![](bg8.png)
6 CONCLUSIONS
Computer derived features from 2D ultrasound im-
ages of the thyroid glands were used as part of a pro-
totype biometric system. These features are related
to the acoustic impedance, texture and morphology of
the thyroid tissue.
Good results were achieved with the MAP clas-
sifier, when using the three most discriminant fea-
tures, computed by PCA. Moreover, reasonably high
identification rates were also achieved with the en-
tropy distance classifiers, suggesting that the acoustic
impedance, or reflectivity, of the tissues is a relevant
feature to discriminate between individuals. Analy-
sis of thyroid echo-morphology should be further ex-
ploited because it appears to be very useful not only
as a (soft) biometric system but also as a diagnostic
tool.
Preliminary results, using only 9 parameters ex-
tracted from ultrasoundimages, are encouraging. Fur-
ther studies, involving larger data sets (more individ-
uals and more samples), as well as observations taken
from multiple sessions along distinct time instants,
are required to better establish the accuracy of this
new biometric modality.
REFERENCES
J. Abbot and F. Thurstone. Acoustic speckle: Theory and
experimental analysis. Ultrasound Imaging, 1:303–
324, 1979.
C. Burckhardt. Speckle in ultrasound b-mode scans. IEEE
Transations on Sonics and Ultrasonics, SU-25(1):1–
6, January 1978.
Stephen Boyd and Lieven Vandenberghe. Convex Optimiza-
tion. Cambridge University Press, 2004.
Y. Boykov, O. Veksler, and R. Zabih. Fast approximate en-
ergy minimization via graph cuts. IEEE Trans. Pattern
Anal. Mach. Intell., 23(11):1222–1239, 2001.
S. Catherine, L. Maria, A. Aristides, and V. Lambros. Quan-
titative image analysis in sonograms of the thyroid
gland. Nuclear Instruments and Methods in Physics
Research A, 569:606–609, December 2006.
J. Dias, T. Silva, and J. Leitão. Adaptive restoration
of speckled SAR images using a compound random
markov field. In Procedings IEEE International Con-
ference on Image Processing, Vol.II, pages 79–83,
Chicago, USA, October 1998. IEEE.
R. M. Haralick, Dinstein, and K. Shanmugam. Textural
features for image classification. IEEE Transactions
on Systems, Man, and Cybernetics, SMC-3:610–621,
November 1973.
V. Kolmogorov and R. Zabih. What energy functions can be
minimizedvia graph cuts? IEEE Trans. Pattern Anal.
Mach. Intell., 26(2):147–159, 2004.
Guy Mailloux, Michel Bertrand, Robert Stampfler, and
Serge Ethier. Computer analysis of echographic tex-
tures in hashimoto disease of the thyroid. Journal of
Clinical Ultrasound, 14(7):521–527, 1986.
O. V. Michailovich and A. Tannenbaum. Despeckling of
medical ultrasound images. IEEE Transactions on
Ultrasonics, Ferroelectrics and Frequency Control,
53(1):64–78, 2006.
Salil Prabhakar, Josef Kittler, Davide Maltoni, Lawrence
O’Gorman, and Tieniu Tan. Introduction to the special
issue on biometrics: Progress and directions. IEEE
Trans. Pattern Anal. Mach. Intell., 29(4):513–516,
2007.
M.A. Savelonas, D.K. Iakovidis, N. Dimitropoulos, and
D. Maroulis. Computational characterization of thy-
roid tissue in the radon domain. Computer-Based
Medical Systems, 2007. CBMS ’07. Twentieth IEEE
International Symposium on, pages 189–192, June
2007.
Daniel Smutek, Radim Sara, Petr Sucharda, and Ludvik
Tesar. Different types of image texture features in ul-
trasound of patients with lymphocytic thyroiditis. In
ISICT ’03: Proceedings of the 1st international sym-
posium on Information and communication technolo-
gies, pages 100–102. Trinity College Dublin, 2003.
Daniel Smutek, Radim Sara, Petr Sucharda, and Lud-
vik Tesar. Image texture analysis of sonograms in
chronic inflammations of thyroid gland. Ultrasound
in Medicine and Biology, 29:1531–1543(13), Novem-
ber 2003.
José Seabra, João Xavier, and João Sanches. Convex ultra-
sound image reconstruction with log-euclidean priors.
In In Proc. of the Engineering in Medicine and Biol-
ogy Conference, Vancouver, Canada, 2008.
Tortora Gerard J Tortora, Gerard J. (Gerard Joseph). Prin-
ciples of anatomy and physiology, 2000.
C. M. van Bemmel, L. Spreeuwers, M.A. Viergever, and
W.J. Niessen. Level-set based carotid artery seg-
mentation for stenosis grading. In MICCAI ’02:
Proceedings of the 5th International Conference on
Medical Image Computing and Computer-Assisted
Intervention-Part II, pages 36–43, London, UK, 2002.
Springer-Verlag.
C. Xu and J.L. Prince. Snakes, shapes, and gradient vector
flow. IEEE Transactions on Image Processing, 7(3),
March 1998.
A BIOMETRIC IDENTIFICATION SYSTEM BASED ON THYROID TISSUE ECHO-MORPHOLOGY
193