Gui Haitian, Jiao Han, Li Li, Jiang Xinhua, Su Tao, Pang Zhiyong
School of Biomedical Engineering, Shenzhen Campus of Sun Yat-sen University, Shenzhen 518107, China.
School of Electronics and Information Technology, Sun Yat-sen University, Guangzhou 510006, China.
Bioengineering (Basel). 2024 Dec 1;11(12):1217. doi: 10.3390/bioengineering11121217.
AI-based breast cancer detection can improve the sensitivity and specificity of detection, especially for small lesions, which has clinical value in realizing early detection and treatment so as to reduce mortality. The two-stage detection network performs well; however, it adopts an imprecise ROI during classification, which can easily include surrounding tumor tissues. Additionally, fuzzy noise is a significant contributor to false positives. We adopted Faster RCNN as the architecture, introduced ROI aligning to minimize quantization errors and feature pyramid network (FPN) to extract different resolution features, added a bounding box quadratic regression feature map extraction network and three convolutional layers to reduce interference from tumor surrounding information, and extracted more accurate and deeper feature maps. Our approach outperformed Faster R-CNN, Mask R-CNN, and YOLOv9 in breast cancer detection across 485 internal cases. We achieved superior performance in mAP, sensitivity, and false positive rate ((0.752, 0.950, 0.133) vs. (0.711, 0.950, 0.200) vs. (0.718, 0.880, 0.120) vs. (0.658, 0.680, 405)), which represents a 38.5% reduction in false positives compared to manual detection. Additionally, in a public dataset of 220 cases, our model also demonstrated the best performance. It showed improved sensitivity and specificity, effectively assisting doctors in diagnosing cancer.
基于人工智能的乳腺癌检测可以提高检测的灵敏度和特异性，特别是对于小病灶，这在实现早期检测和治疗以降低死亡率方面具有临床价值。两阶段检测网络表现良好；然而，它在分类过程中采用的感兴趣区域（ROI）不准确，容易包含周围的肿瘤组织。此外，模糊噪声是误报的一个重要因素。我们采用更快的区域卷积神经网络（Faster RCNN）作为架构，引入感兴趣区域对齐（ROI aligning）以最小化量化误差，并使用特征金字塔网络（FPN）来提取不同分辨率的特征，添加了一个边界框二次回归特征图提取网络和三个卷积层以减少肿瘤周围信息的干扰，并提取更准确、更深层次的特征图。在485例内部病例的乳腺癌检测中，我们的方法优于更快的区域卷积神经网络（Faster R-CNN）、掩码区域卷积神经网络（Mask R-CNN）和YOLOv9。我们在平均精度均值（mAP）、灵敏度和误报率方面取得了卓越的性能（（0.752，0.950，0.133）对比（0.711，0.950，0.200）对比（0.718，0.880，0.120）对比（0.658，0.680，4（此处原文有误，推测为0.405））），与人工检测相比，误报率降低了38.5%。此外，在一个包含220例病例的公共数据集中，我们的模型也表现出了最佳性能。它提高了灵敏度和特异性，有效地协助医生诊断癌症。
