Garima Chandel

Bio: Garima Chandel is an academic researcher from Aligarh Muslim University. The author has contributed to research in topics: Wavelet & Electroencephalography. The author has an hindex of 3, co-authored 7 publications receiving 42 citations.

Detection of Seizure Event and Its Onset/Offset Using Orthonormal Triadic Wavelet Based Features

Abstract: IRBM - In Press.Proof corrected by the author Available online since vendredi 28 decembre 2018

A simplified method for classification of epileptic EEG signals

Abstract: Epileptic seizures occur randomly and are difficult to identify in Electroencephalogram (EEG) recording with multiple channels. Most researchers have used large dimension features, complex transformation techniques and non-linear classifier. A new algorithm based on Mean Absolute Deviation (MAD) using lower feature vector dimension and linear classifier is proposed for automatic seizure detection using EEG signals. The proposed method calculates MAD of each channel on frame consisting of 256 samples. In order to reduce the dimension of the feature, mean and maximum value of the MAD for all channels were selected as discriminating parameters. The proposed algorithm is tested on a publicly available Bonn University EEG database for three cases. The accuracy of the algorithm was 100% in all the considered problems. The proposed work outperforms in terms of complexity with respect to the other available state-of-the-art method on the same database. It results in reduction of number of features per frame with less complexity for all considered problems.

Study on circular planar antenna

Abstract: In this paper, the microstrip antenna consist of a planar resonant radiating element parallel to, but separated, from a ground plane by a thin dielectric substrate (t <<λ) was designed and simulated on EM software. This antenna is fabricated on RT Duroid substrate of dielectric 2.2, thickness 0.8 mm. This antenna was feed by microstrip transformer. Linearly polarized circular shape antenna was designed at centre frequency 10 GHz. The E & H — plane radiation patterns were shown

EMD Analysis of EEG Signals for Seizure Detection

Abstract: The Electroencephalogram (EEG) is the electrical signals which contain the information related to activities within the brain. Neurological disorders such as epilepsy can be diagnosed effectively by analyzing EEG signals. In the present work, the empirical mode decomposition (EMD) is applied to EEG recordings for the automated detection of seizures in epileptic patients. For this purpose, intrinsic mode functions (IMFs) from the EMD are processed to extract the features from normal and seizure EEG signals. The extracted features are ranked to select the useful features for classification. The classification was done by using these selected features by Artificial Neural Network (ANN). The EEG dataset used in the present study is the well-known publicly available Bonn University EEG dataset. Three different classification problems were done by using the sets of this data. The classification accuracy achieved for these three cases were 96.1, 96.4, and 99.3%.

Seizure Onset Detection by Analyzing Long-Duration EEG Signals

Abstract: Seizures in epileptic patients affect tremendously their daily life in terms of accidents during driving a vehicle, swimming, using stairs, etc. Automatic seizure detectors are used to detect seizure as early as possible so that an alarm can be given to patient or their family for using anti-epileptic drugs (AEDs). In this paper, an algorithm has been proposed for automatic seizure onset detection by analysis of electroencephalogram (EEG) signals. The method is based on few wavelet transform-based features and two statistical features without wavelet decomposition for improving the performance of detector. The mean, energy, and entropy were calculated on different wavelet decomposed subbands, and mean absolute deviation and interquartile range were calculated on raw signal. Classification between seizure and nonseizure types of EEG signals was done successfully by linear classifier. The algorithm was applied to CHB-MIT EEG dataset for seizure onset detection and achieved 100 % sensitivity with mean latency of 1.9 s.

1D-Local Binary Pattern Based Feature Extraction for Classification of Epileptic EEG Signals, Applied Mathematics and Computation243 (2014): 209-219.

Abstract: In this paper, an effective approach for the feature extraction of raw Electroencephalogram (EEG) signals by means of one-dimensional local binary pattern (1D-LBP) was presented. For the importance of making the right decision, the proposed method was performed to be able to get better features of the EEG signals. The proposed method was consisted of two stages: feature extraction by 1D-LBP and classification by classifier algorithms with features extracted. On the classification stage, the several machine learning methods were employed to uniform and non-uniform 1D-LBP features. The proposed method was also compared with other existing techniques in the literature to find out benchmark for an epileptic data set. The implementation results showed that the proposed technique could acquire high accuracy in classification of epileptic EEG signals. Also, the present paper is an attempt to develop a general-purpose feature extraction scheme, which can be utilized to extract features from different categories of EEG signals.

A comprehensive comparison of handcrafted features and convolutional autoencoders for epileptic seizures detection in EEG signals

Abstract: Epilepsy, a brain disease generally associated with seizures, has tremendous effects on people’s quality of life. Diagnosis of epileptic seizures is commonly performed on electroencephalography (EEG) signals, and by using computer-aided diagnosis systems (CADS), neurologists can diagnose epileptic seizure stages more accurately. In these systems, a mandatory stage is feature extraction, performed by handcrafting features or learning them, ordinarily by a deep neural net. While researches in this field commonly show the value of a group of limited features, yet an accurate comparison between different suggested features is essential. In this article, first, a comparison between the importance of 50 different handcrafted features for seizure detection is presented. Additionally, the computational complexity of features is investigated as well. Then the best features based on Fisher scores are picked to classify signals on a benchmark dataset for evaluation. Additionally, a convolutional autoencoder with five layers is applied to learn features in order to have a complete comparison among feature extraction approaches. Finally, a hybrid method is employed, which combines handcrafted features and encoding of autoencoder to reach high performance in seizure detection in EEG signals.

A hybrid Local Binary Pattern and wavelets based approach for EEG classification for diagnosing epilepsy

Abstract: Epilepsy is one of the grave neurological ailments affecting approximately 70 million people globally. Detection of epileptic attack is commonly carried out by viewing and analysing long-duration multi-channel EEG records. To counter this time-consuming process, a hybrid Local Binary Pattern-Wavelet based approach, classifying EEG in epileptic patients, is adopted in this research. Epilepsy is characterized by multiple ictal patterns in the form of synchronous epileptiform discharge transients. This work attempts to classify seizure from normal EEG recordings using the low-frequency activity. In order to perform this classification, the EEG signal is filtered and then transformed using Local Binary Pattern (LBP) into a new signal. Discrete Wavelet Transform (DWT) is employed to decompose the obtained signal. Wavelet coefficients are calculated to 5 levels of decomposition. A combination of univariate and bivariate features forms the feature set for seizure detection. This feature set extracted from low-frequency band coefficients helps in bringing out the dispersion, symmetry, and peakedness present in the EEG signal. A novel LBP based Spatio-temporal analysis of the continuous EEG signal for epilepsy detection is carried out on 105 seizures from 14 randomly selected subjects of CHB-MIT EEG database. A sensitivity of 100% is achieved on the CHB-MIT database while long term EEG is being tested with Linear Discriminant Analysis (LDA) classifier. The algorithm works well to obtain a false detection rate (FP/Hour) of 0.59. The specificity of 99.8% is attained with a mean accuracy of 99.6% when tested on 498.9 h of EEG data.

Identification of Epileptic EEG Signals Using Convolutional Neural Networks

Abstract: Epilepsy is one of the chronic neurological disorders that is characterized by a sudden burst of excess electricity in the brain. This abnormality appears as a seizure, the detection of which is an important research topic. An important tool used to study brain activity features, neurological disorders and particularly epileptic seizures, is known as electroencephalography (EEG). The visual inspection of epileptic abnormalities in EEG signals by neurologists is time-consuming. Different scientific approaches have been used to accurately detect epileptic seizures from EEG signals, and most of those approaches have obtained good performance. In this study, deep learning based on convolutional neural networks (CNN) was considered to increase the performance of the identification system of epileptic seizures. We applied a cross-validation technique in the design phase of the system. For efficiency, comparative results between other machine-learning approaches and deep CNNs have been obtained. The experiments were performed using standard datasets. The results obtained indicate the efficiency of using CNN in the detection of epilepsy.