Earticle

초록

영어

This study proposes an integrated multi-task learning model that performs both lesion segmentation and lesion-type classification simultaneously in breast ultrasound images. Conventional single-task approaches handle segmentation and classification independently, often leading to information isolation and a failure to leverage complementary features between the tasks. To overcome these limitations, we adopt a two-stage U-Net architecture with a ResNet-101 backbone and introduce the Hierarchical Gating Module (HGM) as the core component of our framework. HGM hierarchically reuses the coarse segmentation probability map generated in Stage 1 to modulate multi-scale encoder features in Stage 2, thereby maximizing the synergistic interaction between segmentation and classification. Experiments conducted on the BUSI breast ultrasound dataset demonstrate that the proposed HGMNet achieves superior performance compared to existing models, recording a Dice coefficient of 0.7431 for segmentation and an accuracy of 0.8500 for classification. These results indicate that the proposed model can effectively integrate the two tasks within a single unified network, thereby enhancing both the accuracy and reliability of breast ultrasound–based diagnosis.

목차

Abstract
I. 서론
II. 연구 방법
1. 데이터셋 구성
2. 제안 모델(HGM-Net) 구조
3. 손실 함수
4. 학습 및 검증 과정
III. 연구 결과 및 고찰
1. 주요 실험 결과
2. Ablation Study
3. 외부 데이터셋 검증 결과
IV. 결론
감사의 말씀
참고문헌

키워드

저자

• 오예인 [ Yein Oh | 동국대학교 바이오헬스의료기기규제과학과 ]
• 고재은 [ Jaeeun Ko | 동국대학교 바이오헬스의료기기규제과학과 ]
• 권지연 [ Jiyean Kwon | 동국대학교 의료기기산업학과 ]
• 김성민 [ Sungmin | 동국대학교 바이오헬스의료기기규제과학과, 동국대학교 의료기기산업학과, 동국대학교 의생명공학과 ] Corresponding Author

참고문헌

발행기관

간행물

표지보기 관심저널 등록

• 간행물명

  KFDC규제과학회지(구 FDC법제연구) [Regulatory Research on Food, Drug and Cosmetic]

• 간기

  반년간

• pISSN

  2799-8940

• 수록기간

  2006~2025

• 십진분류

  KDC 518 DDC 615