Abstract
A method is described to rescale images in order to equalize image noise. Such a scale conversion appears to be an effective way to obtain a uniform sensitivity of feature detection in digitized X-ray images. A robust algorithm is proposed to determine a proper scale conversion from a phantom recording, or from an image to be processed itself. The latter approach, in which noise characteristics are estimated from the image at hand, appeared to be preferable in an application to mammography. In this application the performance of a statistical method for detection of microcalcifications was investigated on a set of 40 digitized mammographic images. It was found that logarithmic scaling, which is often used, is not appropriate because it strongly biases the sensitivity of feature detection. Results of adaptive scaling were superior to those obtained by using a fixed scale transform.
This research was supported by Siemens Medical Systems and by the Dutch Prevention Fund.
This is a preview of subscription content, log in via an institution to check access.
Preview
Unable to display preview. Download preview PDF.
References
Karssemeijer N and van Erning L J Th O: Iso-precision scaling of digitized mammograms to facilitate image analysis. SPIE Med. Im. V, Image Processing, 166–177 (1990)
Jain A K: Fundamentals of digital image processing. New Yersey: Prentice-Hall (1998)
Webb S: The Physics of medical imaging. Bristol: Adam Hilger (1988)
Karssemeijer N: A stochastic model for automated detection of calcifications in digital mammograms. Proc. IPMI 91, 227–238. Reprinted in: Image and Vision Computing 10 369–375 (1992)
Nab H W, Karssemeijer N, van Erning L J Th O, and Hendriks J H C L: Comparison of digital and conventional mammography: a ROC study of 270 mammograms. Med. Inform. 17 125–131 (1992)
Karssemeijer N, Frieling J T M, and Hendriks J H C L: Spatial resolution in digital mammography. Invest. Radiol. (accepted).
Besag J E: On the statistical analysis of dirty pictures. J. Royal. Statist. Soc., Ser. B 48 259–302 (1986)
Dubes R C, Jain A K: Random field models in image analysis. J. Appl. Stat. 16 131–164 (1989)
Egan JP, Greenberg GZ, Schulman AI: Operating characteristics, signal detectability, and the method of free response. J. Acoust. Soc. Am. 33 993–1007 (1961)
Chakraborty D P, Winter L H L: Free response methodology: Alternate analysis and a new observer-performance experiment. Radiology 174 873–881 (1990)
Chan H P, Doi K, Vyborny C J, Lam K L and Schmidt R A: Computer-Aided detection of microcalcifications in mammograms. Invest. Radiol. 23 664–671 (1988)
Davies D H and Dance D R: Automatic computer detection of clustered calcifications in digital mammograms. Phys. Med. Biol. 35 1111–1118 (1990)
Astley S M and Taylor C J: Combining cues for mammographic abnormalities. Proc. 1st. Brit. Mach. Vision Conf. 253–258 (1990)
Author information
Authors and Affiliations
Editor information
Rights and permissions
Copyright information
© 1993 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Karssemeijer, N. (1993). Adaptive noise equalization and image analysis in mammography. In: Barrett, H.H., Gmitro, A.F. (eds) Information Processing in Medical Imaging. IPMI 1993. Lecture Notes in Computer Science, vol 687. Springer, Berlin, Heidelberg. https://doi.org/10.1007/BFb0013806
Download citation
DOI: https://doi.org/10.1007/BFb0013806
Published:
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-56800-1
Online ISBN: 978-3-540-47742-6
eBook Packages: Springer Book Archive