共 46 条
Vision Transformers-Based Transfer Learning for Breast Mass Classification From Multiple Diagnostic Modalities
被引:1
作者:
Ayana, Gelan
[1
,2
]
Choe, Se-woon
[1
,3
]
机构:
[1] Kumoh Natl Inst Technol, Dept Med IT Convergence Engn, Gumi 39253, South Korea
[2] Jimma Univ, Sch Biomed Engn, Jimma 378, Ethiopia
[3] Kumoh Natl Inst Technol, Dept IT Convergence Engn, Gumi 39253, South Korea
基金:
新加坡国家研究基金会;
关键词:
Vision Transformer;
Transfer Learning;
Breast Mass;
Ultrasound;
Mammography;
CANCER;
MAMMOGRAMS;
D O I:
10.1007/s42835-024-01904-w
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:
Breast mass evaluation is crucial for early breast cancer diagnosis via imaging. While Convolutional Neural Network (CNN)-based deep learning (DL) has enhanced this process, it suffers from computational complexity and limited spatial encoding. Vision Transformer (ViT)-based DL, more adept at encoding spatial information, presents a promising alternative. This study introduces a ViT-based transfer learning (TL) method for breast mass classification. Three ViT-based TL architectures pretrained on ImageNet were proposed and evaluated using ultrasound and mammogram datasets. Comparative analysis against ViT trained from scratch and CNN-based TL was conducted. Results showed the ViT-based TL method achieving the highest area under curve (AUC) of 1 +/- 0 for both datasets, outperforming ViT from scratch and yielding similar or better performance compared to CNN-based TL. Despite its computational cost, ViT-based TL demonstrates superior classification capabilities for breast mass images. This research provides a foundational framework for future studies exploring ViT-based TL in breast cancer diagnosis.
引用
收藏
页码:3391 / 3410
页数:20
相关论文