Knowledge-based contour detection in medical imaging using fuzzy logic

Soft computing (e.g. fuzzy logic, neural network, and genetic algorithms) has proved to yield promising results in digital image processing and understanding when missing, ambiguous or distorted data is available according to H. Costin and Cr. Rotariu (2001) and D. Dubois et al. (1993). For biomedical image analysis, archiving and retrieval, the great structural information may be successfully approached by using methods of soft computing. Moreover, symbolic calculus (e.g. predicate logic, semantic nets, frames, scripts) may be used for knowledge representation, thus merging the expert's domain into a decision support system. This paper describes the use of fuzzy logic and semantic knowledge for edge detection and segmentation of magnetic resonance (MR) images of brain. Promising results show the superiority of this knowledge-based approach over best traditional techniques in terms of segmentation errors. The proposed methodology can be successfully used for model-driven in the domain of MRI.