Enas Abdulhay

Seizure classification in EEG signals utilizing Hilbert-Huang transform

BackgroundClassification method capable of recognizing abnormal activities of the brain functionality are either brain imaging or brain signal analysis. The abnormal activity of interest in this study is characterized by a disturbance caused by changes in neuronal electrochemical activity that results in abnormal synchronous discharges. The method aims at helping physicians discriminate between healthy and seizure electroencephalographic (EEG) signals.MethodDiscrimination in this work is achieved by analyzing EEG signals obtained from freely accessible databases. MATLAB has been used to implement and test the proposed classification algorithm. The analysis in question presents a classification of normal and ictal activities using a feature relied on Hilbert-Huang Transform. Through this method, information related to the intrinsic functions contained in the EEG signal has been extracted to track the local amplitude and the frequency of the signal. Based on this local information, weighted frequencies are calculated and a comparison between ictal and seizure-free determinant intrinsic functions is then performed. Methods of comparison used are the t-test and the Euclidean clustering.ResultsThe t-test results in a P-value < 0.02 and the clustering leads to accurate (94%) and specific (96%) results. The proposed method is also contrasted against the Multivariate Empirical Mode Decomposition that reaches 80% accuracy. Comparison results strengthen the contribution of this paper not only from the accuracy point of view but also with respect to its fast response and ease to use.ConclusionAn original tool for EEG signal processing giving physicians the possibility to diagnose brain functionality abnormalities is presented in this paper. The proposed system bears the potential of providing several credible benefits such as fast diagnosis, high accuracy, good sensitivity and specificity, time saving and user friendly. Furthermore, the classification of mode mixing can be achieved using the extracted instantaneous information of every IMF, but it would be most likely a hard task if only the average value is used. Extra benefits of this proposed system include low cost, and ease of interface. All of that indicate the usefulness of the tool and its use as an efficient diagnostic tool.

Classification of focal and non focal EEG using entropies

Entropy based features and non linear features are used for classification.NNGE Classifier with optimized features enhances sensitivity.The computation time is very low leading to real time identification. Electroencephalogram (EEG) is the recording of the electrical activity of the brain which can be used to identify different disease conditions. In the case of a partial epilepsy, some portions of the brain is affected and the EEG measured from that portions are called as Focal EEG and the EEG measured from other regions is termed as Non Focal EEG. The identification of Focal EEG assists the doctors in finding the epileptogenic focus and thereby go for surgical removal of those portions of the brain for those who are having drug resistant epilepsy. In this work, we have proposed a classification methodology to classify Focal and Non Focal EEG. We used the Bern Barcelona database and used entropies such as Approximate entropy (ApEn), Sample entropy (SampEn) and Reynis entropy as features. These features were fed into six different classifiers such as Nave Bayes (NBC), Radial Basis function (RBF), Support Vector Machines (SVM), KNN classifier, Non-Nested Generalized Exemplars classifier (NNge) and Best First Decision Tree (BFDT) classifier. It was found that NNge classifier gave the highest accuracy of 98%, sensitivity of 100% and specificity of 96%, which is the highest comparing to other methods in the literature. In addition to the above, the maximum computation time of our features is 0.054 seconds which opens the window for real time processing. Thus our method can be written as a handy software tool towards assisting the physician.

Computer Aided Solution for Automatic Segmenting and Measurements of Blood Leucocytes Using Static Microscope Images

Blood leucocytes segmentation in medical images is viewed as difficult process due to the variability of blood cells concerning their shape and size and the difficulty towards determining location of Blood Leucocytes. Physical analysis of blood tests to recognize leukocytes is tedious, time-consuming and liable to error because of the various morphological components of the cells. Segmentation of medical imagery has been considered as a difficult task because of complexity of images, and also due to the non-availability of leucocytes models which entirely captures the probable shapes in each structures and also incorporate cell overlapping, the expansive variety of the blood cells concerning their shape and size, various elements influencing the outer appearance of the blood leucocytes, and low Static Microscope Image disparity from extra issues outcoming about because of noise. We suggest a strategy towards segmentation of blood leucocytes using static microscope images which is a resultant of three prevailing systems of computer vision fiction: enhancing the image, Support vector machine for segmenting the image, and filtering out non ROI (region of interest) on the basis of Local binary patterns and texture features. Every one of these strategies are modified for blood leucocytes division issue, in this manner the subsequent techniques are very vigorous when compared with its individual segments. Eventually, we assess framework based by compare the outcome and manual division. The findings outcome from this study have shown a new approach that automatically segments the blood leucocytes and identify it from a static microscope images. Initially, the method uses a trainable segmentation procedure and trained support vector machine classifier to accurately identify the position of the ROI. After that, filtering out non ROI have proposed based on histogram analysis to avoid the non ROI and chose the right object. Finally, identify the blood leucocytes type using the texture feature. The performance of the foreseen approach has been tried in appearing differently in relation to the system against manual examination by a gynaecologist utilizing diverse scales. A total of 100 microscope images were used for the comparison, and the results showed that the proposed solution is a viable alternative to the manual segmentation method for accurately determining the ROI. We have evaluated the blood leucocytes identification using the ROI texture (LBP Feature). The identification accuracy in the technique used is about 95.3%., with 100 sensitivity and 91.66% specificity.

Convolutional neural network for bio-medical image segmentation with hardware acceleration

Abstract Application of artificial intelligence in Bio-Medical image processing is gaining more and more importance in the field of Medical Science. The bio medical images, has to go through several steps before the diagnosis of the disease. Firstly, the images has to be acquired and preprocessing has to be done and the data has to be stored in memory. It requires huge amount of memory and processing time. Among the preprocessing steps, edge detection is one of the major step. Edge detection filters the unwanted details in the image, and preserves the edges of the image, which describe the boundary of the image. In biomedical application, for the detection of the diseases, it is very essential to have the boundary detail of the acquired image of the organ under observation. Thus it is very essential to extract the edges of the images. Power is one of the main parameters that have to be considered while dealing with biomedical instruments. The biomedical signal processing instruments should be capable of operating at low power and also at high speed. In order to segregate the images into different levels or stage, we use convolutional neural networks for classification. By having a hardware architecture for image edge detection, the computational time for pre-processing of the image can be reduced, and the hardware can be a part of acquisition device itself. In this paper a low-power architecture for edge detection to detect the biomedical images are presented. The edge detection output are given to the system, which will diagnose the diseases using image classification using convolutional neural network. In this paper, Sobel and Prewitt, algorithms are used for edge detection using 180 nm technology. The edge detection algorithms are implemented using VLSI, and digital IC design of the architecture is presented. The algorithms for edge detection is co-simulated using MATLAB and Modelsim. The architecture is first simulated using CMOS logic and new method using domino logic is presented for low power consumption.

Gait and tremor investigation using machine learning techniques for the diagnosis of Parkinson disease

Abstract Parkinson’s disease (PD) is a chronic and progressive movement disorder affecting patients in large numbers throughout the world. As PD progresses, the affected person is unable to control movement normally. Individuals affected by Parkinson’s disease exhibit notable symptoms like gait impairments and tremor occurrences during different stages of the disease. In this paper a novel approach has been proposed to diagnose PD using the gait analysis, that consists of the gait cycle, which can be broken down into various phases and periods to determine normative and abnormal gait. Initially, the raw force data obtained from physionet database was filtered using a Chebyshev type II high pass filter with a cut-off frequency 0.8 Hz to remove noises arising from the changes in orientation of the subject’s body and other factors during measurement. The filtered data was used for extracting various gait features using the peak detection and pulse duration measuring techniques. The threshold values of the gait detection algorithm were tuned to individual subjects. From the peak detection algorithm, various kinetic features including the heel and toe forces, and their normalized values were obtained. The pulse duration algorithm was developed to extract different temporal features including the stance and swing phases, and stride time. Tremor is a common symptom in PD. Tremor is an involuntary movement of body parts. At first the tremor may appear in a specific body part like an arm, leg or one side of the body and later it may spread to both sides . This rest tremor is a cardinal sign of PD. An average accuracy of 92.7% is achieved for the diagnosis of PD from gait analysis and tremor analysis is used for knowing the severity of PD.

A Medical Records Managing and Securing Blockchain Based System Supported by a Genetic Algorithm and Discrete Wavelet Transform

Abstract The privacy of patients is jeopardised when medical records and data are spread or shared beyond the protected cloud of institutions. This is because breaches force them to the brink that they start abstaining from full disclosure of their condition. This type of condition has a negative effect on scientific research, patients and all stakeholders. A blockchain-based data sharing system is proposed to tackle this issue, which employs immutability and autonomy properties of the blockchain to sufficiently resolve challenges associated with access control and handle sensitive data. Our proposed system is supported by a Discrete Wavelet Transform to enhance the overall security, and a Genetic Algorithm technique to optimise the queuing optimization technique as well. Introducing this cryptographic key generator enhances the immunity and system access control, which allows verifying users securely in a fast way. This design allows further accountability since all users involved are already known and the blockchain records a log of their actions. Only when the users’ cryptographic keys and identities are confirmed, the system allows requesting data from the shared queuing requests. The achieved execution time per node, confirmation time per node and robust index for block number of 0.19 s, 0.17 s and 20 respectively that based on system evaluation illustrates that our system is robust, efficient, immune and scalable.

An alternative respiratory sounds classification system utilizing artificial neural networks

Background: Computerized lung sound analysis involves recording lung sound via an electronic device, followed by computer analysis and classification based on specific signal characteristics as non-linearity and nonstationarity caused by air turbulence. An automatic analysis is necessary to avoid dependence on expert skills. Methods: This work revolves around exploiting autocorrelation in the feature extraction stage. All process stages were implemented in MATLAB. The classification process was performed comparatively using both artificial neural networks (ANNs) and adaptive neuro-fuzzy inference systems (ANFIS) toolboxes. The methods have been applied to 10 different respiratory sounds for classification. Results: The ANN was superior to the ANFIS system and returned superior performance parameters. Its accuracy, specificity, and sensitivity were 98.6%, 100%, and 97.8%, respectively. The obtained parameters showed superiority to many recent approaches. Conclusions: The promising proposed method is an efficient fast tool for the intended purpose as manifested in the performance parameters, specifically, accuracy, specificity, and sensitivity. Furthermore, it may be added that utilizing the autocorrelation function in the feature extraction in such applications results in enhanced performance and avoids undesired computation complexities compared to other techniques.

Adsorption-Based Model for Single-Wall Carbon Nanotube Response to NO2 Gas

In this work, the response of single-wall carbon nanotube as a resistive NO2 sensor was investigated. A model was developed based on the Freundlich adsorption isotherm and a multi-exponential function was used to describe the relationship between the sensor response and the gas concentration. The model predicts both static and dynamic responses in a closed sampling system. In addition, the model considers the effect of different variables such as the operating temperature, NO2 concentration in air and the two phases (adsorption and desorption) of response. The developed model is in good accordance with the experimental data. This model could be used to design new environmental detection devices and interpret experimental data by providing some insight into the sensor behaviour during the transient phase.

Classification of Normal, Ictal and Inter-ictal EEG via Direct Quadrature and Random Forest Tree

This paper presents an accurate nonlinear classification method that can help physicians diagnose seizure in electroencephalographic (EEG) signal characterized by a disturbance in temporal and spectral content. This is accomplished by applying four steps. First, different EEG signals containing healthy, ictal and seizure-free (inter-ictal) activities are decomposed by empirical mode decomposition method. The instantaneous amplitudes and frequencies of resulted bands (intrinsic mode functions, IMF) are then tracked by the direct quadrature method (DQ). In contrast to other approaches, DQ cancels the effect of amplitude modulation on frequency calculation. The dissociation between instantaneous amplitude and frequency information is therefore fully achieved to avoid features confusion. Afterwards, the Shannon entropy values of both sets of instantaneous values (amplitudes and frequencies)—related to every IMF—are calculated. Finally, the obtained entropy values are classified by random forest tree. The proposed procedure yields 100% accuracy for (healthy)/(ictal) and 98.3–99.7% for (healthy)/(ictal)/(interictal) classification problems. The suggested method is hence robust, accurate, fast, user-friendly, data driven with open access interpretability.

A robust and efficient password-based conditional privacy preserving authentication and group-key agreement protocol for VANETs

Abstract With the rapid growth of information and communication technology (ICT) and Internet of things (IoT), the concept of smart-city is recently introduced by the government of many countries to improve the living environment of urban people. In city areas, the numbers of vehicles are increased exponentially day-by-day. Therefore, it is very difficult to control and manage the city traffic caused by tens of thousands vehicles. The Vehicular Ad-hoc Network (VANET) is used to communicate with the vehicles to give alert for weather conditions, road defects, traffic conditions, etc. and the conditions of the vehicle including location, speed, traffic status, etc. Therefore, the traffic efficiency and safety of the vehicles can be improved with the help of VANET. To serve this purpose, in the literature, many conditional privacy preserving authentication (CPPA) protocols based on CA-PKC (certificate authority-based public key cryptography), and ID-PKC (identity-based public key cryptography) have been put forwarded. In addition, some of these CPPA protocols use elliptic curve or bilinear-pairing for their implementation. The computation cost for bilinear-pairing and elliptic curve is very high compared to the cryptographic general hash function. Therefore, all the earlier protocols suffer from the heavy computational burden and some security weaknesses as well. Therefore, bilinear-pairing-free, robust and efficient CPPA with group-key agreement protocol for VANETs is essential. This paper presents a password-based conditional privacy preserving authentication and group-key generation (PW-CPPA-GKA) protocol for VANETs. Our protocol offers group-key generation, user leaving, user join, and password change facilities. Our protocol is lightweight in terms computation and communication since it can be designed without bilinear-pairing and elliptic curve.