Advertisement

Fracture Detection Using Bayesian Inference

  • Ananda S. Chowdhury
  • Suchendra M. Bhandarkar
Part of the Advances in Computer Vision and Pattern Recognition book series (ACVPR)

Abstract

Introducing automation in the various aspects of reconstructive craniofacial surgery is a highly demanding and technically challenging task. The automatic detection of fracture surfaces, which subsequently serve as input data to the surface matching and virtual reconstruction algorithms, is a critical and integral aspect of this endeavor. The generation of fracture surface data thus far has entailed significant user involvement. The surgeon has had to explicitly identify and manually extract the stable fracture points in the CT image sequence of the fractured craniofacial skeleton. This has proved to be a performance bottleneck in attaining the ultimate goal of in silico surgery with minimal user intervention. Semi-automatic detection reduces considerably the time and cost of the operation. More importantly, reduced time in the operating room results in reduced operative trauma to the patient and reduced risk for potential postoperative complications. In this chapter, we propose a scheme for identification of stable fracture points in well displaced or major fractures in an input CT image sequence. First, a set of candidate fracture points is generated by identifying points of high curvature on the contours of the broken mandibular fragments in the individual 2D CT image slices. Next, the conventional Kalman filter is modeled as a Bayesian inference problem and used for testing the spatial consistency and stability of the candidate fracture points across the CT image sequence of interest. The proposed scheme is shown to be very effective in detecting fracture surfaces in well displaced fractures.

Keywords

Compute Tomography Image Kalman Filter Corner Point Mandibular Fracture Fracture Point 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

References

  1. 54.
    Bhandarkar SM, Chowdhury AS, Tang Y, Yu JC, Tollner EW (2007) Computer vision guided virtual craniofacial reconstruction. Comput Med Imaging Graph 31(6):418–427 CrossRefGoogle Scholar
  2. 107.
    Chowdhury AS, Bhandarkar SM, Robinson RW, Yu JC (2009) Virtual multi-fracture craniofacial reconstruction using computer vision and graph matching. Comput Med Imaging Graph 33(5):333–342 CrossRefGoogle Scholar
  3. 115.
    Ozanian T, Phillips R (2000) Image analysis for computer-assisted surgery of hip fractures. Med Image Anal 4(2):137–159 CrossRefGoogle Scholar
  4. 116.
    Yap DWH, Chen Y, Leow WK, Howe TS, Png MA (2004) Detecting femur fractures by texture analysis of trabeculae. In: Proc int conf on pattern recognition (ICPR), Cambridge, UK, pp 730–733 Google Scholar
  5. 117.
    Syiam M, El-Aziem MA, El-Mensshawy M (2004) AdAgen: Adaptive interface agent for X-ray fracture detection. In: Proc IEEE int conf on electrical, electronic and computer engineering (ICEEC), Cairo, Egypt, pp 354–357 Google Scholar
  6. 118.
    Lum V, Leow W, Chen Y, Howe Y, Png M (2005) Combining classifiers for bone fracture detection in X-ray images. In: Proc int conf on image processing (ICIP), Genoa, Italy, pp 1149–1152 Google Scholar
  7. 119.
    Jia Y, Jiang Y (2006) Active contour model with shape constraints for bone fracture detection. In: Proc European conf on colour in graphics, imaging, and vision (CGIV), Leeds, UK, pp 90–95 Google Scholar
  8. 120.
    He JC, Leow WK, Howe TS (2007) Hierarchical classifiers for detection of fractures in X-ray images. In: Kropatsch WG, Kampel M, Hanbury A (eds) Proc int conf on computer analysis of images and patterns (CAIP), Vienna, Austria. LNCS, vol 4673. Springer, Berlin, pp 962–969 CrossRefGoogle Scholar
  9. 121.
    Donnelley M, Knowles G, Hearn T (2008) A CAD system for long-bone segmentation and fracture detection. In: Elmoataz A et al. (eds) Proc IAPR int conf on image and signal processing (ICISP), Cherbourg-Octeville, Normandy, France. LNCS, vol 5099. Springer, Berlin, pp 153–162 Google Scholar
  10. 122.
    Welch G, Bishop G (2004) An introduction to the Kalman filter. Tech Report, No 95-041, UNC Chapel Hill Google Scholar
  11. 123.
    Meinhold RJ, Singpurwalla ND (1983) Understanding the Kalman filter. Am Stat 37(2):123–127 CrossRefMathSciNetGoogle Scholar
  12. 124.
    Chowdhury AS, Bhandarkar SM, Datta GS, Yu JC (2006) Automated detection of stable fracture points. In computed tomography image sequences. In: Proc IEEE int symp on biomedical imaging (ISBI), Arlington, VA, USA, pp 1320–1323 Google Scholar
  13. 125.
    Jain R, Kasturi R, Schunck BG (1995) Machine vision. McGraw Hill, New York Google Scholar
  14. 126.
    Pena D, Gutman I (1988) Bayesian approach to robustifying the Kalman filter. In: Spall J (ed) Bayesian analysis of time series and dynamic models. Marcel Dekker, New York, pp 227–253 Google Scholar
  15. 127.
    Ye JC, Bresler Y, Moulin P (2000) Asymptotic global confidence regions in parametric shape estimation problems. IEEE Trans Inf Theory 46(5):1881–1895 CrossRefMATHMathSciNetGoogle Scholar
  16. 128.
    Subramanian KR, Brockway JP, Carruthers WB (2004) Interactive detection and visualization of breast lesions from dynamic contrast enhanced MRI volumes. Comput Med Imaging Graph 28(8):435–444 CrossRefGoogle Scholar
  17. 129.
    Simonson KM, Drescher SM Jr, Tanner FR (2007) A statistics-based approach to binary image registration with uncertainty analysis. IEEE Trans Pattern Anal Mach Intell 29(1):112–125 CrossRefGoogle Scholar
  18. 130.
    Gelman A, Carlin JB, Stern HS, Rubin DB (2004) Bayesian data analysis. Chapman & Hall/CRC, Boca Raton MATHGoogle Scholar
  19. 131.
    Neter J, Kutner M, Wasserman W, Nachsteim C (2002) Applied linear statistical models. McGraw Hill, New York Google Scholar
  20. 132.
    Perez P, Vermaak J, Blake A (2004) Data fusion for visual tracking with particles. Proc IEEE 92(3):495–513 CrossRefGoogle Scholar

Copyright information

© Springer-Verlag London Limited 2011

Authors and Affiliations

  • Ananda S. Chowdhury
    • 1
  • Suchendra M. Bhandarkar
    • 2
  1. 1.Department of Electronics & Telecommunication EngineeringJadavpur UniversityKolkataIndia
  2. 2.Department of Computer ScienceThe University of GeorgiaAthensUSA

Personalised recommendations