Coronary centerline extraction is an essential technique for X-ray coronary angiography (XCA) image analysis, which provides qualitative and quantitative guidance for percutaneous coronary intervention (PCI). In this paper, an online deep reinforcement learning method for coronary centerline extraction is proposed based on the prior vascular skeleton. Firstly, with XCA image preprocessing (foreground extraction and vessel segmentation) results, the improved ZhangSuen image thinning algorithm is used to rapidly extract the preliminary vascular skeleton network. On this basis, according to the spatial-temporal and morphological continuity of the angiography image sequence, the connectivity of different branches is determined using k-means clustering, and the vessel segments are then grouped, screened, and reconnected to obtain the aorta and its major branches. Finally, using the previous results as prior information, an online Deep Q-Network (DQN) reinforcement learning method is proposed to optimize each branch simultaneously. It comprehensively considers grayscale intensity and eigenvector continuity to achieve the combination of data-driven and model-driven without pre-training. Experimental results on clinical images and the third-party dataset demonstrate that the proposed method can accurately extract, restructure, and optimize the centerline of XCA images with a higher overall accuracy than the existing state-of-the-art methods.
LoweHCOesterleSNKhachigianLM. Coronary instent restenosis: current status and future strategies. J Am Coll Cardiol2002; 39(2): 183–193.
2.
FuZFuZLvC, et al. Robust implementation of foreground extraction and vessel segmentation for X-ray coronary angiography image sequence. arXiv preprint arXiv:220907237, 2022.
3.
KirişliHASchaapMMetzCT, et al. Standardized evaluation framework for evaluating coronary artery stenosis detection, stenosis quantification and lumen segmentation algorithms in computed tomography angiography. Med Image Anal2013; 17(8): 859–876.
4.
QinBJinMHaoD, et al. Accurate vessel extraction via tensor completion of background layer in X-ray coronary angiograms. Pattern Recognit2019; 87: 38–54.
5.
ZhangXDuHSongG, et al. X-ray coronary centerline extraction based on C-Unet and a multifactor reconnection algorithm. Comput Methods Programs Biomed2022; 226: 107114.
6.
ZhangYLuoGWangW, et al. Branch-aware double dqn for centerline extraction in coronary ct angiography. In: Medical Image Computing and Computer Assisted Intervention - MICCAI 2020: 23rd International Conference, Lima, Peru, October 4–8, 2020, Proceedings, Part VI 23. Springer, pp. 35–44.
7.
EihoSQianY. Detection of coronary artery tree using morphological operator. In: Computers in cardiology 1997, Lund, Sweden, 2002, pp.525–528. IEEE.
8.
FangHAiDCongW, et al. Topology optimization using multiple-possibility fusion for vasculature extraction. IEEE Trans Circuits Syst Video Technol2020; 30(2): 442–456.
9.
FangHZhuJAiD, et al. Greedy soft matching for vascular tracking of coronary angiographic image sequences. IEEE Trans Circuits Syst Video Technol2020; 30(5): 1466–1480.
10.
OlabarriagaSDBreeuwerMNiessenWJ. Minimum cost path algorithm for coronary artery central axis tracking in ct images. In: Medical Image Computing and Computer-Assisted Intervention-MICCAI 2003: 6th International Conference, Montreal, Canada, November 15–18, 2003. Proceedings 6. Springer, pp. 687–694.
11.
MetzCTSchaapMWeustinkAC, et al. Coronary centerline extraction from ct coronary angiography images using a minimum cost path approach. Med Phys2009; 36(12): 5568–5579.
12.
DeschampsTCohenLD. Fast extraction of minimal paths in 3D images and applications to virtual endoscopy. Med Image Anal2001; 5(4): 281–299.
13.
LiHYezziA. Vessels as 4-d curves: global minimal 4-d paths to extract 3-d tubular surfaces and centerlines. IEEE Trans Med Imaging2007; 26(9): 1213–1223.
14.
ChenYZhangYYangJ, et al. Curve-like structure extraction using minimal path propagation with backtracking. IEEE Trans Image Process2016; 25(2): 988–1003.
15.
YangGLvTShenY, et al. Vessel structure extraction using constrained minimal path propagation. Artif Intell Med2020; 105: 101846.
16.
AylwardSRBullittE. Initialization, noise, singularities, and scale in height ridge traversal for tubular object centerline extraction. IEEE Trans Med Imaging2002; 21(2): 61–75.
17.
XuYZhangHLiH, et al. An improved algorithm for vessel centerline tracking in coronary angiograms. Comput Methods Programs Biomed2007; 88(2): 131–143.
18.
WangYNarayanaswamyARoysamB. Novel 4-d open-curve active contour and curve completion approach for automated tree structure extraction. In: CVPR 2011, Colorado Springs, CO, USA, 2011, pp.1105–1112. IEEE.
19.
ZhouSLiBWangY, et al. The line- and block-like structures extraction via ingenious snake. Pattern Recognit Lett2018; 112: 324–331.
20.
WolterinkJMvan HamersveltRWViergeverMA, et al. Coronary artery centerline extraction in cardiac CT angiography using a cnn-based orientation classifier. Med Image Anal2019; 51: 46–60.
21.
ZhangPWangFZhengY. Deep reinforcement learning for vessel centerline tracing in multi-modality 3d volumes. In: Medical Image Computing and Computer Assisted Intervention -MICCAI 2018: 21st International Conference, Granada, Spain, September 16–20, 2018, Proceedings, Part IV 11. Springer, pp. 755–763.
22.
YangHChenJChiY, et al. Discriminative coronary artery tracking via 3d cnn in cardiac ct angiography. In: Medical Image Computing and Computer Assisted Intervention - MICCAI 2019: 22nd International Conference, Shenzhen, China, October 13–17, 2019, Proceedings, Part II 22. Springer, pp. 468 –476.
23.
ZhouSKLeHNLuuK, et al. Deep reinforcement learning in medical imaging: a literature review. Med Image Anal2021; 73: 102193.
24.
ZhangTYSuenCY. A fast parallel algorithm for thinning digital patterns. Commun ACM1984; 27(3): 236–239.
25.
YangGKitslaarPFrenayM, et al. Automatic centerline extraction of coronary arteries in coronary computed tomographic angiography. Int J Cardiovasc Imaging2012; 28(4): 921–933.
26.
LiZZhangYLiuG, et al. A robust coronary artery identification and centerline extraction method in angiographies. Biomed Signal Process Control2015; 16: 1–8.
27.
MnihVKavukcuogluKSilverD, et al. Human-level control through deep reinforcement learning. Nature2015; 518(7540): 529–533.
28.
FrangiAFNiessenWJVinckenKL, et al. Multiscale vessel enhancement filtering. In: International conference on medical image computing and computer-assisted intervention, pp.130–137. Springer.
HilditchCJ. Linear skeleton from square cupboards. Mach Intell1969; 6: 403–420.
31.
RosenfeldA. A characterization of parallel thinning algorithms. Inf Control1975; 29(3): 286–291.
32.
PavlidisT. A thinning algorithm for discrete binary images. Comput Graph Image Process1980; 13(2): 142–157.
33.
BrutzerSHöferlinBHeidemannG. Evaluation of background subtraction techniques for video surveillance. In: CVPR 2011, Colorado Springs, CO, USA, pp.1937–1944. IEEE.
34.
SchaapMMetzCTvan WalsumT, et al. Standardized evaluation methodology and reference database for evaluating coronary artery centerline extraction algorithms. Med Image Anal2009; 13(5): 701–714.
35.
FuZFuZGongZ, et al. Optimization for 3D reconstruction of coronary artery tree by two-stage Levenberg-Marquardt algorithm. In: 2021 27th international conference on mechatronics and machine vision in practice (M2VIP), Shanghai, China, pp.84–89. IEEE.
