Generation of a Fuzzy Classifier Rule Base for Diagnosing Parkinson’s Disease from Handwritten Data

Parkinson’s disease is a neurodegenerative neurological disease whose progression can be slowed if accurate and timely diagnosis is received. In this connection, the development of simple and accessible screening methods is relevant, one of which is the analysis of handwriting and drawing. A variant of performing such a method based on the application of fuzzy classifier is offered in the work. The author’s algorithm of the formation of bases of fuzzy rules which feature consists in application of mining clustering after carrying out of adjustment of parameters on concrete data is offered. To carry out the procedure for finding parameters, Powell’s optimization algorithm is chosen; as the target function, the balanced accuracy and the ratio of the number of rules to the number of training samples is used. The effectiveness of the proposed algorithm is compared with the classical k-means clustering algorithm and the extreme class feature algorithm. The experiment was performed on original datasets obtained by extracting features from publicly available signal databases taken during tests for Parkinson’s disease diagnosis.