Medical decision support system using pattern recognition methods for assessment of dermatoglyphic indices and diagnosis of down's syndrome

Abstract

The development and implementation of the telemedical system for the diagnosis of Down's syndrome is described in the chapter. The system is a tool supporting medical decision by automatic processing of dermatoglyphic prints and detecting features indicating the presence of genetic disorder. The application of image processing methods for the pre-processing and enhancement of dermatoglyphic images has also been presented. Classifiers for the recognition of fingerprint patterns and patterns of the hallucal area of the soles, which are parts of an automatic system for rapid screen diagnosing of trisomy 21 (Down's Syndrome) in infants, are created and discussed. The method and algorithms for the calculation of palmprint's ATD angle are presented then. The images of dermatoglyphic prints are pre-processed before the classification stage to extract features analyzed by Support Vector Machines algorithm. The application of an algorithm based on multi-scale pyramid decomposition of an image is proposed for ridge orientation calculation. RBF and triangular kernel types are used in training of SVM multi-class systems generated with one-vs.-one scheme. A two stage algorithm for the calculation of palmprint's singular points location, based on improved Poincare index and Gaussian-Hermite moments is subsequently discussed. The results of experiments conducted on the database of Collegium Medicum of the Jagiellonian University in Cracow are presented. © 2014 Springer International Publishing Switzerland.