e & i Elektrotechnik und Informationstechnik

, Volume 124, Issue 7–8, pp 248–253

Bayesian inversion method for 3D dental X-ray imaging

  • V. Kolehmainen
  • A. Vanne
  • S. Siltanen
  • S. Järvenpää
  • J. P. Kaipio
  • M. Lassas
  • M. Kalke
Originalarbeit

Summary

Diagnostic and operational tasks in dentistry require three-dimensional (3D) information about tissue. A novel type of low dose dental 3D X-ray imaging is considered. Given projection images taken from a few sparsely distributed directions using the dentist's regular X-ray equipment, the 3D X-ray attenuation function is reconstructed. This is an ill-posed inverse problem, and Bayesian inversion is a well suited framework for reconstruction from such incomplete data. The reconstruction problem is formulated in a well-posed probabilistic form in which a priori information is used to compensate for the incomplete data. A parallelized Bayesian method (implemented for a Beowulf cluster computer) for 3D reconstruction in dental radiology is presented (the method was originally presented in (Kolehmainen et al., 2006)). The prior model for dental structures consists of a weighted l1 and total variation (TV)-prior together with the positivity prior. The inverse problem is stated as finding the maximum a posterior (MAP) estimate. The method is tested with in vivo patient data and shown to outperform the reference method (tomosynthesis).

Keywords

Inverse problems X-ray tomography Dental Beowulf Parallel computing 

Bayes'sche Inversion für dreidimensionale Röntgentomografie in der Zahnmedizin

Zusammenfassung

Diagnostische und operative Zahnmedizin erfordert dreidimensionale (3D) Gewebeinformation. In diesem Beitrag wird eine neue Art der schwachdosierten 3D-Röntgentomografie untersucht. Die Rekonstruktion der 3D-Dämpfungsverteilung erfolgt aus nur wenigen Projektionen, die mit handelsüblichen Röntgenapparaten aufgenommen werden. Das inverse Problem ist als Bayes'sches Schätzproblem formuliert, in dem a priori Information zur Kompensation der unvollständigen Messdaten berücksichtigt wird. Zur Lösung des inversen Problems wird eine parallelisierte Bayes'sche Methode für 3D-Röntgentomografie (implementiert für einen Beowulf-Rechencluster) vorgeschlagen. Diese Art der Rekonstruktion wurde in (Kolehmainen et al., 2006) präsentiert. Das verwendete A priori-Modell besteht aus einer Kombination einer gewichteten (l1-prior, einer total variation (TV) prior und einer Positivitätsprior. Das inverse Problem kann als Suche nach dem Maximum a posterior (MAP)-Zustand betrachtet werden. Die vorgeschlagene Methode wird anhand von In vivo-Patientendaten getestet und zeigt eine signifikante Verbesserung gegenüber der üblichen Methode (tomosynthesis).

Schlüsselwörter

Inverses Problem Röntgentomografie Zahnmedizin Beowulf Parallelverarbeitung 

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. Balay, S., Buchelman, K., Eijkhout, V., Gropp, W. D., Kaushik, D., Knepley, M. G., McInnes, L. C., Smith, B. E., Zhang, H. (1995): PETSc Users Manual. Argonne National Laboratory ANL-95/11, revision 2.1.5 edition.Google Scholar
  2. Barzilai, J., Borwein, J. M. (1988): Two point step size gradient method. IMA J. Numer. Anal. 8: 141–148.MATHCrossRefMathSciNetGoogle Scholar
  3. Bouman, C., Sauer, K. (1993): A generalized Gaussian image model for edge-preserving MAP estimation. IEEE Trans. Image Processing 2: 296–310.CrossRefGoogle Scholar
  4. Brocklebank, L. (1997): Dental Radiology – Understanding the X-ray Image: Oxford University Press. ISBN 0-19-262411-3.Google Scholar
  5. Dobson, D. C., Santosa, F. (1994): An image enhancement technique for electrical impedance tomography. Inv. Probl. 10: 317–334.MATHCrossRefMathSciNetGoogle Scholar
  6. Dobson, D. C., Santosa, F. (1996): Recovery of blocky images from noisy and blurred data. SIAM J. Appl. Math. 56: 1181–1198.MATHCrossRefMathSciNetGoogle Scholar
  7. Donoho, D. L., Johnstone, I. M., Hoch, J. C., Stern, A. S. (1992): Maximum entropy and the nearly black object. J. Roy. Statist. Ser. B 54: 41–81.MATHMathSciNetGoogle Scholar
  8. Ekestubbe, A., Gröndahl, K., Gröndahl, H.-G. (1997): The use of tomography for dental implant planning. Dentomaxillofacial Radiology 26: 206–213.CrossRefGoogle Scholar
  9. Fiacco, A. V., McCormick, G. P. (1990): Nonlinear programming: sequential unconstrained minimization techniques. SIAM.Google Scholar
  10. Frese, T., Bouman, C., Sauer, K. (2002): Adaptive wavelet graph model for Bayesian tomographic reconstruction. IEEE Trans. Image Processing 11: 756–770.CrossRefGoogle Scholar
  11. Gropp, W., Lusk, E., Doss, N., Skjellum, A. (1996): A high-performance, portable implementation of the MPI message passing interface standard. Parallel Computing 22: 789–828.MATHCrossRefGoogle Scholar
  12. Hanson, K. M. (1987): Bayesian and related methods in image reconstruction from incomplete data. Image Recovery: Theory and Applications. Academic, Orlando.Google Scholar
  13. Hanson, K. M., Cunningham, G. S., McKee, R. (1997): Uncertainty assessment for reconstructions based on deformable geometry. Int. J. Imaging Syst. Technol. 8: 506–512.CrossRefGoogle Scholar
  14. Hanson, K. M., Wecksung, G. W. (1983): Bayesian approach to limited-angle reconstruction in computed tomography. J. Opt. Soc. Am. 73: 1501–1509.CrossRefGoogle Scholar
  15. Kaipio, J. P., Somersalo, E. (2004): Statistical and Computational Methods for Inverse Problems. Number 160 in Applied Mathematical Sciences. New York: Springer. ISBN: 0-387-22073-9.Google Scholar
  16. Kolehmainen, V., Siltanen, S., Järvenpää, S., Kaipio, J. P., Koistinen, P., Lassas, M., Pirttila, J., Somersalo, E. (2003): Statistical inversion for medical X-ray tomography with few radiographs II: Application to dental radiology. Phys. Med. Biol. 48: 1465–1490.CrossRefGoogle Scholar
  17. Kolehmainen, V., Vanne, A., Siltanen, S., Järvenpää, S., Kaipio, J. P., Lassas, M., Kalke, M. (2006): Parallelized Bayesian inversion for three-dimensional dental X-ray imaging. IEEE Trans. Med. Im. 25: 218–228.CrossRefGoogle Scholar
  18. Mosegaard, M., Sambridge, M. (2002): Monte Carlo analysis of inverse problems. Inv. Probl. 18: R29–R54.MATHCrossRefMathSciNetGoogle Scholar
  19. Natterer, F. (1986): The Mathematics of Computerized Tomography. Chichester, USA: John Wiley & Son.MATHGoogle Scholar
  20. PaloDEx Group (Finland) (2007): Volumetric Tomography, [Online], Available at http://www.instrumentariumdental.com/
  21. Ramesh, A., Ludlow, J. B., Webber, R. L., Tyndall, D. A., Paquette, D. (2002): Evaluation of tuned-aperture computed tomography in the detection of simulated periodontal defects. Oral and Maxillofacial Radiology 93: 341–349.Google Scholar
  22. Ranggayyan, R. M., Dhawan, A. T., Gordon, R. (1985): Algorithms for limited-view computed tomography: an annotated bibliography and a challenge. Appl. Opt. 24: 4000–4012.Google Scholar
  23. Raydan, M. (1997): The Barzilai and Borwein gradient method for the large scale unconstrained minimization problem. SIAM J. Optim. 7: 26–33.MATHCrossRefMathSciNetGoogle Scholar
  24. Sauer, K., James, S. Jr., Klifa, K. (1994): Bayesian estimation of 3-D objects from few radiographs. IEEE Trans. Nucl. Sci. 41: 1780–1790.CrossRefGoogle Scholar
  25. Siltanen, S., Kolehmainen, V., Järvenpää, S., Kaipio, J. P., Koistinen, P., Lassas, M., Pirttila, J., Somersalo, E. (2003): Statistical inversion for medical X-ray tomography with few radiographs I: General theory. Phys. Med. Biol. 48: 1437–1463.CrossRefGoogle Scholar
  26. Sterling, T., Savarese, D., Becker, D. J., Dorband, J. E., Ranawake, U. A., Packer, C. V. (1995): BEOWULF: A parallel workstation for scientific computation. In Proc. 24th Int. Conf. on Parallel Processing, number I.Google Scholar
  27. Webber, R. L., Horton, R. A., Tyndall, D. A., Ludlow, J. B. (1997): Tuned aperture computed tomography (TACT). Theory and application for three-dimensional dento-alveolar imaging. Dentomaxillofacial Radiology 26: 53–62.CrossRefGoogle Scholar
  28. Webber, R. L., Messura, J. K. (1999): An in vivo comparison of diagnostic information obtained from tuned-aperture computed tomography and conventional dental radiographic imaging modalities. Oral Surgery Oral Medicine Oral Pathology 88: 239–247.Google Scholar
  29. Whaley, R. C., Petitet, A., Dongarra, J. J. (2001): Automated empirical optimization of software and the ATLAS project. Parallel Computing 27: 3–35.MATHCrossRefGoogle Scholar
  30. Yu, D. F., Fessler, J. A. (2002): Edge-preserving tomographic reconstruction with nonlocal regularization. IEEE Trans. Med. Imag. 21: 159–173.CrossRefGoogle Scholar
  31. Zheng, J., Saquib, S. S., Sauer, K., Bouman, C. A. (2000): Parallelizable Bayesian tomography algorithms with rapid, quaranteed convergence. IEEE Trans. Image Processing 9: 1745–1759.MATHCrossRefGoogle Scholar

Copyright information

© Springer-Verlag 2007

Authors and Affiliations

  • V. Kolehmainen
    • 1
  • A. Vanne
    • 1
  • S. Siltanen
    • 4
  • S. Järvenpää
    • 2
  • J. P. Kaipio
    • 1
  • M. Lassas
    • 2
  • M. Kalke
    • 3
  1. 1.University of KuopioKuopioFinland
  2. 2.Helsinki University of TechnologyTKKFinland
  3. 3.PaloDEx GroupTuusulaFinland
  4. 4.Tampere University of TechnologyTampereFinland

Personalised recommendations