Abstract
Tumor detection in automated breast ultrasound (ABUS) images is critical for computer-aided diagnosis. A single deep learning based network is prone to make overconfident prediction. In this paper, a DML-YOLOX model based on deep mutual learning (DML) is proposed. To alleviate the overconfidence of a single network, the dual-model collaborative training strategy is designed based on YOLOX baseline. To implement the interaction between the dual models, the exploration loss is developed and combined with the consistency loss for supervision. The exploration loss aims at encouraging the models to explore and learn different feature representations in the feature extraction stage. The consistency loss aims at constraining the models to have consistent output representations for the same input. According to the ellipse-like characteristic of ABUS tumors, a rotation augmentation method is designed, which can decrease the overestimate of the bounding box. Besides, in order to reduce the regression error caused by large angle rotation, a rotation discriminant (RD) measurement is developed. The proposed method is verified on two ABUS datasets, one of which is the private ABUS dataset with 68 normal volumes and 68 tumor volumes, including 43,248 slices, and the other is the public ABUS dataset. On the private ABUS dataset, it achieves a promising detection result with a sensitivity of 0.90 and the false positives per slice (FPs/S) at 0.15.
![](http://media.springernature.com/m312/springer-static/image/art%3A10.1007%2Fs11042-024-18377-8/MediaObjects/11042_2024_18377_Fig1_HTML.png)
![](http://media.springernature.com/m312/springer-static/image/art%3A10.1007%2Fs11042-024-18377-8/MediaObjects/11042_2024_18377_Fig2_HTML.png)
![](http://media.springernature.com/m312/springer-static/image/art%3A10.1007%2Fs11042-024-18377-8/MediaObjects/11042_2024_18377_Fig3_HTML.png)
![](http://media.springernature.com/m312/springer-static/image/art%3A10.1007%2Fs11042-024-18377-8/MediaObjects/11042_2024_18377_Fig4_HTML.png)
![](http://media.springernature.com/m312/springer-static/image/art%3A10.1007%2Fs11042-024-18377-8/MediaObjects/11042_2024_18377_Fig5_HTML.png)
![](http://media.springernature.com/m312/springer-static/image/art%3A10.1007%2Fs11042-024-18377-8/MediaObjects/11042_2024_18377_Fig6_HTML.png)
![](http://media.springernature.com/m312/springer-static/image/art%3A10.1007%2Fs11042-024-18377-8/MediaObjects/11042_2024_18377_Fig7_HTML.png)
![](http://media.springernature.com/m312/springer-static/image/art%3A10.1007%2Fs11042-024-18377-8/MediaObjects/11042_2024_18377_Fig8_HTML.png)
Similar content being viewed by others
Data availability
The datasets generated during and/or analysed during the current study are not publicly available due to no hospital permission but are available from the corresponding author on reasonable request.
References
Bray F, Ferlay J, Soerjomataram I, Siegel RL, Torre LA, Jemal A (2018) Global cancer statistics 2018: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries. CA Cancer J Clin 68:394–424
Liu Y, Li R, Cao X, Tong Y (2020) Breast tumors recognition based on edge feature extraction using support vector machine. Biomed Signal Process Control 58:101825
Kolb TM, Lichy J, Newhouse JH (2002) Comparison of the performance of screening mammography, physical examination, and breast US and evaluation of factors that influence them: An analysis of 27,825 patient evaluations. Radiology 225(1):165–175
Brem RF, Lenihan MJ, Lieberman J, Torrente J (2015) Screening breast ultrasound: Past, present, and future. Amer J Roentgenol 204(2):234–240
Ikedo Y, Fukuoka D, Hara T, Fujita H, Takada E (2007) Development of a fully automatic scheme for detection of masses in whole breast ultrasound images. Med Phys 34(11):4378–4388
Lo CM, Chen RT, Chang YC, Yang YW et al (2014) Multi-Dimensional Tumor Detection in Automated Whole Breast Ultrasound Using Topographic Watershed. IEEE Trans Med Imaging 33(7):1503–1511
Ikedo Y, Fukuoka D, Hara T et al (2007) Computerized mass detection in whole breast ultrasound images: Reduction of false positives using bilateral subtraction technique. Med Imaging 2007: Comput-Aided Diagnos 6514:570–579
Moon WK, Shen YW, Bae MS et al (2013) Computer-aided tumor detection based on multi-scale blob detection algorithm in automated breast ultrasound images. IEEE Trans Med Imaging 32(7):1191–1200
Tan T, Platel B, Mus R, Tabar L et al (2013) Computer-aided detection of cancer in automated 3-D breast ultrasound. IEEE Trans Med Imaging 32(9):1698–1706
Lo CM, Chen RT, Chang YC et al (2014) Multi-dimensional tumor detection in automated whole breast ultrasound using topographic watershed. IEEE Trans Med Imaging 33(7):1503–1511
Moon WK, Haung YS, Hsu CH et al (2020) Computer-aided tumor detection in automated breast ultrasound using a 3-D convolutional neural network. Comput Methods Programs Biomed 190:105360
Zhang Z, Zhang X, Lin X et al (2019) Ultrasonic Diagnosis of Breast Nodules Using Modified Faster R-CNN. Ultrason Imaging 41(6):353–367
Wang F, Liu X, Qian B, et al. (2019) Spatial attention lesion detection on automated breast ultrasound, International Work-Conference on Bioinformatics and Biomedical Engineering, 216–227. https://doi.org/10.1007/978-3-030-17938-0_20
Zhou Y, Chen H, Li Y et al (2021) 3D multi-view tumor detection in automated whole breast ultrasound using deep convolutional neural network. Expert Syst Appl 168:114410
Wang Y, Wang N, Xu M et al (2020) Deeply supervised networks with threshold loss for cancer detection in automated breast ultrasound. IEEE Trans Med Imaging 39(4):866–876
Li Y, Wu W, Chen H et al (2020) 3D tumor detection in automated breast ultrasound using deep convolutional neural network. Med Phys 47(11):5669–5680
Zhang JX, Tao X, Jiang YH et al (2022) Application of convolution neural network algorithm based on multicenter ABUS images in breast lesion detection. Front Oncol 12:938413
Wang Y, Yao YD (2022) Breast lesion detection using an anchor-free network from ultrasound images with segmentation-based enhancement. Sci Rep 12(1):14720
Zhang Z, Li Y, Wu W et al (2021) Tumor detection using deep learning method in automated breast ultrasound. Biomed Signal Process Control 68(1):102677
Joshi RC, Singh D, Tiwari V, Dutta MK (2022) An efficient deep neural network based abnormality detection and multi class breast tumor classification. Multimed Tools Appl 81:13691–13711
Zou Z, Chen K, Shi Z, Guo Y, Ye J (2023) Object detection in 20 years: a survey. Proc IEEE 111(3):257–276. https://doi.org/10.1109/JPROC.2023.3238524
Girshick R (2015) Fast R-CNN. In: IEEE International Conference on Computer Vision (ICCV), pp 1440–1448. https://doi.org/10.1109/ICCV.2015.169
Ren S, He K, Girshick R et al (2015) Faster R-CNN: Towards real-time object detection with region proposal networks. IEEE Trans Pattern Anal Mach Intell 39(6):1137–1149
Lin TY, Dollár P, Girshick R, He K, Hariharan B, Belongie S (2017) Feature pyramid networks for object detection. In: IEEE Conference on Computer Vision and Pattern Recognition (CVPR), pp 936–944. https://doi.org/10.1109/CVPR.2017.106
Redmon J, Farhadi A (2017) YOLO9000: better, faster, stronger. In: IEEE Conference on Computer Vision and Pattern Recognition (CVPR), pp 6517–6525. https://doi.org/10.1109/CVPR.2017.690
Redmon J, Farhadi A (2018) Yolov3: an incremental improvement. arXiv preprint, arXiv:1804.02767. https://arxiv.org/pdf/1804.02767.pdf
Bochkovskiy A, Wang C, Liao H (2020) Yolov4: optimal speed and accuracy of object detection. arXiv preprint, arXiv:2004.10934. https://arxiv.org/pdf/2004.10934.pdf
Ge Z, Liu S, Wang F, Li Z, Sun J (2021) Yolox: exceeding yolo series in 2021. arXiv preprint, arXiv:2107.08430. https://arxiv.org/pdf/2107.08430v2.pdf
Liu W, Anguelov D, Erhan D, et al. (2016) SSD: Single shot multibox detector, European Conference on Computer Vision (ECCV), 21–37. https://doi.org/10.1007/978-3-319-46448-0_2
Duan K, Bai S, Xie L, Qi H, Huang Q, Tian Q (2019) Centernet: keypoint triplets for object detection. In: IEEE/CVF International Conference on Computer Vision (ICCV), pp 6569–6578. https://doi.org/10.1109/ICCV.2019.00667
Zhang Y, Xiang T, Hospedales TM, Lu H (2018) Deep mutual learning. In: IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), pp 4320–4328. https://doi.org/10.1109/CVPR.2018.00454
Zhao H, Yang G, Wang D, Lu H (2019) Lightweight deep neural network for real-time visual tracking with mutual learning. In: IEEE International Conference on Image Processing (ICIP), pp 3063–3067. https://doi.org/10.1109/ICIP.2019.8803340
Wang Y, Che Z, Wu QM (2020) J, Rong X, Deep mutual learning network for gait recognition. Multimed Tools Appl 79:22653–22672
Feng X, Gu X, Kuang P, Li X, Zhu Y (2021) Pedestrian detection and tracking with deep mutual learning. In: 18th International Computer Conference on Wavelet Active Media Technology and Information Processing (ICCWAMTIP), pp 217–220. https://doi.org/10.1109/ICCWAMTIP53232.2021.9674099
Zhao H, Yang G, Wang D, Lu H (2021) Deep mutual learning for visual object tracking. Pattern Recogn 112:107796
Montserrat D M, Lin Q, Allebach J, et al. (2017) Training object detection and recognition CNN models using data augmentation, Imaging and Multimedia Analytics in a Web and Mobile World (IMAWM), 27–36. https://doi.org/10.2352/ISSN.2470-1173.2017.10.IMAWM-163
Lin TY, Maire M, Belongie S, Hays J, Perona P, Ramanan D, Dollar P, Zitnick CL (2014) Microsoft COCO: Common objects in context, European Conference on Computer Vision (ECCV), 740–755. https://doi.org/10.1007/978-3-319-10602-1_48
Tan M, Pang R, Le QV (2020) Efficientdet: scalable and efficient object detection. In: IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), pp 10781–10790. https://doi.org/10.1109/CVPR42600.2020.01079
Lv W, Zhao Y, Xu S, Wei J, Wang G, Cui C, Du Y, Dang Q, Liu Y (2023) DETRs beat YOLOs on real-time object detection. arXiv preprint, arXiv: 2304.08069. https://arxiv.org/pdf/2304.08069.pdf
Zheng ZH, Ye RG, Hou QB et al (2023) Localization distillation for object detection. IEEE Trans Pattern Anal Mach Intell 45(8):10070–10083
Acknowledgements
The work was supported in part by the National Natural Science Foundation of China (No. 62272027), the Beijing Natural Science Foundation (No. 4232012) and the Fundamental Research Funds for the Central Universities (2021JBM001).
Author information
Authors and Affiliations
Corresponding author
Ethics declarations
Declarations
We would like to confirm that the data collection process for this research has undergone the approval from the local ethics committee. The research was conducted in accordance with the guidelines and regulations set forth by the committee to ensure the protection of human subjects and the ethical handling of data.
We have followed strict protocols and obtained informed consent from all participants involved in the study. The identities of the participants have been anonymized to ensure confidentiality and privacy. We have taken all necessary precautions to minimize any potential harm or risks associated with the research.
Please note that due to the nature of the research and the specific context in which it was conducted, the requirement for a written informed consent form was waived by the ethics committee. The committee acknowledged that the research design and procedures complied with ethical principles and did not pose any significant ethical concerns.
Conflicts of interest
The authors do not have relevant conflicts of interest to disclose.
Additional information
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
About this article
Cite this article
Li, Y., Zhang, Z., Sun, J. et al. Tumor detection based on deep mutual learning in automated breast ultrasound. Multimed Tools Appl 83, 68421–68439 (2024). https://doi.org/10.1007/s11042-024-18377-8
Received:
Revised:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11042-024-18377-8