Wang Linjing, Zheng Chao, Chen Wentao, He Qiang, Li Xin, Zhang Shuxu, Qin Genggeng, Chen Weiguo, Wei Jun, Xie Peiliang, Zhou Linghong, Wang Xuetao, Zhen Xin
Radiotherapy Center, Affiliated Cancer Hospital & Institute of Guangzhou Medical University, Guangzhou, Guangdong 510095 People's Republic of China.
Contributed equally.
Phys Med Biol. 2020 Dec 4;65(23):235045. doi: 10.1088/1361-6560/abaeb7.
To develop and evaluate a multi-path synergic fusion (MSF) deep neural network model for breast mass classification using digital breast tomosynthesis (DBT).
We retrospectively collected 441 patients who had undergone DBT in which the regions of interest (ROIs) covering the malignant/benign breast mass were extracted for model training and validation. In the proposed MSF framework, three multifaceted representations of the breast mass (gross mass, overview, and mass background) are extracted from the ROIs and independently processed by a multi-scale multi-level features enforced DenseNet (MMFED). The three MMFED sub-models are finally fused at the decision level to generate the final prediction. The advantages of the MMFED over the original DenseNet, as well as different fusion strategies embedded in MSF, were comprehensively compared.
The MMFED was observed to be superior to the original DenseNet, and multiple channel fusions in the MSF outperformed the single-channel MMFED and double-channel fusion with the best classification scores of area under the receiver operating characteristic (ROC) curve (87.03%), Accuracy (81.29%), Sensitivity (74.57%), and Specificity (84.53%) via the weighted fusion method embedded in MSF. The decision level fusion-based MSF was significantly better (in terms of the ROC curve) than the feature concatenation-based fusion (p< 0.05), the single MMFED using a fused three-channel image (p< 0.04), and the multiple MMFED end-to-end training (p< 0.004).
Integrating multifaceted representations of the breast mass tends to increase benign/malignant mass classification performance and the proposed methodology was verified to be a promising tool to assist in clinical breast cancer screening.
开发并评估一种用于乳腺肿块分类的多路径协同融合（MSF）深度神经网络模型，该模型使用数字乳腺断层合成（DBT）技术。
我们回顾性收集了441例接受DBT检查的患者，从中提取覆盖恶性/良性乳腺肿块的感兴趣区域（ROI）用于模型训练和验证。在所提出的MSF框架中，从ROI中提取乳腺肿块的三种多方面特征表示（肿块总体、概览和肿块背景），并由多尺度多层次特征增强的密集连接网络（MMFED）独立处理。最后，三个MMFED子模型在决策层面进行融合以生成最终预测。全面比较了MMFED相对于原始密集连接网络的优势，以及MSF中嵌入的不同融合策略。
观察到MMFED优于原始密集连接网络，并且MSF中的多通道融合优于单通道MMFED和双通道融合，通过MSF中嵌入的加权融合方法，其在接受者操作特征（ROC）曲线下面积、准确率、灵敏度和特异性方面的最佳分类得分分别为87.03%、81.29%、74.57%和84.53%。基于决策层面融合的MSF在ROC曲线方面显著优于基于特征拼接的融合（p<0.05）、使用融合三通道图像的单个MMFED（p<0.04）以及多个MMFED的端到端训练（p<0.004）。
整合乳腺肿块的多方面特征表示往往会提高良性/恶性肿块分类性能，并且所提出的方法被验证是一种有前途的辅助临床乳腺癌筛查的工具。
