Information Technology Reference
In-Depth Information
In seven of the ten studied images the ignorance based algorithm obtain better
results than fuzzy algorithm. With these results, we can conclude that for real ul-
trasound images, ignorance based algorithm has a total mean error lower than the
fuzzy algorithm.
12.5
Conclusions
In this chapter we have reviewed the use of fuzzy sets in medical image processing,
mainly in the task of image segmentation.
There exist a large number of successful applications that prove that fuzzy set
theory is a quality tool to handle the fuzziness present in medical images. But an
extensive comparison with classical and new methods of standard image processing
is needed.
Among fuzzy techniques, FCM is a well known and extensively method used in
medical image processing. However, there are other techniques not widely used by
the medical image processing community. For example fuzzy rule based modeling
of organs, which is one of the best advantages of fuzzy set theory has not been ex-
ploited to its full potential. Indeed, fuzzy rule based systems can include knowledge
from physicians and radiologists in the algorithms.
On the other side the definition of new measures of uncertainty or fuzziness, and
their use to minimize fuzziness is a promising topic. These measures can be defined
ad-hoc to each problem so they can represent the fuzziness from their acquisition.
Acknowledgement.
This research was partially supported by grants TIN2010-15505 and
TIN2011-29520 from the Government of Spain.
References
1. Acton, P.D., Pilowsky, L.S., Kung, H.F., Ell, P.J.: Automatic Segmentation of Dynamic
Neuroreceptor Single-photon Emission Tomography Images Using Fuzzy Clustering.
European Journal of Nuclear Medicine 26(6), 581-590 (1999)
2. Amini, L., Soltanian-Zadeh, H., Lucas, C., Gity, M.: Automatic Segmentation of Thala-
mus from Brain MRI Integrating Fuzzy Clustering and Dynamic Contours. IEEE Trans-
actions on Biomedical Engineering 51(5), 800-811 (2004)
3. Awad, J.G.E., Abdel-Galil, T., Salama, M.M.A., Tizhoosh, H.R., Fenster, A., Rizkalla,
K., Downey, D.B.: Prostate's Boundary Detection in Transrectal Ultrasound Images Us-
ing Scanning Technique. In: Olivier, G., Pierre, S., Sood, V.K. (eds.) 15th IEEE Canadian
Conference on Electrical and Computer Engineering CCECE 2003: Canadian Confer-
ence on Electrical and Computer Engineering, Montreal, Canada, May 04-07. Proceed-
ings: Toards a Caring and Humane Technology, pp. 1199-1202. IEEE, IEEE Canada,
IEEE Eastern Canada Council, IEEE Montreal, St Maurice & Quebec Sect, Ecole Poly-
tech Montreal, Manitoba HVDC Res. Ctr., Cooperat Etudiante Ecole Polytech, John Wi-
ley & Sons Canada, McGraw Hill Ryerson Ltd. Cheneliere McGraw Hill, Oxford Univ.
Press, Presses Int. Polytech (2003)
