The detection of stress is important because it contributes to diverse pathophysiological changes including sudden death. Various techniques have been used to evaluate stress in terms of questionnaire or by quantifying the changes of physiological signals. Electroencephalogram signals are highly useful in measuring human stress. Therefore, to solve and detect stress problem, this work had extracted electroencephalogram features of theta, alpha, and beta bands in the frequency domain by wavelet packet transform because these bands are concerned with stress. In this research four features have been supplied to extreme learning machine which gave accuracy of 98.56% of detecting stress from normal state based on db4 with an average sensitivity of 92.52% and specificity of 95.88%. This research studied the stress on 15 students due to mathematical exercises in a noisy environment with different stimulus.

The detection of fatigue level is important because it is the main reason of sudden death. This research depended on the average heartbeat of the electrocardiogram signal, and the features were extracted from its high frequency components. Therefore, there is great need to transform signal into frequency domain by discrete wavelet transform. In this research, 6 features were supplied to probabilistic neural network which gave accuracy of 60.56% of detecting high level among other levels of medium and low fatigue. This research studied the fatigue on 40 students due to mathematical exercises in a noisy environment with different stimuli.

This paper presents the classification of electroencephalogram (EEG) signals using artificial neural network techniques. The signal processing of EEG signal could provide several areas for research in biomedical field. Numerous techniques can be applied to extract out the EEG characteristics in order to study and investigates the problems in the pattern recognition by its features extracted. The interesting site of signal measurement is the temporal lobe which is responsible of T3 and T4 in human electrode placement scalp. In this paper, many subjects were used to test the performance of non-neurophysiologic signals in order to investigate the electrical waves in human brain via the production of numerous EEG signals. A linear method of discrete wavelet transform (DWT) was used to gain classification with accuracy of 94.93% for testing EEG of different samples of music such as rock, jazz, classical and heavy metal using artificial neural network (ANN) with 2000 epoch, 25 nodes, 2 hidden layers. The results showed promisingly valuable EEG signal characteristics which could support the hospital staff to take care of and treat patients in the correct direction.