心血管相關疾病長期以來一直位居全球人口死因之冠，甚至超越了癌症。根據2019年世界衛生組織（WHO）的統計數據，因心臟相關疾病而死亡的人數超過800萬。心血管相關的影像切割技術因此變得更加重要，它可以協助醫生對心血管疾病的治療，使醫生在診斷上更加高效。
儘管在心臟導管檢查中經常使用連續的冠狀動脈攝影（Coronary Angiograms，CAG），然而，由於收集和標記大量連續X射線影像的成本高昂，大部分的血管分割技術僅針對單一影像進行開發。僅使用單一影像進行分割可能會忽略影像之間的重要時間信息。現有的影片分割方法，即使對於日常生活影片能夠達到高準確度，卻不能直接應用於連續的CAG，因為它們未經訓練以捕捉冠狀動脈影像的特殊結構。
在這項研究中，我們探討了在缺乏標記影片數據的情況下，如何有效進行連續CAG分割的訓練策略。透過廣泛的實驗，我們確認了以下幾點：通用的影片分割網絡確實適用於連續CAG分割；我們成功地訓練了模型，即使在沒有明確標記的影片數據情況下，也取得了令人滿意的結果；此外，我們充分運用了CAG數據的特點，進一步提升了訓練和推斷效果。
實驗結果顯示，在良好的訓練策略下，我們可以使用通用的影片分割網絡對連續CAG數據進行分割，而無需明確的標記CAG影片數據。我們訓練的網絡在最佳情況下，平均F1得分可達85.32，平均區域相似性(region similarity)為74.40，分別比最新技術成果高約0.57和1.11。這些改進有助於提高CAG中的血管分割整體效能和準確性，突顯了我們的方法在提升冠狀動脈疾病診斷能力方面的潛力。另外在我們的最後一個實驗也驗證了，我們的模型沒有非常依賴第一幀的好壞，可以期待其在醫療實務上的應用。

Although sequential Coronary Angiograms (CAG) are often used in cardiac catheterizations, most vessel segmentation techniques are developed just for single images, owing to the expense of collecting and labeling huge amounts of sequential X-Ray images. The segmentation using single images could cause the overlook of critical temporal information among images. Existing video segmentation methods, even though can achieve high accuracy for daily life videos, cannot be used directly for continuous CAG, because they are not trained to capture the special structure of coronary articles. In this work, we investigate the effective training strategies for sequential CAG segmentation without labeled video data. We performed extensive experiments to answer the following questions: Can general purpose video segmentation networks work for sequential CAG segmentation? Can the models be trained without explicit labeled video data? How to use the characteristic of CAG data for better training and inference? The experimental results show that with good training strategies, one can use the general purpose video segmentation networks for sequential CAG data without explicit labeled CAG video data. The best result of our trained network can achieve average F1 score 85.32% and the average region similarity is 74.40%, which are about 0.57% and 1.11% higher than the state-of-the-art results respectively. These improvements contribute to the overall effectiveness and accuracy of vessel segmentation in CAG, highlighting the potential of our approach in advancing diagnostic capabilities in coronary diseases.

Bibliography
[1] David Berthelot, Nicholas Carlini, Ian Goodfellow, Nicolas Papernot, Avital
Oliver, and Colin A Raffel. MixMatch: A holistic approach to semi-supervised
learning. In H. Wallach, H. Larochelle, A. Beygelzimer, F. d'Alché-Buc,
E. Fox, and R. Garnett, editors, Advances in Neural Information Process-
ing Systems, volume 32. Curran Associates, Inc., 2019.
[2] Sergi Caelles, Kevis-Kokitsi Maninis, Jordi Pont-Tuset, Laura Leal-Taixe,
Daniel Cremers, and Luc Van Gool. One-Shot video object segmentation. In
Proceedings of the IEEE Conference on Computer Vision and Pattern Recog-
nition (CVPR), July 2017.
[3] Emmanuel J. Candès, Xiaodong Li, Yi Ma, and John Wright. Robust principal
component analysis? J. ACM, 58(3), jun 2011.
[4] Fernando Cervantes-Sanchez, Ivan Cruz-Aceves, Arturo Hernandez-Aguirre,
Martha Alicia Hernandez-Gonzalez, and Sergio Eduardo Solorio-Meza. Auto-
matic segmentation of coronary arteries in X-ray angiograms using multiscale
analysis and artificial neural networks. Applied Sciences, 9(24), 2019.
[5] Yuhua Chen, Jordi Pont-Tuset, Alberto Montes, and Luc Van Gool. Blazingly
fast video object segmentation with pixel-wise metric learning. In Proceedings
of the IEEE conference on computer vision and pattern recognition, pages
1189–1198, 2018.
[6] Ho Kei Cheng and Alexander G. Schwing. XMem: Long-term video ob-
ject segmentation with an Atkinson-Shiffrin memory model. In Shai Avidan,
Gabriel Brostow, Moustapha Cissé, Giovanni Maria Farinella, and Tal Has-
sner, editors, Computer Vision – ECCV 2022, pages 640–658, Cham, 2022.
Springer Nature Switzerland.
[7] Ho Kei Cheng, Yu-Wing Tai, and Chi-Keung Tang. Rethinking space-time
networks with improved memory coverage for efficient video object segmenta-
tion. In M. Ranzato, A. Beygelzimer, Y. Dauphin, P.S. Liang, and J. Wort-
man Vaughan, editors, Advances in Neural Information Processing Systems,
volume 34, pages 11781–11794. Curran Associates, Inc., 2021.
[8] Jingchun Cheng, Yi-Hsuan Tsai, Wei-Chih Hung, Shengjin Wang, and Ming-
Hsuan Yang. Fast and accurate online video object segmentation via tracking
parts. In Proceedings of the IEEE conference on computer vision and pattern
recognition, pages 7415–7424, 2018.
[9] Kevin M. Cherry, Brandon Peplinski, Lauren Kim, Shijun Wang, Le Lu, Wei-
dong Zhang, Jianfei Liu, Zhuoshi Wei, and Ronald M. Summers. Sequential
Monte Carlo tracking of the marginal artery by multiple cue fusion and ran-
dom forest regression. Medical Image Analysis, 19(1):164–175, 2015.
[10] L.D. Cohen and T. Deschamps. Grouping connected components using min-
imal path techniques. application to reconstruction of vessels in 2D and 3D
images, 2001.
[11] Huihui Fang, Danni Ai, Weijian Cong, Siyuan Yang, Jianjun Zhu, Yong
Huang, Hong Song, Yongtian Wang, and Jian Yang. Topology optimization
using multiple-possibility fusion for vasculature extraction, 2019.
[12] M. Sabry Hassouna and A. A. Farag. Multistencils fast marching methods: A
highly accurate solution to the Eikonal equation on cartesian domains, 2007.
[13] Kaiming He, Xiangyu Zhang, Shaoqing Ren, and Jian Sun. Deep residual
learning for image recognition. In Proceedings of the IEEE Conference on
Computer Vision and Pattern Recognition (CVPR), June 2016.
[14] Ping Hu, Gang Wang, Xiangfei Kong, Jason Kuen, and Yap-Peng Tan.
Motion-guided cascaded refinement network for video object segmentation.
In Proceedings of the IEEE conference on computer vision and pattern recog-
nition, pages 1400–1409, 2018.
[15] Yuan-Ting Hu, Jia-Bin Huang, and Alexander Schwing. MaskRNN: Instance
level video object segmentation. In I. Guyon, U. Von Luxburg, S. Bengio,
H. Wallach, R. Fergus, S. Vishwanathan, and R. Garnett, editors, Advances
in Neural Information Processing Systems, volume 30. Curran Associates,
Inc., 2017.
[16] Xuhua Huang, Jiarui Xu, Yu-Wing Tai, and Chi-Keung Tang. Fast video ob-
ject segmentation with temporal aggregation network and dynamic template
matching. In Proceedings of the IEEE/CVF conference on computer vision
and pattern recognition, pages 8879–8889, 2020.
[17] Licheng Jiao, Ruohan Zhang, Fang Liu, Shuyuan Yang, Biao Hou, Lingling
Li, and Xu Tang. New generation deep learning for video object detection: A
survey, 2021.
[18] Mingxin Jin, Dongdong Hao, Song Ding, and Binjie Qin. Low-rank and sparse
decomposition with spatially adaptive filtering for sequential segmentation of
2D+ t vessels, 2018.
[19] Mingxin Jin, Rong Li, Jian Jiang, and Binjie Qin. Extracting contrast-filled
vessels in X-ray angiography by graduated RPCA with motion coherency
constraint, 2017.
[20] Tae Joon Jun, Jihoon Kweon, Young-Hak Kim, and Daeyoung Kim. T-Net:
Nested encoder–decoder architecture for the main vessel segmentation in coro-
nary angiography. Neural Networks, 128:216–233, 2020.
[21] Ahsan Khawaja, Tariq M. Khan, Khuram Naveed, Syed Saud Naqvi,
Naveed Ur Rehman, and Syed Junaid Nawaz. An improved retinal vessel seg-
mentation framework using frangi filter coupled with the probabilistic patch
based denoiser. IEEE Access, 7:164344–164361, 2019.
[22] Diederik P Kingma and Jimmy Ba. Adam: A method for stochastic opti-
mization, 2014.
[23] Xiaoxiao Li and Chen Change Loy. Video object segmentation with joint
re-identification and attention-aware mask propagation. In Proceedings of the
European conference on computer vision (ECCV), pages 90–105, 2018.
[24] Dongxue Liang, Jing Qiu, Lu Wang, Xiaolei Yin, Junhui Xing, Zhiyun Yang,
Jiangzeng Dong, and Zhaoyuan Ma. Coronary angiography video segmen-
tation method for assisting cardiovascular disease interventional treatment.
BMC medical imaging, 20:1–8, 2020.
[25] Yanhui Liang, Fusheng Wang, Darren Treanor, Derek Magee, George
Teodoro, Yangyang Zhu, and Jun Kong. A 3D primary vessel reconstruc-
tion framework with serial microscopy images. In Nassir Navab, Joachim
Hornegger, William M. Wells, and Alejandro F. Frangi, editors, Medical Im-
age Computing and Computer-Assisted Intervention – MICCAI 2015, pages
251–259, Cham, 2015. Springer International Publishing.
[26] Si Liu, Tianrui Hui, Shaofei Huang, Yunchao Wei, Bo Li, and Guanbin Li.
Cross-modal progressive comprehension for referring segmentation. IEEE
Transactions on Pattern Analysis and Machine Intelligence, 44(9):4761–4775,
2021.
[27] Ilya Loshchilov and Frank Hutter. Decoupled weight decay regularization.
CoRR, abs/1711.05101, 2017.
[28] Xiankai Lu, Wenguan Wang, Jianbing Shen, David Crandall, and Jiebo
Luo. Zero-shot video object segmentation with co-attention siamese networks.
IEEE transactions on pattern analysis and machine intelligence, 44(4):2228–
2242, 2020.
[29] Xiankai Lu, Wenguan Wang, Jianbing Shen, David J Crandall, and Luc
Van Gool. Segmenting objects from relational visual data. IEEE transac-
tions on pattern analysis and machine intelligence, 44(11):7885–7897, 2021.
[30] Hua Ma, Ayla Hoogendoorn, Evelyn Regar, Wiro J Niessen, and Theo van
Walsum. Automatic online layer separation for vessel enhancement in X-ray
angiograms for percutaneous coronary interventions, 2017.
[31] K.-K. Maninis, S. Caelles, Y. Chen, J. Pont-Tuset, L. Leal-Taixé, D. Cre-
mers, and L. Van Gool. Video object segmentation without temporal infor-
mation. IEEE Transactions on Pattern Analysis and Machine Intelligence,
41(6):1515–1530, 2019.
[32] Tim Meinhardt and Laura Leal-Taixé. Make One-Shot video object segmen-
tation efficient again. In H. Larochelle, M. Ranzato, R. Hadsell, M.F. Balcan,
and H. Lin, editors, Advances in Neural Information Processing Systems, vol-
ume 33, pages 10607–10619. Curran Associates, Inc., 2020.
[33] Bo Miao, Mohammed Bennamoun, Yongsheng Gao, and Ajmal Mian. Self-
supervised video object segmentation by motion-aware mask propagation. In
2022 IEEE International Conference on Multimedia and Expo (ICME), pages
1–6, 2022.
[34] Takeru Miyato, Shin-Ichi Maeda, Masanori Koyama, and Shin Ishii. Vir-
tual adversarial training: A regularization method for supervised and semi-
supervised learning. IEEE Transactions on Pattern Analysis and Machine
Intelligence, 41(8):1979–1993, 2019.
[35] Sara Moccia, Elena De Momi, Sara El Hadji, and Leonardo S. Mattos. Blood
vessel segmentation algorithms — review of methods, datasets and evaluation
metrics. Computer Methods and Programs in Biomedicine, 158:71–91, 2018.
[36] National Taiwan University Medical Imaging. CAG biomedical-images.
https://scidm.nchc.org.tw/dataset/cag, 2021.
[37] Koen Nieman, Matthijs Oudkerk, Benno J Rensing, Peter van Ooijen, Arie
Munne, Robert-Jan van Geuns, and Pim J de Feyter. Coronary angiography
with multi-slice computed tomography. The Lancet, 357(9256):599–603, 2001.
[38] Seoung Wug Oh, Joon-Young Lee, Kalyan Sunkavalli, and Seon Joo Kim. Fast
video object segmentation by reference-guided mask propagation. In Proceed-
ings of the IEEE conference on computer vision and pattern recognition, pages
7376–7385, 2018.
[39] Seoung Wug Oh, Joon-Young Lee, Ning Xu, and Seon Joo Kim. Video ob-
ject segmentation using space-time memory networks. In Proceedings of the
IEEE/CVF International Conference on Computer Vision, pages 9226–9235,
2019.
[40] Federico Perazzi, Anna Khoreva, Rodrigo Benenson, Bernt Schiele, and
Alexander Sorkine-Hornung. Learning video object segmentation from static
images. In Proceedings of the IEEE conference on computer vision and pattern
recognition, pages 2663–2672, 2017.
[41] Federico Perazzi, Jordi Pont-Tuset, Brian McWilliams, Luc Van Gool, Markus
Gross, and Alexander Sorkine-Hornung. A benchmark dataset and evaluation
methodology for video object segmentation. In Proceedings of the IEEE Con-
ference on Computer Vision and Pattern Recognition (CVPR), June 2016.
[42] Jordi Pont-Tuset, Federico Perazzi, Sergi Caelles, Pablo Arbelaez, Alexander
Sorkine-Hornung, and Luc Van Gool. The 2017 DAVIS challenge on video
object segmentation. CoRR, abs/1704.00675, 2017.
[43] Binjie Qin, Mingxin Jin, Dongdong Hao, Yisong Lv, Qiegen Liu, Yueqi Zhu,
Song Ding, Jun Zhao, and Baowei Fei. Accurate vessel extraction via tensor
completion of background layer in X-ray coronary angiograms, 2019.
[44] Binjie Qin, Haohao Mao, Yiming Liu, Jun Zhao, Yisong Lv, Yueqi Zhu, Song
Ding, and Xu Chen. Robust PCA unrolling network for super-resolution vessel
extraction in X-ray coronary angiography. IEEE Transactions on Medical
Imaging, 41(11):3087–3098, 2022.
[45] Andreas Robinson, Felix Järemo Lawin, Martin Danelljan, Fahad Shahbaz
Khan, and Michael Felsberg. Learning fast and robust target models for
video object segmentation. CoRR, abs/2003.00908, 2020.
[46] Olaf Ronneberger, Philipp Fischer, and Thomas Brox. U-Net: Convolu-
tional networks for biomedical image segmentation. In Nassir Navab, Joachim
Hornegger, William M. Wells, and Alejandro F. Frangi, editors, Medical Im-
age Computing and Computer-Assisted Intervention – MICCAI 2015, pages
234–241, Cham, 2015. Springer International Publishing.
[47] Hackjoon Shim, Dongjin Kwon, Il Dong Yun, and Sang Uk Lee. Robust seg-
mentation of cerebral arterial segments by a sequential Monte Carlo method:
Particle filtering, 2006. Medical Image Segmentation Special Issue.
[48] João Lourenço Silva, Miguel Nobre Menezes, Tiago Rodrigues, Beatriz Silva,
Fausto J. Pinto, and Arlindo L. Oliveira. Encoder-Decoder architectures for
clinically relevant coronary artery segmentation, Jun 2021.
[49] Oren Solomon, Regev Cohen, Yi Zhang, Yi Yang, Qiong He, Jianwen Luo,
Ruud JG van Sloun, and Yonina C Eldar. Deep unfolded robust PCA with
application to clutter suppression in ultrasound, 2019.
[50] Shuang Song, Chenbing Du, Danni Ai, Yong Huang, Hong Song, Yongtian
Wang, and Jian Yang. Spatio-temporal constrained online layer separation
for vascular enhancement in X-ray angiographic image sequence, 2019.
[51] Hui Tang, Theo Van Walsum, Robbert S Van Onkelen, Stefan Klein, Rein-
hard Hameeteman, Michiel Schaap, Quirijn JA Van den Bouwhuijsen, Jacque-
line CM Witteman, Aad Van der Lugt, Lucas J van Vliet, et al. Multispectral
MRI centerline tracking in carotid arteries, 2011.
[52] Antti Tarvainen and Harri Valpola. Mean teachers are better role models:
Weight-averaged consistency targets improve semi-supervised deep learning
results. In I. Guyon, U. Von Luxburg, S. Bengio, H. Wallach, R. Fergus,
S. Vishwanathan, and R. Garnett, editors, Advances in Neural Information
Processing Systems, volume 30. Curran Associates, Inc., 2017.
[53] Carles Ventura, Miriam Bellver, Andreu Girbau, Amaia Salvador, Ferran
Marques, and Xavier Giro-i Nieto. Rvos: End-to-end recurrent network for
video object segmentation. In Proceedings of the IEEE/CVF conference on
computer vision and pattern recognition, pages 5277–5286, 2019.
[54] Paul Voigtlaender, Yuning Chai, Florian Schroff, Hartwig Adam, Bastian
Leibe, and Liang-Chieh Chen. Feelvos: Fast end-to-end embedding learning
for video object segmentation. In Proceedings of the IEEE/CVF Conference
on Computer Vision and Pattern Recognition, pages 9481–9490, 2019.
[55] Paul Voigtlaender and Bastian Leibe. Online adaptation of convolutional
neural networks for video object segmentation. CoRR, abs/1706.09364, 2017.
[56] Chien-Yao Wang, Alexey Bochkovskiy, and Hong-Yuan Mark Liao. YOLOv7:
Trainable bag-of-freebies sets new state-of-the-art for real-time object detec-
tors. In Proceedings of the IEEE/CVF Conference on Computer Vision and
Pattern Recognition, pages 7464–7475, 2023.
[57] Qiang Wang, Li Zhang, Luca Bertinetto, Weiming Hu, and Philip HS Torr.
Fast online object tracking and segmentation: A unifying approach. In Pro-
ceedings of the IEEE/CVF conference on Computer Vision and Pattern Recog-
nition, pages 1328–1338, 2019.
[58] X Wang, T Heimann, P Lo, M Sumkauskaite, M Puderbach, M de Bruijne,
H P Meinzer, and I Wegner. Statistical tracking of tree-like tubular struc-
tures with efficient branching detection in 3D medical image data. Physics in
Medicine Biology, 57(16):5325, aug 2012.
[59] Shaoyan Xia, Haogang Zhu, Xiaoli Liu, Ming Gong, Xiaoyong Huang, Lei
Xu, Hongjia Zhang, and Jialong Guo. Vessel segmentation of X-ray coronary
angiographic image sequence, 2019.
[60] Qizhe Xie, Zihang Dai, Eduard Hovy, Thang Luong, and Quoc Le. Unsuper-
vised data augmentation for consistency training. In H. Larochelle, M. Ran-
zato, R. Hadsell, M.F. Balcan, and H. Lin, editors, Advances in Neural Infor-
mation Processing Systems, volume 33, pages 6256–6268. Curran Associates,
Inc., 2020.
[61] Ning Xu, Linjie Yang, Yuchen Fan, Dingcheng Yue, Yuchen Liang, Jianchao
Yang, and Thomas S. Huang. Youtube-vos: A large-scale video object seg-
mentation benchmark. CoRR, abs/1809.03327, 2018.
[62] Su Yang, Jihoon Kweon, Jae-Hyung Roh, Jae-Hwan Lee, Heejun Kang, Lae-
Jeong Park, Dong Jun Kim, Hyeonkyeong Yang, Jaehee Hur, Do-Yoon Kang,
et al. Deep learning segmentation of major vessels in X-ray coronary angiog-
raphy. Scientific reports, 9(1):16897, 2019.
[63] Zongxin Yang, Yunchao Wei, and Yi Yang. Collaborative video object seg-
mentation by foreground-background integration. In Andrea Vedaldi, Horst
Bischof, Thomas Brox, and Jan-Michael Frahm, editors, Computer Vision –
ECCV 2020, pages 332–348, Cham, 2020. Springer International Publishing.
[64] Zongxin Yang, Yunchao Wei, and Yi Yang. Associating objects with trans-
formers for video object segmentation. In M. Ranzato, A. Beygelzimer,
Y. Dauphin, P.S. Liang, and J. Wortman Vaughan, editors, Advances in
Neural Information Processing Systems, volume 34, pages 2491–2502. Cur-
ran Associates, Inc., 2021.
[65] Yizhuo Zhang, Zhirong Wu, Houwen Peng, and Stephen Lin. A transductive
approach for video object segmentation. In Proceedings of the IEEE/CVF
Conference on Computer Vision and Pattern Recognition, pages 6949–6958,
2020.
[66] Tianfei Zhou, Wenguan Wang, Ender Konukoglu, and Luc Van Gool. Re-
thinking semantic segmentation: A prototype view. In Proceedings of
the IEEE/CVF Conference on Computer Vision and Pattern Recognition
(CVPR), pages 2582–2593, June 2022.
[67] Xiangjun Zhu, Zhong Xue, Xin Gao, Yisheng Zhu, and Stephen T.C. Wong.
VOLES: Vascularity-oriented level set algorithm for pulmonary vessel seg-
mentation in image guided intervention therapy, 2009.
[68] Xiliang Zhu, Zhaoyun Cheng, Sheng Wang, Xianjie Chen, and Guoqing Lu.
Coronary angiography image segmentation based on PSPNet. Computer
Methods and Programs in Biomedicine, 200:105897, 2021.