This work shows an application of algorithms in which fuzzy techniques are used. It is focused on the automation of image analysis for use with a non-invasive technique, as magnetic resonance, in multiple sclerosis patients, and specifically in detection of the smallest lesions. The typical uncertainty in the definition of these lesions lead us to consider that a fuzzy approach is a good solution to the problem.
The design of the algorithm is based on the definition of a rule set, which enable feature extraction and data analysis. The fuzzification process is solved by means of probability density functions. In this way we obtain OWA operators that achieve a high degree of detection in these lesions.
The proposed design resolves the problem of false detections by the use of various filters implemented from new rules.
@article{urn:eudml:doc:39159,
title = {Use of fuzzy techniques for detection of multiple sclerosis small lesions.},
journal = {Mathware and Soft Computing},
volume = {5},
year = {1998},
pages = {355-363},
zbl = {0949.92502},
language = {en},
url = {http://dml.mathdoc.fr/item/urn:eudml:doc:39159}
}
Aymerich, F. Xavier; Sobrevilla, Pilar; Gili, Jaume; Montseny, Eduard. Use of fuzzy techniques for detection of multiple sclerosis small lesions.. Mathware and Soft Computing, Tome 5 (1998) pp. 355-363. http://gdmltest.u-ga.fr/item/urn:eudml:doc:39159/