Research on the IVUS Border Detection Method Based on Improved TransUnet
Wang Yuanyuan1&, Dong Fang1&, Shang Lina1, Zhang Cui1, Xia Shunren1,2#*
1(School of Information and Electrical Engineering, Hangzhou City University, Hangzhou 310000, China) 2(Key Laboratory of Biomedical Engineering of Ministry of Education, Zhejiang University, Hangzhou 310000, China)
Abstract:Intravascular ultrasound (IVUS) is a popular imaging technique that can observe the internal structure of blood vessels. The extraction of lumen and media-adventitia border in the IVUS image is vitally important for the accurate diagnosis of coronary atherosclerosis. To solve the problems of complex structure, low contrast and difficult boundary detection in the IVUS images, a modified TransUnet network was proposed in this paper to achieve pixel-wise classification results. Firstly, to overcome the difficulty of IVUS image border detection, data augmentation strategy is performed based on four kinds of image structure models, including edge vessel, vessel bifurcation, guide wire artifact and shadow. Then, a position named Polar-bias was proposed in the TransUnet network. The Polar-bias combines the ring structure distribution characteristics of IVUS images. The modified TransUnet network was applied to classify the IVUS images in pixel-wise level. Finally, the classified results were employed to optimize the external force field of GVF Snake model, which was used to detect the lumen and media-adventitia border in IVUS images. Two public datasets with different ultrasonic imaging center frequencies (a total of 512 images) were used for testing and verification, and three evaluation indexes of JMard measure (JM), Hausdorff distance (HD) and percentage of area difference (PAD) were introduced. The proposed method has achieved JM (0.87), HD (0.87) and PAD (0.18) in dataset A, and JM(0.91), HD(0.25) and PAD(0.08) in dataset B. The experimental results demonstrated that the proposed method outperformed other state-of-the-art ones in the border detection of two datasets.
王媛媛, 董芳, 尚丽娜, 张翠, 夏顺仁. 基于改进TransUnet网络的血管内超声图像边界提取方法研究[J]. 中国生物医学工程学报, 2023, 42(1): 41-50.
Wang Yuanyuan, Dong Fang, Shang Lina, Zhang Cui, Xia Shunren. Research on the IVUS Border Detection Method Based on Improved TransUnet. Chinese Journal of Biomedical Engineering, 2023, 42(1): 41-50.
[1] 国家心血管病中心.《中国心血管健康与疾病报告2019》[M],上海:科学出版社, 2020: 135-137. [2] Ong HG, Kang SJ, Mintz GS. Value of intravascular ultrasound in guiding coronary interventions [J]. Echocardiography, 2018, 35(4): 520-533. [3] 霍勇, 葛均波, 方唯一, 等. 冠状动脉疾病影像学[M], 北京:北京大学医学出版社, 2015: 28-39. [4] Slager CJ, Wentzel JJ, Schuurbiers JC, et al. True 3-dimensional reconstruction of coronary arteries in patients by fusion of angiography and IVUS (ANGUS) and its quantitative validation [J]. Circulation, 2000, 102(5): 511-516. [5] Sonka M, Zhang X, Siebes M, et al. Segmentation of intravascular ultrasound images: A knowledge-based approach [J]. IEEE T Med Imaging, 1995, 14(4): 719-732. [6] Von BC, Di MC, Li W, et al. Morphometric analysis in three-dimensional intracoronary ultrasound: an in vitro and in vivo study performed with a novel system for the contour detection of lumen and plaque [J]. Am Heart J, 1996, 132(3): 516-527. [7] Sun S, Sonka M, Beichel RR. Graph-based IVUS segmentation with efficient computer-aided refinement [J]. IEEE Trans Med Imaging, 2013, 32(8): 1536-1549. [8] Plissiti ME, Fotiadis DI, Michalis LK, et al. An automated method for lumen and media-adventitia border detection in a sequence of IVUS frames [J]. IEEE Trans Inf Technol Biomed, 2004, 8(2): 131-141. [9] Zhu Xinjian, Zhang Pengfei, Shao Jinhua, et al. A snake-based method for segmentation of intravascular ultrasound images and its in vivo validation [J]. Ultrasonics, 2011, 51(2): 181-189. [10] Ciompi F, Pujol O, Gatta C, et al. HoliMAb: a holistic approach for Media–Adventitia border detection in intravascular ultrasound [J]. Med Image Anal, 2012, 16(6): 1085-1100. [11] Faraji M, Cheng I, Naudin I, et al. Segmentation of arterial walls in intravascular ultrasound cross-sectional images using extremal region selection[J]. Ultrasonics, 2018, 84: 356-365. [12] Huang Yi, Yan Wenjun, Xia Menghua, et al. Vessel membrane segmentation and calcification location in intravascular ultrasound images using a region detector and an effective selection strategy [J]. Comput Methods Programs Biomed, 2020, 189(105):339-351. [13] Jodas D S, Pereira A S, Tavares J. Automatic segmentation of the lumen region in intravascular images of the coronary artery[J]. Med Image Anal, 2017, 40(2017):60-79. [14] Xia Zhaode, Yan Weijun, Huang Liheng, et al. IVUS image segmentation using superpixel-wise fuzzy clustering and level set evolution [J]. Appl Sci, 2019, 9(22): 1-14. [15] Kumar RB, Marikkannu P, Vetrivel K. Plaque mass border detection and classification in intravascular ultrasound images using plaque mass weight-K-nearest-neighbor classifier [J]. J Med Imaging Health Inform, 2019, 9(9): 1838-1843. [16] Wang Yuanyuan, Peng Wenxian, Qiu Chenhui, et al. Fractional-order Darwinian PSO-based feature selection for media-adventitia border detection in intravascular ultrasound images [J]. Ultrasonics, 2018, 92(1): 1-12. [17] Wang Yuanyuan, Qiu Chenhui, Jiang Jun, et al. Detecting the media-adventitia border in intravascular ultrasound images through a classification-based approach [J]. Ultrasonic Imaging, 2019, 41(2): 1-16. [18] Tong Jijun, Li Kai, Lin Wei, et al. Automatic lumen border detection in IVUS images using dictionary learning and kernel sparse representation[J]. Biomed Signal Process Control, 2021, 66(6):102489. [19] Ronneberger O, Fischer P, Brox T. U-Net: convolutional networks for biomedical image segmentation[C]//International Conference on Medical Image Computing and Computer-Assisted Intervention. Munich: Springer, 2015: 234-241. [20] Yang Ji, Tong Lin, Faraji M, et al. IVUS-Net: an intravascular ultrasound segmentation network [C]// International Conference on Smart Multimedia. Toulon: Springer, 2018: 367-377. [21] Yang Ji, Faraji M, Basu A. Robust segmentation of arterial walls in intravascular ultrasound images using Dual Path U-Net[J]. Ultrasonics, 2019, 96(1): 24-33. [22] Li Yichen, Shen TY, Chen CC, et al. Automatic detection of atherosclerotic plaque and calcification from intravascular ultrasound images by using deep convolutional neural networks[J]. IEEE Trans Ultrason Ferroelectr Freq Control, 2021, 68(5):1762-1726. [23] Xia Menghua, Yan Wenjun, Huang Yi, et al. Extracting membrane borders in IVUS images using a multi-scale feature aggregated u-net[C]//2020 42nd Annual International Conference of the IEEE Engineering in Medicine & Biology Society (EMBC). Montreal: IEEE, 2020: 1650-1653. [24] Li Kai, Tong Jijun, Zhu Xia, et al. Automatic lumen border detection in IVUS images using deep learning model and handcrafted features[J]. Ultrasonic Imaging, 2021, 43(2):324-338. [25] Blanco PJ, Ziemer PG, Bulant CA, et al. Fully automated lumen and vessel contour segmentation in intravascular ultrasound datasets [J]. Med Image Anal, 2022, 75(2022): 102262. [26] Deepak M, Santanu C, Mukul S, et al. Ultrasound image segmentation: a deeply supervised network with attention to boundaries[J]. IEEE Trans Biomed Eng, 2018, 66(6): 1637-1648. [27] 袁绍锋, 杨丰, 徐琳, 等.有条件生成对抗网络的IVUS图像内膜与中-外膜边界检测[J]. 中国生物医学工程学报, 2019, 38(2):146-156. [28] Vaswani A, Shazeer N, Parmar N, et al. Attention is all you need[C]// The 31st Conference on Neural Information Processing Systems. Long Beach: Neural Information Processing Systems. 2017:1-11. [29] Dosovitskiy A, Beyer L, Kolesnikov A, et al. An image is worth 16x16 words: transformers for image recognition at scale[C]//The 9th International Conference on Learning Representations. Vienna: Elsevier, 2021:1-21. [30] Chen Jieneng, Lu Yongyi, Yu Qihang, et al. TransUNet: transformers make strong encoders for medical image segmentation[EB/OL]. arXiv preprint arXiv:2102.04306v1, 2021-02-08/2022-03-29. [31] Balocco S, Gatta C, Ciompi F, et al. Standardized evaluation methodology and reference database for evaluating IVUS image segmentation [J]. Comput Med Imaging Graph, 2014, 38(2): 70-90.