Affiliations: [a] Department of Medical Imaging, Affiliated Hospital of Putian University, Fujian, China
| [b] Department of Medical Imaging, The First Hospital of Putian City, Fujian, China
| [c] Radiology Department, Peking University First Hospital, Beijing, China
Correspondence:
[*]
Corresponding author. Jianxing Qiu, Radiology Department, Peking University First Hospital, Beijing 100034, China. E-mail: quanzhe82675549@163.com.
Note: [#] These authors contributed equally to this work and should be considered co-first authors.
Abstract: The early hidden characteristics of breast tumors make their features difficult to be effectively identified. In order to improve the detection accuracy of breast tumors, this study combined with computer-aided diagnosis techniques such as machine learning and computer vision and used X-ray analysis to study breast tumor diagnosis techniques. Moreover, this study combines breast tumor diagnostic images to determine various parameters of the image. At the same time, through experimental research and analysis of the region segmentation method and preprocessing method of breast detection images, the best diagnostic images are obtained, and the influence of background and other noise on the image diagnosis results is effectively proposed. In addition, this study proposes a method for detecting the distortion of the mammogram image structure, which accurately detects the structural distortion and reduces the interference of various influencing factors. Finally, this paper designs experiments to study the effects of the diagnostic method of this paper. Through comparative analysis, it can be seen that the results of this study have certain advantages in accuracy and image clarity, and have certain clinical significance, and can provide theoretical reference for subsequent related research.
Keywords: X-ray analysis, breast neoplasms, diagnosis, image, machine learning
