GPU accelerated segmentation and centerline extraction of tubular structures from medical images

Purpose   To create a fast and generic method with sufficient quality for extracting tubular structures such as blood vessels and airways from different modalities (CT, MR and US) and organs (brain, lungs and liver) by utilizing the computational power of graphic processing units (GPUs).Methods   A cropping algorithm is used to remove unnecessary data from the datasets on the GPU. A model-based tube detection filter combined with a new parallel centerline extraction algorithm and a parallelized region growing segmentation algorithm is used to extract the tubular structures completely on the GPU. Accuracy of the proposed GPU method and centerline algorithm is compared with the ridge traversal and skeletonization/thinning methods using synthetic vascular datasets.Results   The implementation is tested on several datasets from three different modalities: airways from CT, blood vessels from MR, and 3D Doppler Ultrasound. The results show that the method is able to extract airways and vessels in 3–5 s on a modern GPU and is less sensitive to noise than other centerline extraction methods.Conclusions   Tubular structures such as blood vessels and airways can be extracted from various organs imaged by different modalities in a matter of seconds, even for large datasets.

[1]  Horst Bischof,et al.  Airway Tree Reconstruction Based on Tube Detection , 2009, MICCAI 2009.

[2]  Francis K. H. Quek,et al.  A review of vessel extraction techniques and algorithms , 2004, CSUR.

[3]  Barbara Cutler,et al.  Robust Adaptive 3-D Segmentation of Vessel Laminae From Fluorescence Confocal Microscope Images and Parallel GPU Implementation , 2010, IEEE Transactions on Medical Imaging.

[4]  Eric A. Hoffman,et al.  Extraction of Airways From CT (EXACT'09) , 2012, IEEE Transactions on Medical Imaging.

[5]  D. Louis Collins,et al.  Clinical validation of vessel-based registration for correction of brain-shift , 2007, Medical Image Anal..

[6]  Horst Bischof,et al.  Segmentation of Airways Based on Gradient Vector Flow , 2009, MICCAI 2009.

[7]  Frank Lindseth,et al.  Real-time gradient vector flow on GPUs using OpenCL , 2015, Journal of Real-Time Image Processing.

[8]  Bram van Ginneken,et al.  Computer analysis of computed tomography scans of the lung: a survey , 2006, IEEE Transactions on Medical Imaging.

[9]  Rangasami L. Kashyap,et al.  Building Skeleton Models via 3-D Medial Surface/Axis Thinning Algorithms , 1994, CVGIP Graph. Model. Image Process..

[10]  Horst Bischof,et al.  A Novel Approach for Detection of Tubular Objects and Its Application to Medical Image Analysis , 2008, DAGM-Symposium.

[11]  Lin Shi,et al.  A survey of GPU-based medical image computing techniques. , 2012, Quantitative imaging in medicine and surgery.

[12]  L. Cohen,et al.  Segmentation of 3D tubular objects with adaptive front propagation and minimal tree extraction for 3D medical imaging , 2007, Computer methods in biomechanics and biomedical engineering.

[13]  Baba C. Vemuri,et al.  Shape Modeling with Front Propagation: A Level Set Approach , 1995, IEEE Trans. Pattern Anal. Mach. Intell..

[14]  Karl Rohr,et al.  Robust segmentation of tubular structures in 3-D medical images by parametric object detection and tracking , 2003, IEEE Trans. Syst. Man Cybern. Part B.

[15]  Isabelle Bloch,et al.  A review of 3D vessel lumen segmentation techniques: Models, features and extraction schemes , 2009, Medical Image Anal..

[16]  Demetri Terzopoulos,et al.  Snakes: Active contour models , 2004, International Journal of Computer Vision.

[17]  Christian Bauer,et al.  Segmentation of 3D Tubular Tree Structures in Medical Images , 2010 .

[18]  Marius Erdt,et al.  Automatic Hepatic Vessel Segmentation Using Graphics Hardware , 2008, MIAR.

[19]  Ulf Assarsson,et al.  Efficient stream compaction on wide SIMD many-core architectures , 2009, High Performance Graphics.

[20]  H. Bischof,et al.  Edge Based Tube Detection for Coronary Artery Centerline Extraction , 2008, The MIDAS Journal.

[21]  W. Brent Seales,et al.  Endoscopic Video Texture Mapping on Pre-Built 3-D Anatomical Objects Without Camera Tracking , 2010, IEEE Transactions on Medical Imaging.

[22]  Marleen de Bruijne,et al.  Vessel-guided airway tree segmentation: A voxel classification approach , 2010, Medical Image Anal..

[23]  Frank Lindseth,et al.  GPU-Based Airway Segmentation and Centerline Extraction for Image Guided Bronchoscopy , 2012 .

[24]  Kenneth A. Hawick,et al.  Parallel graph component labelling with GPUs and CUDA , 2010, Parallel Comput..

[25]  Horst Bischof,et al.  Extracting Curve Skeletons from Gray Value Images for Virtual Endoscopy , 2008, MIAR.

[26]  Laurent D. Cohen,et al.  Tubular Structure Segmentation Based on Minimal Path Method and Anisotropic Enhancement , 2011, International Journal of Computer Vision.

[27]  Kaleem Siddiqi,et al.  Flux Maximizing Geometric Flows , 2001, ICCV.

[28]  Stephen R. Aylward,et al.  Initialization, noise, singularities, and scale in height ridge traversal for tubular object centerline extraction , 2002, IEEE Transactions on Medical Imaging.

[29]  Falko Kuester,et al.  GPU-Based Active Contour Segmentation Using Gradient Vector Flow , 2006, ISVC.

[30]  Zuoyong Zheng,et al.  A Fast GVF Snake Algorithm on the GPU , 2013 .

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

[32]  William E. Higgins,et al.  Robust 3-D Airway Tree Segmentation for Image-Guided Peripheral Bronchoscopy , 2010, IEEE Transactions on Medical Imaging.

[33]  Ghassan Hamarneh,et al.  VascuSynth: Simulating vascular trees for generating volumetric image data with ground-truth segmentation and tree analysis , 2010, Comput. Medical Imaging Graph..

[34]  Ole Christian Eidheim,et al.  Real-time analysis of ultrasound images using GPU , 2005 .

[35]  Bram van Ginneken,et al.  Robust Segmentation and Anatomical Labeling of the Airway Tree from Thoracic CT Scans , 2008, MICCAI.

[36]  Hans-Peter Seidel,et al.  On-the-fly Point Clouds through Histogram Pyramids , 2006 .

[37]  Frank Lindseth,et al.  Real-Time Surface Extraction and Visualization of Medical Images using OpenCL and GPUs , 2012 .

[38]  Gábor Székely,et al.  Fast and robust extraction of centerlines in 3D tubular structures using a scattered-snakelet approach , 2006, SPIE Medical Imaging.

[39]  V.R.S Mani,et al.  Survey of Medical Image Registration , 2013 .

[40]  PhengAnn Heng,et al.  Automated extraction of bronchus from 3D CT images of lung based on genetic algorithm and 3D region growing , 2000, Medical Imaging: Image Processing.

[41]  Anthony J. Yezzi,et al.  Vessels as 4-D Curves: Global Minimal 4-D Paths to Extract 3-D Tubular Surfaces and Centerlines , 2007, IEEE Transactions on Medical Imaging.

[42]  Horst Bischof,et al.  Pulmonary Vascular Tree Segmentation from Contrast-Enhanced CT Images , 2013, ArXiv.

[43]  Jerry L. Prince,et al.  Snakes, shapes, and gradient vector flow , 1998, IEEE Trans. Image Process..

[44]  Aly A. Farag,et al.  On the Extraction of Curve Skeletons using Gradient Vector Flow , 2007, 2007 IEEE 11th International Conference on Computer Vision.

[45]  Olivier D. Faugeras,et al.  Codimension-two geodesic active contours for the segmentation of tubular structures , 2000, Proceedings IEEE Conference on Computer Vision and Pattern Recognition. CVPR 2000 (Cat. No.PR00662).

[46]  Li Yuan-yuan.  A SURVEY OF MEDICAL IMAGE REGISTRATION , 2006 .

[47]  Ghassan Hamarneh,et al.  VascuSynth: Vascular Tree Synthesis Software , 2011 .

[48]  Nicholas Ayache,et al.  Model-Based Detection of Tubular Structures in 3D Images , 2000, Comput. Vis. Image Underst..