JointVesselNet: Joint Volume-Projection Convolutional Embedding Networks for 3D Cerebrovascular Segmentation

In this paper, we present an end-to-end deep learning method, JointVesselNet, for robust extraction of 3D sparse vascular structure through embedding the image composition, generated by maximum intensity projection (MIP), into the 3D magnetic resonance angiography (MRA) volumetric image learning process to enhance the overall performance. The MIP embedding features can strengthen the local vessel signal and adapt to the geometric variability and scalability of vessels. Therefore, the proposed framework can better capture the small vessels and improve the vessel connectivity. To our knowledge, this is the first time that a deep learning framework is proposed to construct a joint convolutional embedding space, where the computed joint vessel probabilities from 2D projection and 3D volume can be integrated synergistically. Experimental results are evaluated and compared with the traditional 3D vessel segmentation methods and the state-of-the-art in deep learning, by using both public and real patient cerebrovascular image datasets.

[1]  Anthony J. Yezzi,et al.  Vessel Segmentation Using a Shape Driven Flow , 2004, MICCAI.

[2]  Nils Daniel Forkert,et al.  3D cerebrovascular segmentation combining fuzzy vessel enhancement and level-sets with anisotropic energy weights. , 2013, Magnetic resonance imaging.

[3]  Anil A. Bharath,et al.  Retinal Blood Vessel Segmentation by Means of Scale-Space Analysis and Region Growing , 1999, MICCAI.

[4]  Bjoern H. Menze,et al.  DeepVesselNet: Vessel Segmentation, Centerline Prediction, and Bifurcation Detection in 3-D Angiographic Volumes , 2018, Frontiers in Neuroscience.

[5]  Il Dong Yun,et al.  Deep Vessel Segmentation By Learning Graphical Connectivity , 2018, Medical Image Anal..

[6]  Alejandro F. Frangi,et al.  Muliscale Vessel Enhancement Filtering , 1998, MICCAI.

[7]  Nikos Paragios,et al.  Globally Optimal Active Contours, Sequential Monte Carlo and On-Line Learning for Vessel Segmentation , 2006, ECCV.

[8]  Lei Zhang,et al.  Multi-level deep supervised networks for retinal vessel segmentation , 2017, International Journal of Computer Assisted Radiology and Surgery.

[9]  Bojana Stefanovic,et al.  Amyloid-β-dependent compromise of microvascular structure and function in a model of Alzheimer's disease. , 2012, Brain : a journal of neurology.

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

[11]  G D Rubin,et al.  CT angiography with spiral CT and maximum intensity projection. , 1992, Radiology.

[12]  William R. Brown,et al.  Cerebral microvascular pathology in aging and neurodegeneration , 2011 .

[13]  Benoît Naegel,et al.  Cerebrovascular Network Segmentation of MRA Images With Deep Learning , 2018, 2019 IEEE 16th International Symposium on Biomedical Imaging (ISBI 2019).

[14]  Maxime Descoteaux,et al.  A geometric flow for segmenting vasculature in proton-density weighted MRI , 2008, Medical Image Anal..

[15]  Karl Rohr,et al.  Globally Optimal Curvature-Regularized Fast Marching for Vessel Segmentation , 2013, MICCAI.

[16]  Wei Xiong,et al.  VesselNet: A deep convolutional neural network with multi pathways for robust hepatic vessel segmentation , 2019, Comput. Medical Imaging Graph..

[17]  Sergey Ioffe,et al.  Inception-v4, Inception-ResNet and the Impact of Residual Connections on Learning , 2016, AAAI.

[18]  W. M. van der Flier,et al.  Heterogeneity of small vessel disease: a systematic review of MRI and histopathology correlations , 2010, Journal of Neurology, Neurosurgery & Psychiatry.

[19]  Thomas Brox,et al.  3D U-Net: Learning Dense Volumetric Segmentation from Sparse Annotation , 2016, MICCAI.

[20]  W. Brown,et al.  Review: Cerebral microvascular pathology in ageing and neurodegeneration , 2011, Neuropathology and applied neurobiology.

[21]  Stephen Lin,et al.  DeepVessel: Retinal Vessel Segmentation via Deep Learning and Conditional Random Field , 2016, MICCAI.

[22]  Tianfu Wang,et al.  A Cross-Modality Learning Approach for Vessel Segmentation in Retinal Images , 2016, IEEE Transactions on Medical Imaging.

[23]  Meghan Mott,et al.  Small blood vessels: big health problems: National Institute of Neurological Disorders and Stroke update. , 2014, Stroke.

[24]  Ronald M. Summers,et al.  Sequential Monte Carlo Tracking for Marginal Artery Segmentation on CT Angiography by Multiple Cue Fusion , 2013, MICCAI.

[25]  Yongquan Ye,et al.  Noncontrast‐enhanced magnetic resonance angiography and venography imaging with enhanced angiography , 2013, Journal of magnetic resonance imaging : JMRI.

[26]  Krzysztof Krawiec,et al.  Segmenting Retinal Blood Vessels With Deep Neural Networks , 2016, IEEE Transactions on Medical Imaging.

[27]  J. Alison Noble,et al.  Statistical 3D Vessel Segmentation Using a Rician Distribution , 1999, MICCAI.