Automatic stent reconstruction in optical coherence tomography based on a deep convolutional model.

Intravascular optical coherence tomography (IVOCT) can accurately assess stent apposition and expansion, thus enabling the optimisation of a stenting procedure to minimize the risk of device failure. This paper presents a deep convolutional based model for automatic detection and segmentation of stent struts. The input of pseudo-3D images aggregated the information from adjacent frames to refine the probability of strut detection. In addition, multi-scale shortcut connections were implemented to minimize the loss of spatial resolution and refine the segmentation of strut contours. After training, the model was independently tested in 21,363 cross-sectional images from 170 IVOCT image pullbacks. The proposed model obtained excellent segmentation (0.907 Dice and 0.838 Jaccard) and detection metrics (0.943 precision, 0.940 recall and 0.936 F1-score), significantly better than conventional features-based algorithms. This performance was robust and homogenous among IVOCT pullbacks with different sources of acquisition (clinical centres, imaging operators, type of stent, time of acquisition and challenging scenarios). In addition, excellent agreement between the model and a commercialized software was observed in the quantification of clinically relevant parameters. In conclusion, the deep-convolutional model can accurately detect stent struts in IVOCT images, thus enabling the fully-automatic quantification of stent parameters in an extremely short time. It might facilitate the application of quantitative IVOCT analysis in real-world clinical scenarios.

[1]  Fuhua Yan,et al.  Diagnostic accuracy of intracoronary optical coherence tomography-derived fractional flow reserve for assessment of coronary stenosis severity. , 2019, EuroIntervention : journal of EuroPCR in collaboration with the Working Group on Interventional Cardiology of the European Society of Cardiology.

[2]  Andrew M. Rollins,et al.  Automated stent coverage analysis in intravascular OCT (IVOCT) image volumes using a support vector machine and mesh growing. , 2019, Biomedical optics express.

[3]  P. Seferovic,et al.  2018 ESC/EACTS Guidelines on myocardial revascularization. , 2019, EuroIntervention : journal of EuroPCR in collaboration with the Working Group on Interventional Cardiology of the European Society of Cardiology.

[4]  Yundai Chen,et al.  First Presentation of Integration of Intravascular Optical Coherence Tomography and Computational Fractional Flow Reserve , 2018, The International Journal of Cardiovascular Imaging.

[5]  Antonio Colombo,et al.  Clinical use of intracoronary imaging. Part 1: guidance and optimization of coronary interventions. An expert consensus document of the European Association of Percutaneous Cardiovascular Interventions , 2018, European heart journal.

[6]  Volkmar Falk,et al.  2018 ESC/EACTS Guidelines on myocardial revascularization. , 2018, European heart journal.

[7]  Trevor Darrell,et al.  Deep Layer Aggregation , 2017, 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition.

[8]  A. Kirtane,et al.  Characteristics of early versus late in-stent restenosis in second-generation drug-eluting stents: an optical coherence tomography study. , 2017, EuroIntervention : journal of EuroPCR in collaboration with the Working Group on Interventional Cardiology of the European Society of Cardiology.

[9]  Deniz Erdogmus,et al.  Tversky Loss Function for Image Segmentation Using 3D Fully Convolutional Deep Networks , 2017, MLMI@MICCAI.

[10]  I. Reid,et al.  RefineNet: Multi-path Refinement Networks for High-Resolution Semantic Segmentation , 2016, 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[11]  Olivier Morel,et al.  Optical Coherence Tomography to Optimize Results of Percutaneous Coronary Intervention in Patients with Non–ST-Elevation Acute Coronary Syndrome: Results of the Multicenter, Randomized DOCTORS Study (Does Optical Coherence Tomography Optimize Results of Stenting) , 2016, Circulation.

[12]  Hongki Yoo,et al.  Automated detection of vessel lumen and stent struts in intravascular optical coherence tomography to evaluate stent apposition and neointimal coverage. , 2016, Medical physics.

[13]  Jian Sun,et al.  Deep Residual Learning for Image Recognition , 2015, 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[14]  Thomas Brox,et al.  U-Net: Convolutional Networks for Biomedical Image Segmentation , 2015, MICCAI.

[15]  Andrew M. Rollins,et al.  3-D Stent Detection in Intravascular OCT Using a Bayesian Network and Graph Search , 2015, IEEE Transactions on Medical Imaging.

[16]  Sergey Ioffe,et al.  Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift , 2015, ICML.

[17]  Trevor Darrell,et al.  Fully convolutional networks for semantic segmentation , 2014, Computer Vision and Pattern Recognition.

[18]  Seung‐Jung Park,et al.  Intravascular ultrasound assessment of optimal stent area to prevent in‐stent restenosis after zotarolimus‐, everolimus‐, and sirolimus‐eluting stent implantation , 2014, Catheterization and cardiovascular interventions : official journal of the Society for Cardiac Angiography & Interventions.

[19]  Ryo Torii,et al.  Incomplete Stent Apposition Causes High Shear Flow Disturbances and Delay in Neointimal Coverage as a Function of Strut to Wall Detachment Distance: Implications for the Management of Incomplete Stent Apposition , 2014, Circulation. Cardiovascular interventions.

[20]  Geoffrey E. Hinton,et al.  ImageNet classification with deep convolutional neural networks , 2012, Commun. ACM.

[21]  Andrew M. Rollins,et al.  Automatic stent detection in intravascular OCT images using bagged decision trees , 2012, Biomedical optics express.

[22]  Johan H. C. Reiber,et al.  Automatic stent strut detection in intravascular optical coherence tomographic pullback runs , 2012, The International Journal of Cardiovascular Imaging.

[23]  P. Serruys,et al.  Reproducibility of qualitative assessment of stent struts coverage by optical coherence tomography , 2012, The International Journal of Cardiovascular Imaging.

[24]  Juan Luis Gutiérrez-Chico,et al.  Optical coherence tomography: from research to practice , 2012, European heart journal cardiovascular Imaging.

[25]  P. Serruys,et al.  Vascular Tissue Reaction to Acute Malapposition in Human Coronary Arteries: Sequential Assessment With Optical Coherence Tomography , 2012, Circulation. Cardiovascular interventions.

[26]  W. Desmet,et al.  Automatic segmentation of in-vivo intra-coronary optical coherence tomography images to assess stent strut apposition and coverage , 2012, The International Journal of Cardiovascular Imaging.

[27]  Stavros Tsantis,et al.  Automatic vessel lumen segmentation and stent strut detection in intravascular optical coherence tomography. , 2011, Medical physics.

[28]  Gary S. Mintz,et al.  Comprehensive Intravascular Ultrasound Assessment of Stent Area and Its Impact on Restenosis and Adverse Cardiac Events in 403 Patients With Unprotected Left Main Disease , 2011, Circulation. Cardiovascular interventions.

[29]  P. Serruys,et al.  Delayed coverage in malapposed and side-branch struts with respect to well-apposed struts in drug-eluting stents: in vivo assessment with optical coherence tomography. , 2011, Circulation.

[30]  Seung‐Jung Park,et al.  Mechanisms of In-Stent Restenosis After Drug-Eluting Stent Implantation: Intravascular Ultrasound Analysis , 2011, Circulation. Cardiovascular interventions.

[31]  C. Mathers,et al.  Global and regional causes of death. , 2009, British medical bulletin.

[32]  A. Rollins,et al.  Intracoronary optical coherence tomography: a comprehensive review clinical and research applications. , 2009, JACC. Cardiovascular interventions.

[33]  R. Virmani,et al.  Pathological Correlates of Late Drug-Eluting Stent Thrombosis: Strut Coverage as a Marker of Endothelialization , 2007, Circulation.

[34]  Guido Gerig,et al.  User-guided 3D active contour segmentation of anatomical structures: Significantly improved efficiency and reliability , 2006, NeuroImage.

[35]  Stéphane G Carlier,et al.  Clinical researchInterventional cardiologyStent underexpansion and residual reference segment stenosis are related to stent thrombosis after sirolimus-eluting stent implantation: An intravascular ultrasound study , 2005 .

[36]  P. Fitzgerald,et al.  Impact of final stent dimensions on long-term results following sirolimus-eluting stent implantation: serial intravascular ultrasound analysis from the sirius trial. , 2004, Journal of the American College of Cardiology.

[37]  Antonio Colombo,et al.  Terminology for high-risk and vulnerable coronary artery plaques. Report of a meeting on the vulnerable plaque, June 17 and 18, 2003, Santorini, Greece. , 2004, European heart journal.

[38]  R. Virmani,et al.  Lessons from sudden coronary death: a comprehensive morphological classification scheme for atherosclerotic lesions. , 2000, Arteriosclerosis, thrombosis, and vascular biology.

[39]  Alan D. Lopez,et al.  Alternative projections of mortality and disability by cause 1990–2020: Global Burden of Disease Study , 1997, The Lancet.

[40]  Alan D. Lopez,et al.  Mortality by cause for eight regions of the world: Global Burden of Disease Study , 1997, The Lancet.

[41]  W D Wagner,et al.  A definition of advanced types of atherosclerotic lesions and a histological classification of atherosclerosis. A report from the Committee on Vascular Lesions of the Council on Arteriosclerosis, American Heart Association. , 1995, Arteriosclerosis, thrombosis, and vascular biology.

[42]  W D Wagner,et al.  A definition of initial, fatty streak, and intermediate lesions of atherosclerosis. A report from the Committee on Vascular Lesions of the Council on Arteriosclerosis, American Heart Association. , 1994, Arteriosclerosis and thrombosis : a journal of vascular biology.

[43]  Lawrence D. Jackel,et al.  Backpropagation Applied to Handwritten Zip Code Recognition , 1989, Neural Computation.

[44]  Nitish Srivastava,et al.  Dropout: a simple way to prevent neural networks from overfitting , 2014, J. Mach. Learn. Res..

[45]  Andrew L. Maas.  Rectifier Nonlinearities Improve Neural Network Acoustic Models , 2013 .

[46]  D H Blankenhorn,et al.  A definition of the intima of human arteries and of its atherosclerosis-prone regions. A report from the Committee on Vascular Lesions of the Council on Arteriosclerosis, American Heart Association. , 1992, Circulation.