About Me |
---|
I have completed my Ph. D. in the area of Biomedical Signal processing from the International Institute of Information Technology, Bhubaneswar. I have joined Centurion University of Management and Technology in December 2020. Apart from teaching, I am interested in novel research techniques that could bring a change and motivate people around us. I am currently working in the area of biomedical image processing using deep learning techniques. |
Thesis: Development of efficient algorithms for artifact removal from single-channel EEG.
Electroencephalogram (EEG) serves as an acknowledged tool to study and analyze the neuronal functions and neurophysiological activity of the human brain. While recording, the EEG may get contaminated from non-physiological artifacts and physiological artifacts (electrocardiogram (ECG), electrooculogram (EOG), and electromyogram (EMG). This dissertation is focused on developing algorithms for pre-processing of the single-channel EEG signals for artifact detection and removal in the absence of reference artifact channel/s. Also, the algorithms ensure introducing the least distortion
to the spectral bandwidths of EEG where artifact spectral content is not present dominantly. Algorithms are proposed for non-physiological artifacts like Flatline (FL), baseline wandering (BLW), motion artifacts (MA), power line interference (PLI), and abrupt slopes (AS) that are likely to occur in the longterm EEG recordings, need to be detected for quality enhancement of the acquired EEG. For ECG artifact removal two algorithms based on modified S-transform (MST) and modified variational mode decomposition (mVMD) are proposed in the absence of a reference ECG channel. Further, the algorithm for the ocular artifacts (OAs) like blink, lateral eye movement (LEM),
utter removal from single-channel EEG in the absence of coherent channels is proposed using VMD followed by regression. Lastly, muscle artifact removal from single-channel EEG using singular spectrum analysis (SSA) is proposed which uses a mobility parameter to identify the artifact and a trained neural network regressor (NNR) for different contamination levels.
The thesis is compiled from 5 research articles and 5 conference articles.
Sl. No. | Title | Issuer |
---|
Electroencephalograms (EEGs) signal, obtained by recording the
brain waves are used to analyse health problems related to neurology and clinical neurophysiology. This signal is often contaminated by a range of physiological and non-physiological artefacts, which leads to a misinterpretation in EEG signal analysis. Hence, artefact removal is one of the pre-processing step required for clinical usefulness of the EEG signal. One of the physiological artefact, i.e., electrocardiogram (ECG) contaminated EEG can
affect the clinical analysis and diagnosis of brain health in various ways. This paper presents a review of engineering approaches adopted till date for ECG artefact identification and removal from contaminated EEG signal. In addition, the technical approach, computational extensiveness, input requirement and the results achieved with every method is discussed. Along with that, the feasibility study for real-time implementation of the algorithms is discussed. Also, an analysis of these methods has been reported based on their performance.
The electroencephalogram (EEG) electrodes are susceptible to electrocardiogram (ECG) artifacts, misleading the analysis and diagnosis from the EEG recording. This paper mainly focuses on the detection and correction of contaminating ECG artifact of various strengths, from a single channel EEG in the absence of a reference ECG channel. The algorithm involves two stages i.e. detection of ECG artifacts and correction of those artifacts. For detection of ECG artifact, modified S-transform (MST) is used on the bandpass filtered contaminated EEG to localize the higher energy QRS segments in time scale. For improved energy concentration of MST around the instantaneous frequency, artifact detection proximity in time scaleis.05s. To correct ECG artifacts from EEG, a modified ensembled average subtraction is proposed, which restricts the overcompensation of the EEG in the correction process. The proposed algorithm is tested on MIT/BIH Polysomnography dataset and synthetic data set with various contamination strengths, together more than 45 hours of EEG data. The proposed method achieves mean positive predictive value (PPV) and failed detection rate (FDR) of 97.87% and 2.13% respectively for MIT/BIH Polysomnography EEG record-ings. The mean PPV and FDR for the CAP sleep EEG recordings with various contamination levels of ECG are found to be 98.77% and 1.22% respectively. The proposed algorithm is compared against existing Continuous wavelet transform (CWT) and Empirical ensembled mode decomposition (EEMD) based algorithms where the proposed algorithm found to be performing better in terms of the spike to energy ratio(SER), spike to background energy ratio (SBR), correlation factors, Mutual information (MI), Mean absolute error (MAE) and Mean square error (MSE) along with a lower change in power spectral density (PSD)in various brain frequency bands.
The Electroencephalogram (EEG) recordings from the frontal lobe of the human brain help in analyzing several important brain functions like motor functions, problem-solving skills, etc. or brain disorders. These recordings are often contaminated by high amplitude and long duration ocular artifacts (OAs) like eye blinks,
flutters, and lateral eye movements (LEMs), hence corrupting a considerable segment of EEG. In this study, an enhanced version of signal decomposition scheme i.e. Variational Mode Decomposition (VMD) based algorithm is used for suppression of OAs. The signal decomposition is preceded by the identification of ocular artifact corrupted segment using Multiscale modified sample entropy (mMSE). The band-limited intrinsic mode functions (BLIMFs) are obtained using predefined K (number of required BLIMFs) and a (balancing parameter). These parameters help to detrend the EEG
segment in yielding the low frequency and high amplitude BLIMFs related to OA efficiently.
Upon identifying OA components from the BLIMFs and estimating OA, it is regressed with
the contaminated EEG to obtain the clean EEG. The proposed VMD based algorithm provides
an improved performance in comparison with the existing single channel algorithms based
on Empirical mode decomposition (EMD) and Ensembled EMD (EEMD) and multi-channel
algorithms like Independent component analysis (ICA) and wavelet enhanced ICA for
artifact suppression and is also able to overcome their limitations. The significance of
the algorithm are: (1) no additional reference EOG channel requirement, (2) OA artifact
based thresholds for identification and estimation from the mode functions obtained using
VMD, and (3) also address the
flutter artifacts.