Medical Decision Support System Architecture for Diagnosis of Down's Syndrome

The paper presents the development of a new system that is used to solve the problem of the recognition of the dermatoglyphic pattern and the understanding of the classification process of the symptoms of Down's syndrome. The method used in the system for diagnosing Down's syndrome in infants is based on the combination of text knowledge found in the scientific literature describing Down's syndrome with the knowledge obtained from the analysis of dermatoglyphic indices characteristic of Down's syndrome with the use of digital pattern recognition techniques. The scientific goal is to design a classifier system that realizes automatic medical diagnosis through the application of an expert system designed on the basis of knowledge included in the scientific text descriptions of the Down's syndrome. One other aim is the application of the pattern recognition algorithms to the analysis of indices present in the images of dermatoglyphic patterns. This approach, similar to the approach used by anthropologists, is realized by the system through the juxtaposition of the knowledge described in the form of expert system rules and the information provided by the appropriate digital equipment, and on the basis of this juxtaposition an arbitrary classification of the investigated patterns is performed.