Development of features quantifying respiratory dysfunctions in Parkinson’s disease patients

In the beginning of the thesis, Parkinson's disease and hypokinetic dysarthria are briefly described, which have a negative effect on speech production and cause breathing problems during speech in sick patients. The aim of the thesis is to create an algorithm for automated detection of breaths and the design of parameters for the quantification of respiratory disorders in patients with Parkinson's disease. In the MATLAB environment, the recordings of the researched subjects were processed and an algorithm was created for the detection of breaths, which used the logistic regression method. Based on the predicted breaths, proposed parameters were extracted from the recordings, which were then statistically analyzed and compared in healthy controls and patients with Parkinson's disease. By using a machine learning model, it was possible to predict the clinical data of patients from the proposed parameters to a certain extent. The average accuracy of the model for predicting puffs was 0.85. Of the 14 proposed parameters, 6 were suitable for quantifying respiratory disorders associated with hypokinetic dysarthria. The result of the work is a functional algorithm for the automated detection of breaths in the speech signal and proposed parameters that could be useful for the quantification of respiratory disorders in patients with Parkinson's disease