Classification of schizophrenia patients through empirical wavelet transformation using electroencephalogram signals
Smith K. Khare, Varun Bajaj, Siuly Siuly, G R Sinha
Abstract
In this chapter, empirical wavelet transformation is used to decompose the highly nonstationary electroencephalogram signals into modes in a Fourier spectrum. Linear and non-linear time domain features are extracted from the modes. Highly discriminant features are selected using the Kruskal–Wallis test. Different types of classification techniques are employed to classify the healthy and patients with schizophrenia. The effectiveness of the system is measured by evaluating various performance parameters such as accuracy, sensitivity, precision and specificity.
Topics & Concepts
Pattern recognition (psychology)WaveletTransformation (genetics)Artificial intelligenceLinear discriminant analysisSchizophrenia (object-oriented programming)ElectroencephalographyComputer scienceSensitivity (control systems)Wavelet transformSpeech recognitionMathematicsPsychologyEngineeringElectronic engineeringNeuroscienceChemistryProgramming languageBiochemistryGeneEEG and Brain-Computer InterfacesECG Monitoring and AnalysisFault Detection and Control Systems