Preliminary statistical analysis tool


Ing. Zoltán Galáž, Ing. Jiří Mekyska, prof. Ing. Zdeněk Smékal, CSc.


It is possible to download software here.

Publication to be cited

GALÁŽ, Z. Preliminary Acoustic Analysis of Noise Components in Patients In Parkinsons Disease. In Proceedings of the 21st Conference STUDENT EEICT 2015. Brno: 2015. p. 476-480. ISBN: 978-80-214-5148- 3.

Mekyska J., Galáž Z., Mžourek Z., Smékal Z., Rektorová I., et al. (2015) Perceptual Features as Markers of Parkinson’s Disease: The Issue of Clinical Interpretability. International Conference on Non-Linear Speech Processing (NOLISP 2015): in press.


Preliminary statistical analysis is used as a tool to obtain a statistical insight into the analysed dataset. In the field of pathological signal processing the common methodology is to perform so called parametrization process in order to extract clinically useful information from the data to describe the deterioration of the speech production. The output of this process is the feature matrix, constructed of the computed feature labels for all speakers (speech recordings) in the dataset. This software can be used to calculate the correlation coefficient and the mutual information between the feature vectors and the associated clinical diagnosis. It can also be used to perform the individual feature analysis in the direction of the binary classification or regression task (continuous clinical scale estimation) computation.

The actual version of PSA software toolbox provides the possibility of using 6 worldwide used classification techniques (Support Vector Machines, Naive Bayes Networks, Discriminant Analysis, k-Nearest Neighbour, Classification Trees and Gaussian Mixture Models) one regression technique based on the Classification and Regression Trees algorithm. The software also provides 18 metrics to evaluate the classification process (classification accuracy, sensitivity, specificity, etc.) and 10 metrics to evaluate the regression task (gini index, absolute error, root mean squared error, etc.). The testing scripts demo_classification.m and demo_regression.m are also provided. The scripts load the data from the test_classification.mat and test_regression.mat files, which include the parametrization matrix “feat_matrix”: rows are determined for the observations; columns are determined for the parameters, and the vector of labels “labels” (e.g. for the classification task: 0/1 – healthy/disordered and for the regression task it is the numeric continuous scale, and the vector of feature names).


This work was supported by projects NT13499, VG20102014033 and FEKT-S-14-2335. The described research was performed in laboratories supported by the SIX project; the registration number CZ.1.05/2.1.00/03.0072, the operational program Research and Development for Innovation.


To negotiate the license terms of use of this software please contact the responsible person Ing. Lukas Novak at Technology Transfer Office, Brno University of Technology, Kounicova 966/67a, Veveří, 60200, Brno, Czech Republic,