浏览全部资源
扫码关注微信
收稿：2020-05-30，
修回：2020-7-7，
录用：2020-7-14，
纸质出版：2020-10-16
移动端阅览
目的
2
超声检查是诊断甲状腺疾病的主要影像学方法之一，但由于超声图像中斑点强度具有随机性、组织器官复杂等问题，导致甲状腺在不同数据源间的形态、大小和纹理差异性较大，容易导致观察者视觉疲劳。针对甲状腺超声成像存在斑点强度随机性以及周边组织复杂性的问题，为了更准确地描述出器官与病理性病变的解剖边界，提出一种基于频域增强和局部注意力机制的甲状腺超声分割网络。
方法
2
针对原始数据采用高低通滤波器获取高低频段的图像信息，整合高频段细节特征与低频段边缘特征，增强图像前背景的对比度，降低图像间的差异性。根据卷积网络中网络深度所提取特征信息量的不同，采用局部注意力机制对高低维特征信息进行自适应激活，增强低维特征的细节信息，弱化对非目标区域的关注，增强高维特征的全局信息，弱化冗余信息对网络的干扰，增强前背景分类以及对非显著性目标检测的能力。采用金字塔级联空洞卷积获取不同感受野的特征信息，解决数据源间图像差异较大的问题。
结果
2
实验结果表明，本文方法在11~16 MHz时采集的16个手绘甲状腺超声公开数据集中，通过10折交叉验证显示准确率为0.989，召回率为0.849，精准率为0.940，Dice系数为0.812，效果优于当前其他医学图像分割网络。通过消融实验，证明本文的几个模块对超声图像分割确实具有一定的提升效果。
结论
2
本文所提分割网络，结合深度学习模型及传统图像处理模型的优点，能较好地处理超声图像随机斑点并且提升非显著性组织分割效果。
Objective
2
Ultrasound is a main imaging method used for the diagnosis of thyroid diseases. It is convenient for the diagnosis of medical results through the real-time study of its internal anatomical structure. In computer vision
the segmentation of image tissue and organ is the pre background classification of the pixels in the image. The final segmentation image boundary is the combination of the target pixels. The research on medical image segmentation has received much attention
which is mainly divided into two ideas
where the first idea is to obtain the target area by analyzing the pixel value of a given image through computer vision technology. However
the generalization ability of the given image analysis is poor
and the segmentation effect is unremarkable because of the interference of random noise in the ultrasonic image. The second idea is to use deep learning for obtaining the target area through the background information before deep convolution classification. However
the target area may be insignificant using the depth learning model because of the complexity of tissue and organs
the evident surrounding tissues
and the lack of background information before the image
making the abstract features obtained by the depth network mostly the surrounding non target area and causing the segmentation effect of the original target unideal. A thyroid image is different in shape
size
and texture among different data sources. To solve the two problems
a thyroid ultrasound segmentation network based on frequency domain enhancement and local attention mechanism is proposed to solve the problem of random noise interference and insignificant target.
Method
2
First
high and low pass filters are used to obtain the image information of high- and low-frequency bands
and the detail features of high frequency band and the edge feature of low frequency band are integrated to enhance the contrast of background and reduce the difference between images. Second
a local attention mechanism is used to adaptively activate the high- and low-dimensional feature information in accordance with the different information amounts of the features extracted by the network depth in the convolution network. This mechanmism can enhance the detailed information of low-dimensional features
weaken the attention to nontarget areas
enhance the global information of high-dimensional features
and weaken the interference of redundant information on the network
thereby enhancing the ability of background classification and nonsignificant target detection. Finally
a pyramid cascading hole is used
and convolution is utilized to obtain the feature information of different receptive fields and solve the problem of large image difference between data sources. In the training process
a mixed loss function is used to regress the network training effect
and pixel level loss (binary cross entropy) and image similarity loss (structural similarity) can better evaluate the segmentation prediction results. This paper uses the ResNet34 network
which is trained in advance to fine tune
to train the model of the network. The training set adopts the open data set of the network and selects approximately 3 500 images through the screening of appropriate images. During the training
one NVIDIA P100 graphics processing unit(GPU) server is used
the network training of approximately 10 epochs can achieve a better and stable effect
and the total training time is approximately 120 min.
Result
2
Experimental results show that the accuracy of the proposed method is 0.989
the recall rate is 0.849
the specificity is 0.94
and the Dice coefficient is 0.812
which is better than the current methods of medical image segmentation network
such as U-Net and CE-Net network
and is more accurate and special in the effect of ultrasound thyroid image segmentation. A significant improvement is found in heterosexuality and is better than the evaluation result for the network using the same dataset
such as sumNet. At the same time
the ablation experiments show that the proposed modules have a certain improvement effect on ultrasound image segmentation.
Conclusion
2
The proposed segmentation model combined with the advantages of deep learning model and traditional image processing model can better deal with ultrasound image random spots and improve the results of nonsignificant tissue segmentation.
Alom M Z, Yakopcic C, Taha T M and Asari V K. 2018. Nuclei segmentation with recurrent residual convolutional neural networks based U-Net (R2U-Net)//Proceedings of NAECON 2018 IEEE National Aerospace and Electronics Conference. Dayton: IEEE: 228-233[ DOI: 10.1109/NAECON.2018.8556686 http://dx.doi.org/10.1109/NAECON.2018.8556686 ]
Chi J N, Yu X S and Zhang Y F. 2018. Thyroid nodule malignantrisk detection in ultrasound image by fusing deep and texture features. Journal of Image and Graphics, 23(10):1582-1593
迟剑宁, 于晓升, 张艺菲. 2018.融合深度网络和浅层纹理特征的甲状腺结节癌变超声图像诊断.中国图象图形学报, 23(10):1582-1593[DOI:10.11834/jig.180232]
Goyal M, Yap M H and Hassanpour S. 2020. Multi-class semantic segmentation of skin lesions via fully convolutional networks[EB/OL].[2020-03-13] . https://arxiv.org/pdf/1711.10449.pdf https://arxiv.org/pdf/1711.10449.pdf
Gu Z W, Cheng J, Fu H Z, Zhou K, Hao H Y, Zhao Y T, Zhang T Y, Gao S H and Liu J. 2019. CE-Net:context encoder network for 2D medical image segmentation. IEEE Transactions on Medical Imaging, 38(10):2281-2292[DOI:10.1109/TMI.2019.2903562]
Hu J, Shen L and Sun G. 2018. Squeeze-and-excitation networks//Proceedings of 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition. Salt Lake City: IEEE: 7132-7141[ DOI: 10.1109/CVPR.2018.00745 http://dx.doi.org/10.1109/CVPR.2018.00745 ]
Illanes A, Esmaeili N, Poudel P, Balakrishnan S and Friebe M. 2019. Parametrical modelling for texture characterization-a novel approach applied to ultrasound thyroid segmentation. PLoS One, 14(1):e0211215[DOI:10.1371/journal.pone.0211215]
Isensee F, Petersen J, Klein A, Zimmerer D, Jaeger P F, Kohl S, Wasserthal J, Köhler G, Norajitra T, Wirkert S and Maier-Hein K H. 2018. nnU-Net: self-adapting framework for U-Net-based medical image segmentation[EB/OL].[2018-09-27] . https://arxiv.org/pdf/1809.10486.pdf https://arxiv.org/pdf/1809.10486.pdf
Li X G, Fu C P, Li X L and Wang Z H. 2019. Improved faster R-CNN algorithm for multi-scale target detection. Journal of Computer-Aided Design and Computer Graphics, 31(7):1095-1101
李晓光, 付陈平, 李晓莉, 王章辉. 2019.面向多尺度目标检测的改进Faster R-CNN算法.计算机辅助设计与图形学学报, 31(7):1095-1101[DOI:10.3724/SP.J.1089.2019.17283]
Lian J, Ma Y D, Ma Y R, Shi B, Liu J Z, Yang Z and Guo Y N. 2017. Automatic gallbladder and gallstone regions segmentation in ultrasound image. International Journal of Computer Assisted Radiology and Surgery, 12(4):553-568[DOI:10.1007/s11548-016-1515-z]
Nandamuri S, China D, Mitra P and Sheet D. 2019. SUMNet: fully convolutional model for fast segmentation of anatomical structures in ultrasound volumes//The 16th IEEE International Symposium on Biomedical Imaging. Venice: IEEE: 1729-1732[ DOI: 10.1109/ISBI.2019.8759210 http://dx.doi.org/10.1109/ISBI.2019.8759210 ]
Peng C, Zhang X Y, Yu G, Luo G M and Sun J. 2017. Large kernel matters-improve semantic segmentation by global convolutional network//Proceedings of 2017 IEEE Conference on Computer Vision and Pattern Recognition. Honolulu: IEEE: 1743-1751[ DOI: 10.1109/CVPR.2017.189 http://dx.doi.org/10.1109/CVPR.2017.189 ]
Peng W X, Liu C B, Xia S E, Chen Y H and Liu R. 2017. Statistic texture feature based thyroid nodule recognition on CT images. Space Medicine and Medical Engineering, 30(4):258-262
彭文献, 刘晨彬, 夏顺仁, 陈益红, 刘蕊. 2017.基于CT图像统计纹理特征的甲状腺结节识别技术.航天医学与医学工程, 30(4):258-262[DOI:10.16289/j.cnki.1002-0837.2017.04.005]
Poudel P, Illanes A, Ataide E J G, Esmaeili N, Balakrishnan S and Friebe M. 2019. Thyroid ultrasound texture classification using autoregressive features in conjunction with machine learning approaches. IEEE Access, 7:79354-79365[DOI:10.1109/ACCESS.2019.2923547]
Poudel R, Lamata P and Montana G. 2017. Recurrent fully convolutional neural networks for multi-slice MRI cardiac segmentation//Zuluaga M A, Bhatia K, Kainz B, Moghari M H and Pace D F, eds. Reconstruction, Segmentation, and Analysis of Medical Images. Cham: Springer: 83-94[ DOI: 10.1007/978-3-319-52280-7_8 http://dx.doi.org/10.1007/978-3-319-52280-7_8 ]
Qin P L, Wu K, Hu Y S, Zeng J C and Chai X F. 2020. Diagnosis of benign and malignant thyroid nodules using combined conventional ultrasound and ultrasound elasticity imaging. IEEE Journal of Biomedical and Health Informatics, 24(4):1028-1036[DOI:10.1109/JBHI.2019.2950994]
Qin X B, Zhang Z C, Huang C Y, Gao C, Dehghan M and Jagersand M. 2019. BASNet: boundary-aware salient object detection//Proceedings of 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR). Las Vegas: IEEE: 7479-7489[ DOI: 10.1109/CVPR.2019.00766 http://dx.doi.org/10.1109/CVPR.2019.00766 ]
Quan L, Zhang D, Yang Y, Liu Y and Qin Q Q. 2013. Segmentation of tumor ultrasound image via region-based Ncut method. Wuhan University Journal of Natural Sciences, 18(4):313-318[DOI:10.1007/s11859-013-0934-8]
Tran P V. 2017. A fully convolutional neural network for cardiac segmentation in short-axis MRI[EB/OL].[2020-05-01] . https://arxiv.org/pdf/1604.00494.pdf https://arxiv.org/pdf/1604.00494.pdf
Wang P Q, Chen P F, Yuan Y, Liu D, Huang Z H, Hou X D and Cottrell G. 2018. Understanding convolution for semantic segmentation//Proceedings of 2018 IEEE Winter Conference on Applications of Computer Vision (WACV 2018). Lake Tahoe: IEEE: 1451-1460[ DOI: 10.1109/WACV.2018.00163 http://dx.doi.org/10.1109/WACV.2018.00163 ]
Wang Z, Simoncelli E P and Bovik A C. 2003. Multiscale structural similarity for image quality assessment//Proceedings of the 37th Asilomar Conference on Signals, Systems and Computers. Pacific Grove: IEEE: 1398-1402[ DOI: 10.1109/ACSSC.2003.1292216 http://dx.doi.org/10.1109/ACSSC.2003.1292216 ]
Wunderling T, Golla B, Poudel P, Arens C, Friebe M and Hansen C. 2017. Comparison of thyroid segmentation techniques for 3D ultrasound//Proceedings Volume 10133, Medical Imaging 2017:Image Processing. Orlando:SPIE, 10133:1013317[DOI:10.1117/12.2254234]
Yan J Y, Lv D and Cui Y Y. 2017. A novel segmentation approach for intravascular ultrasound images. Journal of Medical and Biological Engineering, 37(3):386-394[DOI:10.1007/s40846-017-0233-5]
Zhao T and Wu X Q. 2019. Pyramid feature attention network for saliency detection//Proceedings of 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR). Long Beach: IEEE: 3080-3089[ DOI: 10.1109/CVPR.2019.00320 http://dx.doi.org/10.1109/CVPR.2019.00320 ]
Zhou B L, Khosla A, Lapedriza A, Oliva A and Torralba A. 2016. Learning deep features for discriminative localization//Proceedings of 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR). Las Vegas: IEEE: 2921-2929[ DOI: 10.1109/CVPR.2016.319 http://dx.doi.org/10.1109/CVPR.2016.319 ]
Zhuang Z M, Lei N H, Raj A N J and Qiu S M. 2019. Application of fractal theory and fuzzy enhancement in ultrasound image segmentation. Medical and Biological Engineering and Computing, 57(3):623-632[DOI:10.1007/s11517-018-1907-z]
相关作者
相关机构
京公网安备11010802024621
