Introduction: Arterial brain vessel assessment is crucial for the diagnostic process in patients with cerebrovascular disease. Non-invasive neuroimaging techniques, such as time-of-flight (TOF) magnetic resonance angiography (MRA) imaging are applied in the clinical routine to depict arteries. They are, however, only visually assessed. Fully automated vessel segmentation integrated into the clinical routine could facilitate the time-critical diagnosis of vessel abnormalities and might facilitate the identification of valuable biomarkers for cerebrovascular events. In the present work, we developed and validated a new deep learning model for vessel segmentation, coined BRAVE-NET, on a large aggregated dataset of patients with cerebrovascular diseases. Methods: BRAVE-NET is a multiscale 3-D convolutional neural network (CNN) model developed on a dataset of 264 patients from three different studies enrolling patients with cerebrovascular diseases. A context path, dually capturing high- and low-resolution volumes, and deep supervision were implemented. The BRAVE-NET model was compared to a baseline Unet model and variants with only context paths and deep supervision, respectively. The models were developed and validated using high-quality manual labels as ground truth. Next to precision and recall, the performance was assessed quantitatively by Dice coefficient (DSC); average Hausdorff distance (AVD); 95-percentile Hausdorff distance (95HD); and via visual qualitative rating. Results: The BRAVE-NET performance surpassed the other models for arterial brain vessel segmentation with a DSC = 0.931, AVD = 0.165, and 95HD = 29.153. The BRAVE-NET model was also the most resistant toward false labelings as revealed by the visual analysis. The performance improvement is primarily attributed to the integration of the multiscaling context path into the 3-D Unet and to a lesser extent to the deep supervision architectural component. Discussion: We present a new state-of-the-art of arterial brain vessel segmentation tailored to cerebrovascular pathology. We provide an extensive experimental validation of the model using a large aggregated dataset encompassing a large variability of cerebrovascular disease and an external set of healthy volunteers. The framework provides the technological foundation for improving the clinical workflow and can serve as a biomarker extraction tool in cerebrovascular diseases.

Berlin School of Mind and Brain Mind Brain Body Institute Humboldt Universität Berlin Berlin Germany

Berlin Ultrahigh Field Facility Max Delbrück Center for Molecular Medicine in the Helmholtz Association Berlin Germany

Biomedical Innovation Academy Berlin Institute of Health Berlin Germany

Centre for Stroke Research Berlin Charité Universitätsmedizin Berlin Berlin Germany

CLAIM Charité Lab for Artificial Intelligence in Medicine Charité Universitätsmedizin Berlin Berlin Germany

Department Biomedical Engineering Medical Image Analysis Center AG University of Basel Basel Switzerland

Department of Neurology 1st Faculty of Medicine Centre of Clinical Neuroscience General University Hospital Prague Charles University Prague Czechia

Department of Neurology Max Planck Institute for Human Cognitive and Brain Sciences Leipzig Germany

Faculty of Computing Engineering and the Built Environment School of Computing and Digital Technology Birmingham City University Birmingham United Kingdom

Johanna Etienne Hospital Neuss Germany

Research Studio Data Science Research Studios Austria Salzburg Austria

Zobrazit více v PubMed

Abadi M., Agarwal A., Barham P., Brevdo E., Chen Z., Citro C., et al. (2015). TensorFlow: large-scale machine learning on heterogeneous distributed systems. arXiv 1603.04467.

Alom M. Z., Hasan M., Yakopcic C., Taha T. M., Asari V. K. (2018). Recurrent residual convolutional neural network based on U-net (R2U-Net) for medical image segmentation. arXiv 1802.06955. PubMed

Babin D., PiŽurica A., Vylder J. D., Vansteenkiste E., Philips W. (2013). Brain blood vessel segmentation using line-shaped profiles. Phys. Med. Biol. 58, 8041–8061. 10.1088/0031-9155/58/22/8041 PubMed DOI

Badrinarayanan V., Kendall A., Cipolla R. (2017). SegNet: a deep convolutional encoder-decoder architecture for image segmentation. IEEE Trans. Pattern Anal. Mach. Intell. 39, 2481–2495. 10.1109/TPAMI.2016.2644615 PubMed DOI

Bullitt E., Zeng D., Gerig G., Aylward S., Joshi S., Smith J. K., et al. . (2005). Vessel tortuosity and brain tumor malignancy: a blinded study. Acad. Radiol. 12, 1232–1240. 10.1016/j.acra.2005.05.027 PubMed DOI PMC

Chen L., Xie Y., Sun J., Balu N., Mossa-Basha M., Pimentel K., et al. (2017). Y-net: 3D intracranial artery segmentation using a convolutional autoencoder. arXiv 1712.07194.

Chen L.-C., Papandreou G., Kokkinos I., Murphy K., Yuille A. L. (2014). Semantic image segmentation with deep convolutional nets and fully connected CRFs. arXiv 1412.7062. PubMed

Dengler N. F., Madai V. I., Wuerfel J., von Samson-Himmelstjerna F. C., Dusek P., Niendorf T., et al. . (2016). Moyamoya vessel pathology imaged by ultra-high-field magnetic resonance imaging at 7.0 T. J. Stroke Cerebrovasc. Dis. 25, 1544–1551. 10.1016/j.jstrokecerebrovasdis.2016.01.041 PubMed DOI

Dong H., Yang G., Liu F., Mo Y., Guo Y. (2017). Automatic brain tumor detection and segmentation using U-net based fully convolutional networks, in Medical Image Understanding and Analysis Communications in Computer and Information Science, eds Valdés Hernández M., González-Castro V. (Oxford: Springer International Publishing; ), 506–517.

Dou Q., Chen H., Jin Y., Yu L., Qin J., Heng P.-A. (2016). 3D deeply supervised network for automatic liver segmentation from CT volumes, in Medical Image Computing and Computer-Assisted Intervention—MICCAI 2016, Lecture Notes in Computer Science, eds Ourselin S., Joskowicz L., Sabuncu M. R., Unal G., Wells W. (Athens: Springer International Publishing; ), 149–157.

Dutra B. G., Tolhuisen M. L., Alves H. C. B. R., Treurniet K. M., Kappelhof M., Yoo A. J., et al. . (2019). Thrombus imaging characteristics and outcomes in acute ischemic stroke patients undergoing endovascular treatment. Stroke 50, 2057–2064. 10.1161/STROKEAHA.118.024247 PubMed DOI

Folle L., Vesal S., Ravikumar N., Maier A. (2019). Dilated deeply supervised networks for hippocampus segmentation in MRI, in Bildverarbeitung für die Medizin 2019 Informatik Aktuell, eds Handels H., Deserno T. M., Maier A., Maier-Hein K. H., Palm C., Tolxdorff T. (Wiesbaden: Springer Fachmedien Wiesbaden; ), 68–73.

Glorot X., Bengio Y. (2010). Understanding the difficulty of training deep feedforward neural networks, in Proceedings of the Thirteenth International Conference on Artificial Intelligence and Statistics (Sardinia: ), 249–256.

Gutierrez J., Cheung K., Bagci A., Rundek T., Alperin N., Sacco Ralph L., et al. . (2015). Brain arterial diameters as a risk factor for vascular events. J. Am. Heart Assoc. 4:e002289. 10.1161/JAHA.115.002289 PubMed DOI PMC

Havaei M., Davy A., Warde-Farley D., Biard A., Courville A., Bengio Y., et al. . (2016). Brain tumor segmentation with deep neural networks. arXiv preprint arXiv:1505.03540. PubMed

He J., Deng Z., Zhou L., Wang Y., Qiao Y. (2019). Adaptive pyramid context network for semantic segmentation, in IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) (Long Beach, CA: ), 7511–7520. 10.1109/CVPR.2019.00770 DOI

Higgins D., Madai V. I. (2020). From bit to bedside: a practical framework for artificial intelligence product development in healthcare. Adv. Intell. Syst. [Epub ahead of print]. 10.1002/aisy.202000052 DOI

Hilbert A., Madai V. I., Akay E. M., Aydin O. U., Behland J., Sobesky J., et al. (2020). Ground truth labels for “BRAVE-NET: Fully Automated Arterial Brain Vessel Segmentation In Patients with Cerebrovascular Disease” (Version 1). Zenodo. PubMed PMC

Ho C. W. L., Soon D., Caals K., Kapur J. (2019). Governance of automated image analysis and artificial intelligence analytics in healthcare. Clin. Radiol. 74, 329–337. 10.1016/j.crad.2019.02.005 PubMed DOI

Hofmanninger J., Prayer F., Pan J., Rohrich S., Prosch H., Langs G. (2020). Automatic lung segmentation in routine imaging is a data diversity problem, not a methodology problem. Eur. Radiol. Exp. 4:50. 10.1186/s41747-020-00173-2 PubMed DOI PMC

Hotter B., Pittl S., Ebinger M., Oepen G., Jegzentis K., Kudo K., et al. . (2009). Prospective study on the mismatch concept in acute stroke patients within the first 24 h after symptom onset−1000Plus study. BMC Neurol. 9:60. 10.1186/1471-2377-9-60 PubMed DOI PMC

Huang H., Lanfen L., Ruofeng T., Hongjie H., Qiaowei Z., Yutaro I., et al. (2020). UNet 3+: a full-scale connected UNet for medical image segmentation. arXiv 2004.08790.

Huang Q., Sun J., Ding H., Wang X., Wang G. (2018). Robust liver vessel extraction using 3D U-Net with variant dice loss function. Comput. Biol. Med. 101, 153–162. 10.1016/j.compbiomed.2018.08.018 PubMed DOI

Ioffe S., Szegedy C. (2015). Batch Normalization: accelerating deep network training by reducing internal covariate shift. arXiv 1502.03167.

Kamnitsas K., Ledig C., Newcombe V. F. J., Simpson J. P., Kane A. D., Menon D. K., et al. . (2017). Efficient multi-scale 3D CNN with fully connected CRF for accurate brain lesion segmentation. Med. Image Anal. 36, 61–78. 10.1016/j.media.2016.10.004 PubMed DOI

Kingma D. P., Ba J. (2014). Adam: a method for stochastic optimization. arXiv 1412.6980.

Kokkinos I. (2015). Pushing the boundaries of boundary detection using deep learning. arXiv 1511.07386.

Lee C., Xie S., Gallagher P., Zhang Z., Tu Z. (2015). Deeply-supvervised nets. arXiv 1409.5185.

Lesage D., Angelini E. D., Bloch I., Funka-Lea G. (2009). A review of 3D vessel lumen segmentation techniques: Models, features and extraction schemes. Med. Image Anal. 13, 819–845. 10.1016/j.media.2009.07.011 PubMed DOI

Livne M, Rieger J, Aydin OU, Taha AA, Akay EM, Kossen T. (2019b). Model data for “A U-net deep learning framework for high performance vessel segmentation in patients with cerebrovascular disease.” 10.5281/zenodo.3433818 PubMed DOI PMC

Livne M., Rieger J., Aydin O. U., Taha A. A., Akay E. M., Kossen T., et al. . (2019a). A U-Net deep learning framework for high performance vessel segmentation in patients with cerebrovascular disease. Front. Neurosci. 13:97. 10.3389/fnins.2019.00097 PubMed DOI PMC

Long J., Shelhamer E., Darrell T. (2014). Fully convolutional networks for semantic segmentation. arXiv 1411.4038. PubMed

Martin S. Z., Madai V. I., von Samson-Himmelstjerna F. C., Mutke M. A., Bauer M., Herzig C. X., et al. . (2015). 3D GRASE pulsed arterial spin labeling at multiple inflow times in patients with long arterial transit times: comparison with dynamic susceptibility-weighted contrast-enhanced MRI at 3 tesla. J. Cereb. Blood Flow Metab. 35, 392–401. 10.1038/jcbfm.2014.200 PubMed DOI PMC

Meijs M., Patel A., Leemput S. C., van de Prokop M., Dijk E. J., van Leeuw F.-E., et al. . (2017). Robust segmentation of the full cerebral vasculature in 4D CT of suspected stroke patients. Sci. Rep. 7:15617. 10.1038/s41598-017-15617-w PubMed DOI PMC

Milletari F., Navab N., Ahmadi S.-A. (2016). V-net: fully convolutional neural networks for volumetric medical image segmentation. arXiv 1606.04797.

Mitchell M., Wu S., Zaldivar A., Barnes P., Vasserman L., Hutchinson B., et al. (2019). Model cards for model reporting, in Proceedings of the Conference on Fairness, Accountability, and Transparency FAT* '19 (Atlanta, GA: Association for Computing Machinery; ), 220–229.

Moccia S., De Momi E., El Hadji S., Mattos L. S. (2018). Blood vessel segmentation algorithms—review of methods, datasets and evaluation metrics. Comput. Methods Programs Biomed. 158, 71–91. 10.1016/j.cmpb.2018.02.001 PubMed DOI

Murray N. M., Unberath M., Hager G. D., Hui F. K. (2019). Artificial intelligence to diagnose ischemic stroke and identify large vessel occlusions: a systematic review. J. NeuroInterv. Surg. 12, 156–164. 10.1136/neurintsurg-2019-015135 PubMed DOI

Ni J., Wu J., Wang H., Tong J., Chen Z., Wong K. K. L., et al. . (2020). Global channel attention networks for intracranial vessel segmentation. Comput. Biol. Med. 118:103639. 10.1016/j.compbiomed.2020.103639 PubMed DOI

Norman B., Pedoia V., Majumdar S. (2018). Use of 2D U-net convolutional neural networks for automated cartilage and meniscus segmentation of knee MR imaging data to determine relaxometry and morphometry. Radiology 288, 177–185. 10.1148/radiol.2018172322 PubMed DOI PMC

Passat N., Ronse C., Baruthio J., Armspach J.-P., Maillot C., Jahn C. (2005). Region-growing segmentation of brain vessels: an atlas-based automatic approach. J. Magn. Reson. Imaging 21, 715–725. 10.1002/jmri.20307 PubMed DOI

Patel T. R., Paliwal N., Jaiswal P., Waqas M., Mokin M., Siddiqui A. H., et al. (2020). Multi-resolution CNN for brain vessel segmentation from cerebrovascular images of intracranial aneurysm: a comparison of U-Net and DeepMedic, in Medical Imaging 2020: Computer-Aided Diagnosis International Society for Optics and Photonics (Houston, TX: SPIE Medical Imaging; ), 113142W.

Ronneberger O., Fischer P., Brox T. (2015). U-net: convolutional networks for biomedical image segmentatio, in Medical Image Computing and Computer-Assisted Intervention—MICCAI 2015, Lecture Notes in Computer Science, eds. Navab N., Hornegger J., Wells W. M., Frangi A. F. (Munich: Springer International Publishing; ), 234–241.

Santos E. M. M., Marquering H. A., den Blanken M. D., Berkhemer O. A., Boers A. M. M., Yoo A. J., et al. (2016). Thrombus permeability is associated with improved functional outcome and recanalization in patients with ischemic stroke. Stroke 47, 732–741. 10.1161/STROKEAHA.115.011187 PubMed DOI

Sinha A., Jose D. (2020). Multi-scale self-guided attention for medical image segmentation. arXiv 1906.02849. PubMed

Srivastava N., Hinton G., Krizhevsky A., Sutskever I., Salakhutdinov R. (2014). Dropout: a simple way to prevent neural networks from overfitting. J. Mach. Learn. Res. 15, 1929–1958. Available online at: https://dl.acm.org/doi/10.5555/2627435.2670313 DOI

Stawiaski J. (2017). A multiscale patch based convolutional network for brain tumor segmentation. arXiv 1710.02316.

Szegedy C., Liu W., Jia Y., Sermanet P., Reed S., Anguelov D., et al. (2015). Going deeper with convolutions, in 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) (Boston, MA: ), 1–9.

Taha A. A., Hanbury A. (2015). Metrics for evaluating 3D medical image segmentation: analysis, selection, and tool. BMC Med. Imaging 15:29. 10.1186/s12880-015-0068-x PubMed DOI PMC

Taher F., Soliman A., Kandil H., Mahmoud A., Shalaby A., Gimelrfarb G., et al. (2020). Accurate segmentation of cerebrovasculature from TOF-MRA images using appearance descriptors. IEEE Access 8, 96139–96149. 10.1109/ACCESS.2020.2982869 DOI

Tetteh G., Efremov V., Forkert N. D., Schneider M., Kirschke J., Weber B., et al. (2018). DeepVesselNet: vessel segmentation, centerline prediction, and bifurcation detection in 3-D angiographic volumes. arXiv 1803.09340. PubMed PMC

Turc G., Bhogal P., Fischer U., Khatri P., Lobotesis K., Mazighi M., et al. . (2019). European Stroke Organisation (ESO)—European Society for Minimally Invasive Neurological Therapy (ESMINT) guidelines on mechanical thrombectomy in acute ischaemic strokeendorsed by Stroke Alliance for Europe (SAFE). Eur. Stroke J. 4, 6–12. 10.1177/2396987319832140 PubMed DOI PMC

Wang L., Lee C.-Y., Tu Z., Lazebnik S. (2015a). Training deeper convolutional networks with deep supervision. arXiv 1505.02496.

Wang R., Li C., Wang J., Wei X., Li Y., Zhu Y., et al. . (2015b). Threshold segmentation algorithm for automatic extraction of cerebral vessels from brain magnetic resonance angiography images. J. Neurosci. Methods 241, 30–36. 10.1016/j.jneumeth.2014.12.003 PubMed DOI

WHO EMRO (2019). Stroke, Cerebrovascular Accident|Health Topics. Available online at: http://www.emro.who.int/health-topics/stroke-cerebrovascular-accident/index.html (accessed September 20, 2019).

Yoo J., Baek J.-H., Park H., Song D., Kim K., Hwang I. G., et al. . (2018). Thrombus volume as a predictor of nonrecanalization after intravenous thrombolysis in acute stroke. Stroke 49, 2108–2115. 10.1161/STROKEAHA.118.021864 PubMed DOI

Yue K., Zou B., Chen Z., Liu Q. (2019). Retinal vessel segmentation using dense U-net with multiscale inputs. J. Med. Imaging 6:034004. 10.1117/1.JMI.6.3.034004 PubMed DOI PMC

Yushkevich P. A., Piven J., Hazlett H. C., Smith R. G., Ho S., Gee J. C., et al. . (2006). User-guided 3D active contour segmentation of anatomical structures: Significantly improved efficiency and reliability. Neuroimage 31, 1116–1128. 10.1016/j.neuroimage.2006.01.015 PubMed DOI

Zhang J., Jin Y., Xu J., Xu X., Zhang Y. (2018). MDU-Net: multi-scale densely connected U-Net for biomedical image segmentation. arXiv 1812.00352. PubMed PMC

Zhou Z., Siddiquee M. M. R., Tajbakhsh N., Liang J. (2020). UNet++: redesigning skip connections to exploit multiscale features in image segmentation. arXiv 1912.05074. PubMed PMC

BRAVE-NET: Fully Automated Arterial Brain Vessel Segmentation in Patients With Cerebrovascular Disease