Time frequency analysis of bioelectrical signals using LabVIEW

Today, the decision-making process depends on people has decreased dramatically with the coming of digital signal processing techniques in biomedical field. To understand the information contained in the signal, different methods have been investigated. The studies have accelerated with understanding the importance of digital signal processing techniques in this field. The aim of this thesis is to implement signal processing techniques in the biomedical field which is one of the applications of electronic systems so as to provide ease of use. In bioelectrical signal processing applications often text-based programs such as MATLAB are used. The major disadvantage in these text-based languages is requiring code memorization. In this thesis, a new and alternative method for signal processing using LabVIEW that is a graphical language based method was investigated. Thanks to the graphic encoding method, disadvantages of the text based language that is tried to be minimized. In this study, LabVIEW is used as a simulation environment. Bioelectric signals Electroencephalography (EEG) and Electrocardiogram (ECG) were performed in three different applications. In the first study, number of heart beats attempts to find from the ECG signal. In this study, as differently, different data sets have been used. A system for detecting QRS from ECG signal via wavelet transform has been developed. In second application, an interface have been created for extract parameters from RR intervals which inside in ECG signal. Signals from MIT-BIH Congestive Heart Failure RR Interval and Normal Sinus Rhythms RR Interval of thirty participants were used to obtain parameters for time and frequency domains. Obtained parameters were applied to principal component analysis. Then the obtained normalized data were applied to k-mean classification algorithm so accuracy rate of extraction parameters were determined. In latter study, EEG signals from the database DEAP of four individuals were examined. EEG recordings of the subjects were recorded during watched videos to evoke four different emotions. The wavelet transform is applied to the recorded EEG signals. The changes on EEG caused by the watched video to create different feelings were examined. Power Spectrum Density (PSD) was created to examine four different frequency ranges in the EEG.