-ray images. Specifically, this paper trained the model on the GRAZPEDWRI-DX dataset and extended the training set using data augmentation techniques to improve the model performance. Experimental results demonstrate that compared to the mAP 50-95 of the current state-of-the-art model, the YOLOv9 model increased the value from 42.16% to 43.73%, with an improvement of 3.7%. The implementation code is publicly available at . YOLOv9 is the latest version of the you only look once (YOLO) series of object detection algorithms released in February 2024. This paper presents a framework for pediatric wrist fracture detection utilizing YOLOv9, which enhances model performance to achieve the state-of-the-art level through training on the GRAZPEDWRI-DX dataset. This is significant because misinterpretation of fracture X-ray images may lead to surgery failure and more harm to the patients. The hope is to use artificial intelligence technology to prevent these incidents from occurring. image
引用
收藏
页数:3
相关论文
共 24 条
[11]   Microsoft COCO: Common Objects in Context [J].
Lin, Tsung-Yi ;
Maire, Michael ;
Belongie, Serge ;
Hays, James ;
Perona, Pietro ;
Ramanan, Deva ;
Dollar, Piotr ;
Zitnick, C. Lawrence .
COMPUTER VISION - ECCV 2014, PT V, 2014, 8693 :740-755
[12]  
Liu Y., 2021, arXiv, DOI DOI 10.48550/ARXIV.2112.05561
[13]   Toward Computing Cross-Modality Symmetric Non-Rigid Medical Image Registration [J].
Mohanty, Snigdha ;
Dakua, Sarada Prasad .
IEEE ACCESS, 2022, 10 :24528-24539
[14]   A pediatric wrist trauma X-ray dataset (GRAZPEDWRI-DX) for machine learning [J].
Nagy, Eszter ;
Janisch, Michael ;
Hrzic, Franko ;
Sorantin, Erich ;
Tschauner, Sebastian .
SCIENTIFIC DATA, 2022, 9 (01)
[15]   You Only Look Once: Unified, Real-Time Object Detection [J].
Redmon, Joseph ;
Divvala, Santosh ;
Girshick, Ross ;
Farhadi, Ali .
2016 IEEE CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION (CVPR), 2016, :779-788
[16]   Radiologist shortage leaves patient care at risk, warns royal college [J].
Rimmer, Abi .
BMJ-BRITISH MEDICAL JOURNAL, 2017, 359
[17]   Detection of solder paste defects with an optimization-based deep learning model using image processing techniques [J].
Sezer, Au ;
Altan, Aytac .
SOLDERING & SURFACE MOUNT TECHNOLOGY, 2021, 33 (05) :291-298
[18]  
Wang CY, 2024, Arxiv, DOI [arXiv:2402.13616, DOI 10.48550/ARXIV.2402.13616]
[19]  
Wang CY, 2023, J INF SCI ENG, V39, P975, DOI [10.6688/JISE.202307, 10.6688/JISE.202307_39(4).0016]
[20]   CSPNet: A New Backbone that can Enhance Learning Capability of CNN [J].
Wang, Chien-Yao ;
Liao, Hong-Yuan Mark ;
Wu, Yueh-Hua ;
Chen, Ping-Yang ;
Hsieh, Jun-Wei ;
Yeh, I-Hau .
2020 IEEE/CVF CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION WORKSHOPS (CVPRW 2020), 2020, :1571-1580