Abstract
Global polynomial (GP) methods have been widely used to correct geometric image distortion of small-size (up to 30 cm) X-ray image intensifiers (XRIIs). This work confirms that this kind of approach is suitable for 40 cm XRIIs (now increasingly used). Nonetheless, two local methods, namely 3rd-order local un-warping polynomials (LUPs) and hierarchical radial basis function (HRBF) networks are proposed as alternative solutions. Extensive experimental tests were carried out to compare these methods with classical low-order local polynomial and GP techniques, in terms of residual error (RMSE) measured at points not used for parameter estimation. Simulations showed that the LUP and HRBF methods had accuracies comparable with that attained using GP methods. In detail, the LUP method (0.353 μm) performed worse than HRBF (0.348 μm) only for small grid spacing (15×15 control points); the accuracy of both HRBF (0.157 μm) and LUP (0.160 μm) methods was little affected by local distortions (30×30 control points); weak local distortions made the GP method poorer (0.320 μm). Tests on real data showed that LUP and HRBF had accuracies comparable with that of GP for both 30 cm (GP: 0.238 μm; LUP: 0.240 μm; HRBF: 0.238 μm) and 40 cm (GP: 0.164 μm; LUP: 0.164 μm; HRBF: 0.164 μm) XRIIs. The LUP-based distortion correction was implemented in real time for image correction in digital tomography applications.
This is a preview of subscription content, log in via an institution to check access.
Similar content being viewed by others
References
Borghese, N. A., andFerrari, S. (1998): ‘Hierarchical RBF networks in function approximation’,Neurocomputing,19, pp. 259–283
Cerveri, P., Ferrari, S., andBorghese, N. A. (1997): ‘Camera calibration through RBF networks’, Proc. SPIE97, Application of Soft Computing. S. Diego, CA, pp. 312–318.
Chakraborty, D. P. (1987): ‘Image intensifier distortion correction’,Med. Phys.,14, pp. 249–252.
Chotas, H. G., Dobbins, J. T. III, andRavin, C. E. (1999): ‘Principles of digital radiography with large-area, electronically readable detectors: a review of the basics’,Radiology,210, pp. 595–599.
Cosby, S., Leszczynski, K., Dennie, J., andDunscombe, P. (1994): ‘Image intensifier distortion correction for large field of view digital radiation therapy simulation’,Med. Phys.,21, p. 886.
Fahrig, R., Morea, M., andHoldsworth, D. W. (1997): ‘Three-dimensional computed tomographic reconstruction using a C-arm mounted XRII: correction of image intensifier distortion’,Med. Phys.,24, pp. 1097–1106.
Fahrig, R., Butts, K., Saunders, R.,et al. (2001): ‘A truly hybrid interventional MR/X-ray system: feasibility demonstration’,J. Magn. Reson. Imaging,13, pp. 294–300.
Forlani, C., andFerrigno, G. (2001): ‘Automatic real-time local distortion correction method for digital linear tomography’ inVassil, N. A., Dongarra, J. J., Juliano, B. J., Renner, R. S., andKenneth Tan, C. J. (Eds): ‘ICCS 2001 Computational Science Proceedings (2)’, (Springer, 2001), pp. 23–26.
Girosi, F., Jones, M., andPoggio, T. (1995): ‘Regularization theory and neural networks architectures’,Neural Comput. 7, pp. 219–269.
Gronenschild, E. (1993). ‘Effect of spatial orientation on the image-intensifier distortion in quantitative coronary angiography’.Proc. Computers in Cardiology’93, pp. 595–598.
Gronenschild, E. (1997): ‘The accuracy and reproducibility of a global method to correct for geometric distortion in the x-ray chain’,Med. Phys.,24, pp. 1875–1888.
Gronenschild, E. (1999): ‘Correction for geometric image distortion in the x-ray imaging chain: local technique versus global technique’,Med. Phys.,26, pp. 2602–2616.
Liu, R. R., Rudin, S., Bednarek, D. R. (1999): ‘Super-global distortion correction for a rotational C-arm x-ray image intensifier’,Med. Phys.,26, pp. 1802–1810.
Martinez, T. M., Berkovich, S. G., Schulten, K. J. (1993): ‘Neural-gas network for vector quantization and its application to time-series prediction’,IEEE Trans. Neural Netw.,4, p. 558.
Park, J., andSandberg, I. W. (1991): ‘Universal approximation using radial-basis-function networks’,Neural Comput.,3, p. 246.
Reimann, D., andFlynn, M. J. (1992): ‘Automated distortion correction of x-ray image intensifier’, IEEE Nuclear Science Symposium and Medical Imaging Conference Record, Orlando, Florida, pp. 1339–1341.
Rudin, S., Bednarck, D. R., andWong, R. (1991): ‘Accurate characterization of image intensifier distortion’,Med. Phys.,18, pp. 1145–1151.
Sanner, R. M., andSlotine, J. E. (1992): ‘Gaussian networks for direct adaptive control’,IEEE Trans. Neural Networks,3, p. 837.
Shah, S., andAggarwal, J. K. (1996): ‘Intrinsic parameter calibration procedure for a (high distortion) fish-eye lens camera with distortion model and accuracy estimation’,Pattern Recognit.,29, pp. 1775–1788.
Shih, S.-W., Hung, Y.-P., andLin, W.-S. (1995): ‘When should consider lens distortion in camera calibration’,Pattern Recognit.,28, pp. 447–461.
Solzbach, U., Wollschläger, H., Zeiher, A., andJust, H. (1989): ‘Computers in cardiology 1988’ (IEEE Computer Society Press, Los Alamitos, 1989), pp. 355–357.
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Cerveri, P., Forlani, C., Pedotti, A. et al. Hierarchical radial basis function networks and local polynomial un-warping for X-ray image intensifier distortion correction: A comparison with global techniques. Med. Biol. Eng. Comput. 41, 151–163 (2003). https://doi.org/10.1007/BF02344883
Received:
Accepted:
Issue Date:
DOI: https://doi.org/10.1007/BF02344883