Advertisement

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

  • Erik Smistad
  • Anne C. Elster
  • Frank Lindseth
Original Article

Abstract

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.

Keywords

Segmentation Centerline extraction Vessel Airway GPU Parallel 

Notes

Acknowledgments

Thanks to the people of the Heterogeneous and Parallel Computing Lab at NTNU for all their assistance and St. Olav’s University Hospital for the datasets. The authors would also like to convey thanks to NTNU and NVIDIA’s CUDA Research Center Program for their hardware contributions to the HPC Lab. Without their continued support, this project would not have been possible.

Conflict of interest

Erik Smistad, Anne C. Elster and Frank Lindseth declare that they have no conflict of interest.

References

  1. 1.
    AMD. AMD Accelerated Parallel Processing OpenCL Programming Guide. Technical Report December, 2012. http://developer.amd.com/download/AMD_Accelerated_Parallel_Processing_OpenCL_Programming_Guide.pdf. Accessed 4th July 2013
  2. 2.
    Aylward SR, Bullitt E (2002) Initialization, noise, singularities, and scale in height ridge traversal for tubular object centerline extraction. IEEE Trans Med Imaging 21(2):61–75PubMedCrossRefGoogle Scholar
  3. 3.
    Bauer C (2010) Segmentation of 3D tubular tree structures in medical images. PhD thesis, Graz University of TechnologyGoogle Scholar
  4. 4.
    Bauer C, Bischof V (2008) A novel approach for detection of tubular objects and its application to medical image analysis. In: Proceedings of the 30th DAGM symposium on pattern recognition. Springer, pp 163–172Google Scholar
  5. 5.
    Bauer C, Bischof H (2008) Edge based tube detection for coronary artery centerline extraction. MIDAS J. http://www.midasjournal.org/browse/publication/577
  6. 6.
    Bauer C, Bischof H (2008) Extracting curve skeletons from gray value images for virtual endoscopy. In: Proceedings of the 4th international workshop on medical imaging and augmented reality. Springer, pp 393–402Google Scholar
  7. 7.
    Bauer C, Bischof H, Beichel R (2009) Segmentation of airways based on gradient vector flow. In: Proceedings of the 2nd international workshop on pulmonary image analysis. MICCAI, Citeseer, pp 191–201Google Scholar
  8. 8.
    Bauer C, Pock T, Bischof H, Beichel R (2009) Airway tree reconstruction based on tube detection. In: Proceedings of the 2nd international workshop on pulmonary image analysis. MICCAI, Citeseer, pp 203–214Google Scholar
  9. 9.
    Behrens T, Rohr K, Stiehl HS (2003) Robust segmentation of tubular structures in 3-D medical images by parametric object detection and tracking. IEEE Trans Syst Man Cybern Part B Cybern Publ IEEE Syst Man Cybern Soc 33(4):554–61Google Scholar
  10. 10.
    Benmansour F, Cohen LD (2010) Tubular structure segmentation based on minimal path method and anisotropic enhancement. Int J Comput Vis 92(2):192–210CrossRefGoogle Scholar
  11. 11.
    Billeter M, Olsson O, Assarsson U (2009) Efficient stream compaction on wide SIMD many-core architectures. In: Proceedings of the conference on high performance graphics, pp 159–166Google Scholar
  12. 12.
    Cohen LD, Deschamps T (2007) Segmentation of 3D tubular objects with adaptive front propagation and minimal tree extraction for 3D medical imaging. Comput Method Biomech Biomed Eng 10(4):289–305CrossRefGoogle Scholar
  13. 13.
    Eidheim O, Skjermo J, Aurdal L (2005) Real-time analysis of ultrasound images using GPU. Int Congr Ser 1281:284–289CrossRefGoogle Scholar
  14. 14.
    Erdt M, Raspe M, Suehling M (2008) Automatic hepatic vessel segmentation using graphics hardware. In: Proceedings of the 4th international workshop on medical imaging and augmented reality, pp 403–412Google Scholar
  15. 15.
    Frangi A, Niessen W, Vincken K, Viergever M (1998) Multiscale vessel enhancement filtering. Med Image Comput Comput Assist Interv 1496:130–137Google Scholar
  16. 16.
    Graham MW, Gibbs JD, Cornish DC (2010) Robust 3-D airway tree segmentation for image-guided peripheral bronchoscopy. IEEE Trans Med Imaging 29(4):982–997PubMedCrossRefGoogle Scholar
  17. 17.
    Hamarneh G, Jassi P (2010) VascuSynth: simulating vascular trees for generating volumetric image data with ground-truth segmentation and tree analysis. Comput Med Imaging Graph 34(8):605–616PubMedCrossRefGoogle Scholar
  18. 18.
    Hassouna M., Farag A. (2007) On the extraction of curve skeletons using gradient vector flow. In: IEEE 11th international conference on computer vision. IEEE, pp 1–8Google Scholar
  19. 19.
    Hawick K, Leist A, Playne D (2010) Parallel graph component labelling with GPUs and CUDA. Parallel Comput 36(12):655–678CrossRefGoogle Scholar
  20. 20.
    He Z, Kuester F (2006) GPU-based active contour segmentation using gradient vector flow. In: Advances in visual, computing, pp 191–201Google Scholar
  21. 21.
    Helmberger M, Urschler M, Pienn M, Bálint Z, Olschewski A, Bischof H (2013) Pulmonary vascular tree segmentation from contrast-enhanced CT images. In: Proceedings of the 37th annual workshop of the austrian association for, pattern recognition, pp 1–10Google Scholar
  22. 22.
    Homann H (2007) Implementation of a 3D thinning algorithm. Insight J. http://www.insight-journal.org/browse/publication/181
  23. 23.
    Jassi P, Hamarneh G (2011) VascuSynth: vascular tree synthesis software. Insight J. http://www.insight-journal.org/browse/publication/794
  24. 24.
    Kass M, Witkin A, Terzopoulos D (1988) Snakes: active contour models. Int J Comput Vis 1(4):321–331CrossRefGoogle Scholar
  25. 25.
    Kirbas C, Quek F (2004) A review of vessel extraction techniques and algorithms. ACM Comput Surv 36(2):81–121CrossRefGoogle Scholar
  26. 26.
    Krissian K, Malandain G, Ayache N (2000) Model-based detection of tubular structures in 3D images. Comput Vis Image Underst 80(2):130–171CrossRefGoogle Scholar
  27. 27.
    Law T-Y, Heng PA (2000) Automated extraction of bronchus from 3D CT images of lung based on genetic algorithm and 3D region growing. Proc SPIE 3979:906–916CrossRefGoogle Scholar
  28. 28.
    Lee T, Kashyap R, Chu C (1994) Building skeleton models via 3-D medial surface/axis thinning algorithms. CVGIP Graph Model Image Processing 56(6):462–478CrossRefGoogle Scholar
  29. 29.
    Lesage D, Angelini ED, Bloch I, Funka-Lea G (2009) A review of 3D vessel lumen segmentation techniques: models, features and extraction schemes. Med Image Anal 13(6):819–845PubMedCrossRefGoogle Scholar
  30. 30.
    Li H, Yezzi A (2007) Vessels as 4-D curves: global minimal 4-D paths to extract 3-D tubular surfaces and centerlines. IEEE Trans Med Imaging 29(9):1213–1223CrossRefGoogle Scholar
  31. 31.
    Lo P, Ginneken BV, Reinhardt JM, de Bruijne M (2009) Extraction of airways from CT (EXACT’09) . In: Second international workshop on pulmonary image, analysis, pp 175–189Google Scholar
  32. 32.
    Lo P, Sporring J, Ashraf H, Pedersen JJH, de Bruijne M (2010) Vessel-guided airway tree segmentation: a voxel classification approach. Med Image Anal 14(4):527–538PubMedCrossRefGoogle Scholar
  33. 33.
    Lorigo L, Faugeras O (2000) Codimension-two geodesic active contours for the segmentation of tubular structures. Comput Vis Pattern Recognit, 444–451 Google Scholar
  34. 34.
    Maintz JBA, Viergever MA (1998) A survey of medical image registration. Med Image Anal 2(1):1–36Google Scholar
  35. 35.
    Malladi R, Sethian J, Vemuri B (1995) Shape modeling with front propagation: a level set approach. IEEE Trans Pattern Anal Machine Intell 17(2):158–175CrossRefGoogle Scholar
  36. 36.
    Narayanaswamy A, Dwarakapuram S, Bjornsson CS, Cutler BM, Shain W, Roysam B (2010) Robust adaptive 3-D segmentation of vessel laminae from fluorescence confocal microscope images and parallel GPU implementation. IEEE Trans Med Imaging 29(3):583–597PubMedCentralPubMedCrossRefGoogle Scholar
  37. 37.
    NVIDIA. OpenCL Best Practices Guide. Technical report, 2010. http://www.nvidia.com/content/cudazone/CUDABrowser/downloads/papers/NVIDIA_OpenCL_BestPracticesGuide.pdf Accessed 4. July 2013
  38. 38.
    Reinertsen I, Lindseth F, Unsgaard G, Collins DL (2007) Clinical validation of vessel-based registration for correction of brain-shift. Med Image Anal 11(6):673–684PubMedCrossRefGoogle Scholar
  39. 39.
    Shi L, Liu W, Zhang H, Xie Y, Wang D (2012) A survey of GPU-based medical image computing techniques. Quant Imaging Med Surg 2(3):188–206PubMedCentralPubMedGoogle Scholar
  40. 40.
    Sluimer I, Schilham A, Prokop M, van Ginneken B (2006) Computer analysis of computed tomography scans of the lung: a survey. IEEE Trans Med Imaging 25(4):385–405PubMedCrossRefGoogle Scholar
  41. 41.
    Smistad E, Elster AC, Lindseth F (2012) GPU-based airway segmentation and centerline extraction for image guided bronchoscopy. In Norsk informatikkonferanse. Akademika forlag, pp 129–140Google Scholar
  42. 42.
    Smistad E, Elster AC, Lindseth F (2012) Real-time gradient vector flow on GPUs using OpenCL. J Real-Time Image Processing. http://link.springer.com/article/10.1007%2Fs11554-012-0257-6
  43. 43.
    Smistad E, Elster AC, Lindseth F (2012) Real-time surface extraction and visualization of medical images using OpenCL and GPUs. In: Norsk informatikkonferanse. Akademika forlag, pp 141–152Google Scholar
  44. 44.
    Spuhler C, Harders M, Székely G (2006) Fast and robust extraction of centerlines in 3D tubular structures using a scattered–snakelet approach. Proc SPIE 6144, March 2006Google Scholar
  45. 45.
    van Ginneken B, Baggerman W, van Rikxoort EM (2008) Robust segmentation and anatomical labeling of the airway tree from thoracic CT scans. Int Conf Med Image Comput Comput Assist Interv 11:219–226Google Scholar
  46. 46.
    Vasilevskiy A, Siddiqi K (2002) Flux maximizing geometric flows. IEEE Trans Pattern Anal Mach Intell 24(12):1565–1578CrossRefGoogle Scholar
  47. 47.
    Xu C, Prince J (1998) Snakes, shapes, and gradient vector flow. IEEE Trans Image Processing 7(3):359–369CrossRefGoogle Scholar
  48. 48.
    Zheng Z, Zhang R (2012) A fast GVF snake algorithm on the GPU. Res J Appl Sci Eng Technol 4(24):5565–5571Google Scholar
  49. 49.
    Ziegler G, Tevs A, Theobalt C, Seidel H (2006) On-the-fly point clouds through histogram pyramids. In Vision, modeling, and visualization 2006: proceedings, Nov 22–24, 2006. IOS Press, Aachen, Germany, pp 137Google Scholar

Copyright information

© CARS 2013

Authors and Affiliations

  • Erik Smistad
    • 1
  • Anne C. Elster
    • 1
  • Frank Lindseth
    • 1
    • 2
  1. 1.Department of Computer and Information ScienceNorwegian University of Science and TechnologyTrondheimNorway
  2. 2.SINTEF Medical TechnologyTrondheimNorway

Personalised recommendations