발간년도 : [2023]
| 논문정보 |
|
| 논문명(한글) |
[Vol.18, No.6] A Study on the Image Generation with Residual U-Net Generator of CycleGAN in Digital Breast Tomography |
|
| 논문투고자 |
Hyun-seong Lee, Ho-Jung Song, Yong-Do Her |
|
| 논문내용 |
In recent years, the field of deep learning has experienced rapid growth and widespread applicationacross various domains, including cybersecurity, natural language processing, bioinformatics, and medicalinformation processing. Data in the medical domain faces challenges such as anonymizing personalinformation and labeling cost issues, leading to research considering GAN(Generative AdversarialNetworks)-based data synthesis and augmentation as a solution. Not only is data augmentation valuable,but tasks such as synthesizing CT images from MRI images, utilizing deep learning, prove useful inovercoming constraints imposed by the unique characteristics of medical images. Furthermore, deeplearning has seen extensive application such as lesion Segmentation and Denoising in Digital BreastTomosynthesis (DBT) images. DBT images excel in diagnosing cancer that may not be visible intraditional breast imaging methods. In this study, we develop a GAN-based model for synthesizingimages from an imbalanced dataset of DBT (Digital Breast Tomosynthesis) images under un-pairedconditions. To address the imbalance, a GAN model utilizing the Cycle-consistency concept is employed.For performance improvement, the generator's architecture combines the U-net structure, known for highperformance in the medical domain, with Residual Blocks. The model's performance is evaluatedquantitatively using metrics for the generated images, and a visual assessment is conducted. Anticipatedis a positive impact on data augmentation for imbalanced medical datasets under un-paired conditionsthrough such image synthesis models. |
|
| 첨부논문 |
|
|
|
|
|
자료실 > 논문지 
