|
|
金字塔卷积融合YOLOv4的乳腺肿块检测算法
|
Abstract:
针对乳腺肿块与腺体对比度低而导致检测精度低，假阳性率较高的问题，提出了一种基于改进的YOLOv4网络的乳腺肿块检测方法。首先引入金字塔卷积，通过不同大小和深度的卷积核，对输入的特征进行提取；其次，将原特征提取网络中的普通卷积替换为深度可分离卷积，减少网络训练参数，提升网络训练速度。实验结果表明，改进的YOLOv4算法在测试集上的敏感性达到81.49%，较原网络高了2.81%，AP值达到86.85%，比原网络的AP值高了4.27%；平均每幅图假阳性个数为0.418个，相较原网络降低了0.028个，该算法的检测性能较YOLOv4有明显提升。
Aiming at the problems of low detection accuracy and high false positive rate caused by low contrast between breast mass and gland, a breast mass detection method based on improved YOLOv4 is proposed. Firstly, Pyramid Convolution is introduced to extract the input features through convolution kernels of different sizes and depths; Secondly, the ordinary convolution in the original feature extraction network is replaced by Depthwise Separable Convolution to reduce the network training parameters and improve the network training speed. The experimental results show that the sensitivity of the improved YOLOv4 algorithm on the test set is 81.49%, which is 2.81% higher than that of the original network, and the AP value is 86.85%, which is 4.27% higher than that of the original network; The average number of false positives per image is 0.418, which is 0.028 lower than the original network. The detection performance of the algorithm is significantly improved compared with YOLOv4.
| [1] | Siegel, R.L., Miller, K.D., Fuchs, H. and Jemal, A. (2021) Cancer Statistics. CA: A Cancer Journal for Clinicians, 71, 7-33. https://doi.org/10.3322/caac.21654 |
| [2] | Chung, J., Rajagopal, V., Nielsen, P., et al. (2008) A Biomechanical Model of Mammographic Compressions. Biomechanics and Modeling in Mechanobiology, 7, 43-52. https://doi.org/10.1007/s10237-006-0074-6 |
| [3] | Girshick, R., Donahue, J., Darrell, T., et al. (2013) Rich Feature Hierarchies for Accurate Object Detection and Semantic Segmentation. 2014 IEEE Conference on Computer Vision and Pattern Recognition, Columbus, 23-28 June 2014, 580-587. https://doi.org/10.1109/CVPR.2014.81 |
| [4] | Girshick, R. (2015) Fast R-CNN. IEEE International Conference on Computer Vision, Santiago, 7-13 December 2015, 1440-1448. https://doi.org/10.1109/ICCV.2015.169 |
| [5] | Ren, S., He, K., Girshick, R., et al. (2015) Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks. International Conference on Neural Information Processing Systems, Montreal, 7-12 December 2015, 91-99. |
| [6] | 郑杰, 郭朋. 一种改进Faster R-CNN检测乳腺肿块的方法[J]. 影像研究与医学应用, 2020, 4(14): 139-142. |
| [7] | Liu, W., Anguelov, D., Erhan, D., et al. (2016) SSD: Single Shot MultiBox Detector. European Conference on Computer Vision, Amsterdam, 11-14 October 2016, 21-37. https://doi.org/10.1007/978-3-319-46448-0_2 |
| [8] | Redmon, J., Divvala, S., Girshick, R., et al. (2016) You Only Look Once: Unified, Real-Time Object Detection. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, Las Vegas, 27-30 June 2016, 779-788.
https://doi.org/10.1109/CVPR.2016.91 |
| [9] | Redmon, J. and Farhadi, A. (2017) YOL09000: Better, Faster, Stronger. IEEE Conference on Computer Vision and Pattern Recognition, Honolulu, 21-26 July 2017, 6517-6525. https://doi.org/10.1109/CVPR.2017.690 |
| [10] | Redmon, J. and Farhadi, A. (2018) YOLOv3: An Incremental Improvement. 2018 IEEE Conference on Computer Vision and Pattern Recognition, Honolulu, 21-26 July 2017, 1-6. |
| [11] | 蒋慧琴, 王博霖, 马岭, 等. 一种双视图信息融合的乳腺肿块自动检测算法[J]. 郑州大学学报(理学版), 2020, 52(4): 28-36. |
| [12] | Bochkovskiy, A., Wang, C.Y. and Liao, H.Y.M. (2020) YOLOv4: Optimal Speed and Accuracy of Object Detection. |
| [13] | Duta, I.C., Liu, L., Zhu, F., et al. (2020) Pyramidal Convolution: Rethinking Convolutional Neural Networks for Visual Recognition. |
| [14] | Wang, C.Y., Liao, H., Yeh, I.H., et al. (2020) CSPNet: A New Backbone That Can Enhance Learning Capability of CNN. Conference on Computer Vision and Pattern Recognition Workshops, Seattle, 14-19 June 2020, 1571-1580.
https://doi.org/10.1109/CVPRW50498.2020.00203 |
| [15] | Howard, A.G., Zhu, M., Chen, B., et al. (2017) MobileNets: Efficient Convolutional Neural Networks for Mobile Vision Applications. |
| [16] | Lee, R.S., Gimenez, F., Hoogi, A., et al. (2017) A Curated Mammography Data Set for Use in Computer-Aided Detection and Diagnosis Research. Scientific Data, 4, Article ID: 170177. https://doi.org/10.1038/sdata.2017.177 |
| [17] | Heath, M., Bowyer, K., Kopans, D., et al. (2001) The Digital Database for Screening Mammography. Digital Mammography. Springer, Berlin. |
