Saeid Moslehpour

Sleep Stage Classification Using EEG Signal Analysis: A Comprehensive Survey and New Investigation

Sleep specialists often conduct manual sleep stage scoring by visually inspecting the patient’s neurophysiological signals collected at sleep labs. This is, generally, a very difficult, tedious and time-consuming task. The limitations of manual sleep stage scoring have escalated the demand for developing Automatic Sleep Stage Classification (ASSC) systems. Sleep stage classification refers to identifying the various stages of sleep and is a critical step in an effort to assist physicians in the diagnosis and treatment of related sleep disorders. The aim of this paper is to survey the progress and challenges in various existing Electroencephalogram (EEG) signal-based methods used for sleep stage identification at each phase; including pre-processing, feature extraction and classification; in an attempt to find the research gaps and possibly introduce a reasonable solution. Many of the prior and current related studies use multiple EEG channels, and are based on 30 s or 20 s epoch lengths which affect the feasibility and speed of ASSC for real-time applications. Thus, in this paper, we also present a novel and efficient technique that can be implemented in an embedded hardware device to identify sleep stages using new statistical features applied to 10 s epochs of single-channel EEG signals. In this study, the PhysioNet Sleep European Data Format (EDF) Database was used. The proposed methodology achieves an average classification sensitivity, specificity and accuracy of 89.06%, 98.61% and 93.13%, respectively, when the decision tree classifier is applied. Finally, our new method is compared with those in recently published studies, which reiterates the high classification accuracy performance.

Stand-Alone Surface Roughness Analyzer

This paper details the design and implementation of a noncontact surface roughness probe from a PC-based data-acquisition system to a stand-alone measurement instrument system. A Cadence layout for the fabrication of the printed circuit board (PCB), which interfaces and drives the surface roughness probe, was used to prototype this project.

An Intelligence-Based Model for Condition Monitoring Using Artificial Neural Networks

This paper reports a newly developed Condition-Based Maintenance CBM model based on Artificial Neural Networks ANNs which takes into account a feature e.g., vibration signals from a machine to classify the condition into normal or abnormal. The model can reduce equipment downtime, production loss, and maintenance cost based on a change in equipment condition e.g., changes in vibration, power usage, operating performance, temperatures, noise levels, chemical composition, debris content, and volume of material. The model can effectively determine the maintenance/service time that leads to a low maintenance cost in comparison to other types of maintenance strategy. Neural Networks tool NNTool in Matlab is used to apply the model and an illustrative example is discussed.

Design of the Nios II System for the Playing of Wave Files on an Altera DE2 Board

The motivation behind this study is the impact made on todays social media by music players, which include features like portability, size and equalizer functionality for the best possible sound output quality. The objective of this study was to develop a system that reads the wave files present on the Secure Data (SD) card, adjust the equalizer settings incorporated on it, and play them on the speaker with the best possible quality sound output. This was done with the help of the SD card slot provided on the DE2 board, and its implementation on the board using Alteras SoPC (System-on- a-Programmable-Chip) Builder in the Altera Quartus 9.1 environment. Nios II is a 32-bit soft-core embedded processor architecture designed specifically for the Altera family of FPGAs. Programming of the Nios II processor is done using the Nios II 9.1 IDE tool. An extension of this work could include incorporation of video—along with the audio on the DE2 board since there is a VGA slot included on the board features and the movement of the wave files on the SD card with the help of the keys present on the DE2 board .

Design of RISC Processor Using VHDL and Cadence

The project deals about development of a basic RISC processor. The processor is designed with basic architecture consisting of internal modules like clock generator, memory, program counter, instruction register, accumulator, arithmetic and logic unit and decoder. This processor is mainly used for simple general purpose like arithmetic operations and which can be further developed for general purpose processor by increasing the size of the instruction register. The processor is designed in VHDL by using Xilinx 8.1i version. The present project also serves as an application of the knowledge gained from past studies of the PSPICE program. The study will show how PSPICE can be used to simplify massive complex circuits designed in VHDL Synthesis. The purpose of the project is to explore the designed RISC model piece by piece, examine and understand the Input/ Output pins, and to show how the VHDL synthesis code can be converted to a simplified PSPICE model. The project will also serve as a collection of various research materials about the pieces of the circuit.

Various Methods of Economical Load Distribution in Power Plant Units in Comparison to Neural Networks Method

Initial capital investment for setting up an electrical power plant is huge, and the annual operating costs of major and preventive maintenance, administration, labor, fuel consumption, etc., amount to a very large sum. Therefore, optimized operation of the power plants and accomplishment of the highest level of utilization is of utmost importance.

Mosfet Amplifier Using Advanced Analysis in PSpice

This paper incorporates the use of Advanced Analysis [3] of PSpice for exposing the students to some involving procedures of circuit building. After reading through the various properties that have been displayed, it would seem as a compulsory learning and assessment tool for the students which not only helps them interact with the circuitry, but, become much more involved with the practical aspects of electrical systems.

Characterization of human oral tissues based on quantitative analysis of optical coherence tomography images

In this paper, five types of tissues, human enamel, human cortical bone, human trabecular bone, muscular tissue, and fatty tissue were imaged ex vivo using optical coherence tomography (OCT). The specimens were prepared in blocks of 5 x 5 x 3 mm (width x length x height). The OCT imaging system was a swept source OCT system operating at wavelengths ranging between 1250 nm and 1360 nm with an average power of 18 mW and a scan rate of 50 to 100 kHz. The imaging probe was placed on top of a 2 x 2 cm stabilizing device to maintain a standard distance from the samples. Ten image samples from each type of tissue were obtained. To acquire images with minimum inhomogeneity, imaging was performed multiple times at different points. Based on the observed texture differences between OCT images of soft and hard tissues, spatial and spectral features were quantitatively extracted from the OCT images. The Radon transform from angles of 0 deg to 90 deg was computed, averaged over all the angles, normalized to peak at unity, and then fitted with Gaussian function. The mean absolute values of the spatial frequency components of the OCT image were considered as a feature, where 2-D fast Fourier transform (FFT) was done to OCT images. These OCT features can reliably differentiate between a range of hard and soft tissues, and could be extremely valuable in assisting dentists for in vivo evaluation of oral tissues and early detection of pathologic changes in tissues.

A near real-time, parallel and distributed adaptive object detection and retraining framework based on AdaBoost algorithm

Object detection (e.g., face detection) using supervised learning often requires extensive training, resulting in long execution times. If the system requires retraining to accommodate a missed detection, waiting several hours or even days in some cases before the system is ready, may not be acceptable in practical implementations. This paper presents a generalized object detection framework such that the system can efficiently adapt to misclassified data and be retrained within a few minutes. The methodology developed here is based on the popular AdaBoost algorithm for object detection. To reduce the learning time in object detection, we develop a highly efficient, parallel, and distributed AdaBoost algorithm that is able to achieve a training execution time of only 1.4 seconds per feature on 25 workstations. Further, we incorporate this parallel object detection algorithm into an adaptive framework such that a much smaller, optimized training subset is used to yield high detection rates while further reducing the retraining execution time. We demonstrate the usefulness of our adaptive framework on face and car detection.

Opto-Mechanical Methodology for Feature Detection of Contours

Precise measurement of curvature is of critical importance in the aerospace industry. This paper presents a methodology for non-contact measurement of radius of curvature on the finished surface of manufactured parts. Unique optical methods including a Line Laser and a backlight are explored. In each case, a light source illuminates the features to be measured and several sets of high-contrast images are produced. For that purpose a low power (2mW) Helium-Neon (HeNe) laser source has been used. The images are processed using specific software and data points are mapped along the region of interest. A mathematical relationship is used to extrapolate the radius of curvature from the measured data.Copyright