Detekce poruch chůze u osob s Parkinsonovou nemocí

Abstract

The main aim of this master thesis is to design algorithm for detection of freezing of gait of patients with parkinson’s disease. Part of the thesis is to design an experiment for obtaining uniform qualifying data frm measurements of 2 groups. Control group and Parkinson’s disease group. The experiment uses Xsens Awinda and Novel Emed-XL. Measurements on these devices are synchronized by hardware and the results are exported and uploaded to MySQL database which was created on and is available on Google Cloud SQL server. The experiment is designed according to the TUG test, where people are asked simple math questions to divert their attention from the measurements being made, so that patients with parkinson’s disease experience freezing during the measurement. Detection of gait is performer on datasets that were available before the measurement started. Neural network time series and neural network pattern recognition were used for detection. In the future, this solution coudl serve to control the patient’s condition.