Abudukelimu Halidanmu, Gao Yuxin, Abulizi Abudukelimu, Musideke Mayilamu, Wu Shuqin, Wang Mengfei, Aizizi Mireguli, Yehaiya Gulimiremu, Abudukelimu Mayila
School of Information Management, Xinjiang University of Finance and Economics, Urumqi, China.
Department of Radiology (CT & MRI), The Friendship Hospital of Ili Kazakh Autonomous Prefecture, Yining, China.
Digit Health. 2025 Sep 2;11:20552076251374192. doi: 10.1177/20552076251374192. eCollection 2025 Jan-Dec.
Accurate segmentation of breast lesions, especially small ones, remains challenging in digital mammography due to complex anatomical structures and low-contrast boundaries. This study proposes DVF-YOLO-Seg, a two-stage segmentation framework designed to improve feature extraction and enhance small-lesion detection performance in mammographic images.
The proposed method integrates an enhanced YOLOv10-based detection module with a segmentation stage based on the Visual Reference Prompt Segment Anything Model (VRP-SAM). A novel DualConv module is introduced to improve spatial, visual, and channel feature representation, while Varifocal Loss addresses class imbalance by emphasizing hard-to-detect lesions. The detection results are used to generate bounding box prompts for VRP-SAM, which refines the final lesion segmentation.
Experiments on the curated breast imaging subset of the digital database for screening mammography dataset demonstrate that DVF-YOLO-Seg achieves a precision of 79.7%, a recall of 81.5%, a dice coefficient of 80.2%, and an F1-score of 80.6%, outperforming baseline models. Particularly for lesions <5 mm, the model shows improved sensitivity. Ablation studies confirm the effectiveness of the DualConv module and Varifocal Loss. Additionally, the framework shows better visual consistency and clearer boundaries in clinician-evaluated results.
DVF-YOLO-Seg significantly enhances the detection and segmentation accuracy for small breast lesions in mammography. By combining improved detection with prompt-based segmentation, this method offers a promising approach for computer-aided diagnosis in breast cancer screening.
由于乳腺解剖结构复杂且边界对比度低，在数字乳腺摄影中准确分割乳腺病变，尤其是小病变，仍然具有挑战性。本研究提出了DVF-YOLO-Seg，这是一种两阶段分割框架，旨在改进特征提取并提高乳腺钼靶图像中小病变的检测性能。
所提出的方法将基于增强型YOLOv10的检测模块与基于视觉参考提示分割一切模型（VRP-SAM）的分割阶段相结合。引入了一种新颖的双卷积模块来改善空间、视觉和通道特征表示，而变焦距损失通过强调难以检测的病变来解决类别不平衡问题。检测结果用于为VRP-SAM生成边界框提示，从而细化最终的病变分割。
在用于乳腺钼靶筛查的数字数据库精选乳腺成像子集中进行的实验表明，DVF-YOLO-Seg的精度达到79.7%，召回率达到81.5%，骰子系数达到80.2%，F1分数达到80.6%，优于基线模型。特别是对于小于5毫米的病变，该模型显示出更高的灵敏度。消融研究证实了双卷积模块和变焦距损失的有效性。此外，在临床医生评估的结果中，该框架显示出更好的视觉一致性和更清晰的边界。
DVF-YOLO-Seg显著提高了乳腺钼靶中小乳腺病变的检测和分割准确性。通过将改进的检测与基于提示的分割相结合，该方法为乳腺癌筛查中的计算机辅助诊断提供了一种有前景的方法。
