Currently in the world there is an alarming number of people who suffer from sleep disorders. A number of biomedical signals, such as EEG, EMG, ECG and EOG are used in sleep labs among others for diagnosis and treatment of sleep related disorders. The usual method for sleep stage classification is visual inspection by a sleep specialist. This is a very time consuming and laborious exercise. Automatic sleep stage classification can facilitate this process. The definition of sleep stages and the sleep literature show that EEG signals are similar in Stage 1 of non-rapid eye movement (NREM) sleep and rapid eye movement (REM) sleep. Therefore, in this work an attempt was made to classify four sleep stages consisting of Awake, Stage 1 + REM, Stage 2 and Slow Wave Stage based on the EEG signal alone. Wavelet packet coefficients and artificial neural networks were deployed for this purpose. Seven all night recordings from Physionet database were used in the study. The results demonstrated that these four sleep stages could be automatically discriminated from each other with a specificity of 94.4 +/- 4.5%, a of sensitivity 84.2+3.9% and an accuracy of 93.0 +/- 4.0%.

No abstract

No abstract

K-complexes like spindles are hallmark patterns of stage 2 sleep. Due to correlation between these patterns and some diseases, it is necessary to develop algorithms to detect them. In this study, a new method is used to detect K-complexes automatically. 10 time-series and chaotic features were used in order to extract the K-complex waves from stage 2 sleep. To use the most effective features, feature space dimension is reduced with Sequential Forward Selection method. The reduced feature space is classified using Generalized Radial Basis Function Extreme Learning Machine (MELM-GRBF) algorithm. GRBFs make the modification of the RBF possible by adjusting a new parameter . We're applied this methodology to K-complex classification for the first time. The classifier gives noticeably better results compared to ELM-RBF method for sensitivity and accuracy .± . and . ± . , respectively.

Lumbar disc diseases are the commonest complaint of Lower Back Pain (LBP). In this paper, a new method for automatic diagnosis of lumbar disc herniation is proposed which is based on clinical Magnetic Resonance Images (MRI) data. We use T2-W sagittal and myelograph images. Our method uses Otsu thresholding method to extract the spinal cord from MR images of Lumbar disc. In the next step, a third-order polynomial is aligned on the extracted spinal cords, and in the end of preprocessing step all the T2-W sagittal images are prepared for extracting disc boundary and labeling. After labeling and extracting a ROI for each disc, intensity and shape features are used for classification. The presented Method is applied on 30 clinical cases, each containing 7 discs (210 lumbar discs) for the herniation diagnosis. The results revealed 92.38% and 93.80% accuracy for Artificial Neural Network and Support Vector Machine (SVM) classifiers, respectively. The results indicate the superiority of the proposed method to those mentioned in similar studies.