Automatic Segmentation of the Encephalic Parenchyma Using Fuzzy Techniques.

摘要

This work shows an automatic, fast and reproducible algorithm to segment the encephalic parenchyma in magnetic resonance (MR) images. The algorithm has been implemented following a rule-based schema in which a fuzzy analysis of MR images information has been introduced to deal with the vagueness associated to the images. The obtention of a fuzzy result helps to determine the accuracy of the classification. The evaluation of the results is based on the use of quality indexes, which allow the comparison with previous works.