Optimizing Object Detection Algorithms for Congenital Heart Diseases in Echocardiography: Exploring Bounding Box Sizes and Data Augmentation Techniques
This study confirms that careful annotation of imaging data is crucial for optimizing the performance of object detection algorithms in medical imaging. These findings suggest potential pathways for refining AI applications in diagnostic cardiology studies.
Authors
- Kai‐Sheng Hsieh (ORCID: https://orcid.org/0000-0002-0294-9872)
- Yihui Chen (ORCID: https://orcid.org/0000-0003-2858-5170)
- Jo-Hsin Shih
- Wenjun Li (ORCID: https://orcid.org/0000-0001-5335-7386)
- Shih-Hsin Chen (ORCID: https://orcid.org/0000-0001-5635-536X)
- Ruyi Zhang (ORCID: https://orcid.org/0000-0003-2228-3913)
- Yangyang Chen
- Ken‐Pen Weng
- Ting-Yi Kao
- Wan-Ru Tsai
Institutions
- Tamkang University (TW)
- Chang Gung University (TW)
- China Medical University (TW)
- Kaohsiung Chang Gung Memorial Hospital (TW)
- Kaohsiung Veterans General Hospital (TW)
Publication Details
- Journal
- Reviews in Cardiovascular Medicine
- Published
- 2024-09-19
- DOI
- https://doi.org/10.31083/j.rcm2509335
- Citations
- 5
- Primary Topic
- Congenital Heart Disease Studies
- Type
- article
- Field-Weighted Citation Impact
- 1.93
Funders
- Chang Gung Medical Foundation
- National Science and Technology Council
- Kaohsiung Chang Gung Memorial Hospital