Browsing by Author "Islam, Md Kafiul"

Now showing 1 - 20 of 20

A 13µW 87dB dynamic range implantable ΔΣ modulator for full-spectrum neural recording
(IEEE, 2013-09-26) Jian, Xu; Islam, Md Kafiul; Wang, Shuo; Yang, Zhi
Experiment analysis on in-vivo data sequences suggests a wide system dynamic range (DR) is required to simultaneously record local field potentials (LFPs), extra-cellular spikes, and artifacts/interferences. In this paper, we present a 13 μW 87 dB DR ΔΣ modulator for full-spectrum neural recording. To achieve a wide DR and low power consumption, a fully-differential topology is used with multi-bit (MB) quantization scheme and switched-opamp (SO) technique. By adopting a novel fully-clocked scheme, a power-efficient current-mirror SO is developed with 50% power saving, which doubles the figure-of-merit (FOM) over its counterpart. A new static power-less multi-bit quantizer with 96% power and 69% area reduction is also introduced. Besides, instead of metal-insulator-metal (MIM) capacitor, three high-density MOS capacitor (MOSCAP) structures are employed to reduce circuit area. Measurement results show a peak signal-to-noise and distortion ratio (SNDR) of 85 dB with 10 kHz bandwidth at 1.0 V supply, corresponding to an FOM of 45 fJ/conv.-step. which is implemented in a 0.18 μm CMOS.
A High Performance Delta-Sigma Modulator for Neurosensing
(Sensors, 2015-08-07) Jian, Xu; Zhao, Menglian; Wu, Xiaobo; Islam, Md Kafiul; Yang, Zhi
Recorded neural data are frequently corrupted by large amplitude artifacts that are triggered by a variety of sources, such as subject movements, organ motions, electromagnetic interferences and discharges at the electrode surface. To prevent the system from saturating and the electronics from malfunctioning due to these large artifacts, a wide dynamic range for data acquisition is demanded, which is quite challenging to achieve and would require excessive circuit area and power for implementation. In this paper, we present a high performance Delta-Sigma modulator along with several design techniques and enabling blocks to reduce circuit area and power. The modulator was fabricated in a 0.18-μm CMOS process. Powered by a 1.0-V supply, the chip can achieve an 85-dB peak signal-to-noise-and-distortion ratio (SNDR) and an 87-dB dynamic range when integrated over a 10-kHz bandwidth. The total power consumption of the modulator is 13 μW, which corresponds to a figure-of-merit (FOM) of 45 fJ/conversion step. These competitive circuit specifications make this design a good candidate for building high precision neurosensors.
A Machine Learning Approach for Multi-Level Anxiety Screening among University-Going Students using Wireless EEG Signals
(IEEE, 2025-06-23) Sakib, Nazmus; Islam, Md Kafiul; Faruk, Tasnuva
Anxiety is a widespread mental health condition affecting millions globally, often resulting in significant emotional and physical symptoms. Accurate detection of anxiety levels is essential to provide timely interventions and prevent severe complications. This study explores a machine learning-based approach for multilevel anxiety classification among young adults using EEG signals. The GAD-7 screening tool was used to assess and categorize participants into different anxiety severity groups. EEG data was then recorded, processed, and segmented into 1, 3, and 5 second segments to evaluate the impact of segment duration on classification accuracy. Four channel combinations were tested for comparisons in performance. Feature extraction included eleven time and frequency domain features. The Bagged Trees classifier was applied to classify anxiety levels based on these features. The findings of this work show the potential of EEG-based systems as non-invasive tools for anxiety screening that could support more precise mental health diagnostics.
A Self-Organizing Diffusion Mobile Adaptive Network for Pursuing a Target
(Scientific & Academic Publishing, 2016) Rastegarnia, Amir; Khalili, Azam; Islam, Md Kafiul
In this paper we focus on designing self-organizing diffusion mobile adaptive networks where the individual agents are allowed to move in pursuit of an target. The well-known Adapt-then-Combine (ATC) algorithm is already available in the literature as a useful distributed diffusion-based adaptive learning network. However, in the ATC diffusion algorithm, fixed step sizes are used in the update equations for velocity vectors and location vectors. When the nodes are too far away from the target, such strategies may require large number of iterations to reach the target. To address this issue, we suggest two modifications on the ATC mobile adaptive network to improve its performance. The proposed modifications include (i) distance-based variable step size adjustment at diffusion algorithms to update velocity vectors and location vectors, (ii) to use a selective cooperation, by choosing the best nodes at every iteration, to reduce the number of communications. The performance of the proposed algorithm is evaluated by simulation tests where the obtained results show the superior performance of the proposed algorithm in comparison with the available ATC mobile adaptive network.
A Wavelet-Based Artifact Reduction from Scalp EEG for Epileptic Seizure Detection
(IEEE, 2015-07-15) Islam, Md Kafiul; Rastegarnia, Amir; Yang, Zhi
This paper presents a method to reduce artifacts from scalp EEG recordings to facilitate seizure diagnosis/detection for epilepsy patients. The proposed method is primarily based on stationary wavelet transform and takes the spectral band of seizure activities (i.e., 0.5-29 Hz) into account to separate artifacts from seizures. Different artifact templates have been simulated to mimic the most commonly appeared artifacts in real EEG recordings. The algorithm is applied on three sets of synthesized data including fully simulated, semi-simulated, and real data to evaluate both the artifact removal performance and seizure detection performance. The EEG features responsible for the detection of seizures from nonseizure epochs have been found to be easily distinguishable after artifacts are removed, and consequently, the false alarms in seizure detection are reduced. Results from an extensive experiment with these datasets prove the efficacy of the proposed algorithm, which makes it possible to use it for artifact removal in epilepsy diagnosis as well as other applications regarding neuroscience studies.
Artifact Characterization and Removal for In-Vivo Neural Recording
(Elsevier, 2014-04-15) Islam, Md Kafiul; Rastegarnia, Amir; Nguyen, Anh Tuan; Yang, Zhi
Background: In vivo neural recordings are often corrupted by different artifacts, especially in a less-constrained recording environment. Due to limited understanding of the artifacts appeared in the in vivo neural data, it is more challenging to identify artifacts from neural signal components compared with other applications. The objective of this work is to analyze artifact characteristics and to develop an algorithm for automatic artifact detection and removal without distorting the signals of interest. New method: The proposed algorithm for artifact detection and removal is based on the stationary wavelet transform with selected frequency bands of neural signals. The selection of frequency bands is based on the spectrum characteristics of in vivo neural data. Further, to make the proposed algorithm robust under different recording conditions, a modified universal-threshold value is proposed. Results: Extensive simulations have been performed to evaluate the performance of the proposed algorithm in terms of both amount of artifact removal and amount of distortion to neural signals. The quantitative results reveal that the algorithm is quite robust for different artifact types and artifact-to-signal ratio. Comparison with existing methods: Both real and synthesized data have been used for testing the proposed algorithm in comparison with other artifact removal algorithms (e.g. ICA, wICA, wCCA, EMD-ICA, and EMD-CCA) found in the literature. Comparative testing results suggest that the proposed algorithm performs better than the available algorithms. Conclusion: Our work is expected to be useful for future research on in vivo neural signal processing and eventually to develop a real-time neural interface for advanced neuroscience and behavioral experiments
Artifact Characterization, Detection and Removal from Neural Signals
(National University of Singapore, 2015-12) Islam, Md Kafiul; Zhi Yang
Artifact detection and removal are important preprocessing steps for neural recordings to decode the neural signals properly and currently an active research problem. In this thesis, for the first time, artifacts found in the in-vivo neural recordings are studied and consequently a systematic artifact characterization is presented. Subsequently, three different artifact removal methods are proposed (one for in-vivo neural signals in general application and two for EEG data for two different application purposes, i.e. seizure detection and brain-computer interface (BCI) experiments). In order to evaluate the proposed methods quantitatively in comparison with available state-of-the-art methods, different artifact types have been extracted from real recordings for simulating artifact templates and hence to build a synthesized neural database on which the methods are applied. In addition, the effects of artifact removal on the later-stage processing have also been evaluated resulting significant performance improvement which proofs the efficacy of the proposed methods.
Design and Implementation of an EOG-based Mouse Cursor Control for Application in Human-Computer Interaction
(IOP, 2020-03-01) Kabir, Ahsan-ul; Bin Shahin, Faisal; Islam, Md Kafiul
Human Computer Interaction (HCI) has turned into an emerging technology due to the advancement in the field artificial intelligence and biomedical engineering. Acquiring different bio-signals such as Electro-oculography (EOG), Electromyography (EMG) and Electroencephalography (EEG) to control external machine or computer is the essence of HCI technology. In this research, we attempt to extract the EOG signal from different ways of eye movements and process it for HCI application. By utilizing Arduino, EOG data can be transmitted to computer and those signal characteristics is analysed through MATLAB. We have designed and implemented hardware and interfaced it with software to control a computer mouse cursor only by eye movement. Certain classification module like Support Vector machine (SVM) and Multilayer Perceptron (MLP) are used to classify different EOG data generated from different eye movement. According to the eye position, cursor automatically moves in that specific direction and PyAutoGUI module is used for this task. Results after experimentations with different subjects to control mouse cursor in real-time show that the average classification accuracy can reach up to 93% across all directions.
Editorial: Recent advances in EEG (non-invasive) based BCI applications
(Frontiers, 2023-03-02) Rastegarnia, Amir; Islam, Md Kafiul
EEG-based Mouse Cursor Control using Motor Imagery Brain-Computer Interface
(IEEE, 2024-05-03) Roja, Saima Tasfia; Bin Rafique, Sayem; Rhaman, Md. Asikur; Sakib, Nazmus; Islam, Md Kafiul
Brain-computer interface (BCI) is a system that collects, analyzes, and transforms brain signals into commands. The brain experiences repetitive, oscillatory electrical changes caused by these activities that have a very low voltage of only a few microvolts (µV).The term electroencephalogram (EEG) refers to the non-invasive recording of this electrical activity from the scalp. The signals are then analyzed in a computer to identify the desired action after signal acquisition. Relevant features are gathered and translated into commands that operate an output device or carry out the command. The user is subsequently provided with feedback to confirm that the command has been carried out correctly This work focuses on the development and implementation of a Mouse Cursor Control system using Motor Imagery (MI) BCI, using data recorded with the Emotiv EPOC+ headset and processed using our algorithm in the MATLAB software. While similar works do exist, most tend to focus on one or more aspects of data processing such as classification. We acquired the data from subjects ourselves, and after processing the data using our algorithm, the system was implemented, and the cursor was moved. This makes our system a semi-online system, as opposed to offline systems. The only limitation of our system is that the system is implemented in semi-real time. Furthermore, accuracy was tested for different frequency bands and the highest accuracy of 93.60% was achieved using the offline dataset.
EEG-Based Preference Classification for Neuromarketing Application
(Hindawi, 2023-03-01) Sourov, Injamamul Haque; Alvi Ahmed, Faiyaz; Opu, Md. Tawhid Islam; Mutasim, Aunnoy K.; Bashar, M. Raihanul; Sardar Tipu, Rayhan; Amin, Md. Ashraful; Islam, Md Kafiul
Neuromarketing is a modern marketing research technique whereby consumers’ behavior is analyzed using neuroscientific approaches. In this work, an EEG database of consumers’ responses to image advertisements was created, processed, and studied with the goal of building predictive models that can classify the consumers’ preference based on their EEG data. Several types of analysis were performed using three classifier algorithms, namely, SVM, KNN, and NN pattern recognition. The maximum accuracy and sensitivity values are reported to be 75.7% and 95.8%, respectively, for the female subjects and the KNN classifier. In addition, the frontal region electrodes yielded the best selective channel performance. Finally, conforming to the obtained results, the KNN classifier is deemed best for preference classification problems. The newly created dataset and the results derived from it will help research communities conduct further studies in neuromarketing.
Effect of artifact removal on EEG based motor imagery BCI applications
(SPIE Digital Library, 2024-01-29) Islam, Md Kafiul; Sakib, Nazmus; Anjum, Maisha
Brain computer interface (BCI) is an emerging technology where the user can establish direct communication between the electrical device and himself without any physical exertion. The EEG signal is a noninvasive and low-cost method to extract brain signal from subject. The EEG signal contains different types of information including motor sensory information originating from the motor cortex region of the brain. Research and study have shown that motor cortex generates signals similar to the signals generated during deliberate limb movements. Therefore, motor imagery (MI) signals if extracted can be utilized to operate any electrical device establishing a BCI system. However, the EEG data can contain lots of artifacts. This degrades the signal quality and also cause false positive command to the connected device. Therefore, it is crucial to remove the artifacts from the EEG signal before classification. In this project, EEG data has been collected from 12 subjects who are instructed to perform MI activity. The EEG signal is then processed and an efficient artifact removal technique has been applied. The artifact removal method applies wavelet transform theorem and artifactual probability mapping method to detect artifactual epochs and eliminate it from the signal. Useful features are then extracted from the signal and artificial neural network (ANN) classifier is applied to it. The classification accuracy has been enhanced by 15-16% on average after removal of artifacts from the EEG recordings for MI-BCI experiments. Afterwards, performance evaluation such as finding signal to noise ratio has been done to evaluate the improvement in the signal after noise removal.
Emotion classification using single-channel scalp-EEG recording
(IEEE, 2016-10-18) Jalilifard, Amir; Pizzolato, Ednaldo Brigante; Islam, Md Kafiul
Several studies have found evidence for corticolimbic Theta electroencephalographic (EEG) oscillation in the neural processing of visual stimuli perceived as fear or threatening scene. Recent studies showed that neural oscillations' patterns in Theta, Alpha, Beta and Gamma sub-bands play a main role in brain's emotional processing. The main goal of this study is to classify two different emotional states by means of EEG data recorded through a single-electrode EEG headset. Nineteen young subjects participated in an EEG experiment while watching a video clip that evoked three emotional states: neutral, relaxation and scary. Following each video clip, participants were asked to report on their subjective affect by giving a score between 0 to 10. First, recorded EEG data were preprocessed by stationary wavelet transform (SWT) based denoising to remove artifacts. Afterward, the distribution of power in time-frequency space was obtained using short-time Fourier transform (STFT) and then, the mean value of energy was calculated for each EEG sub-band. Finally, 46 features, as the mean energy of frequency bands between 4 and 50 Hz, containing 689 instances — for each subject —were collected in order to classify the emotional states. Our experimental results show that EEG dynamics induced by horror and relaxing movies can be classified with average classification rate of 92% using support vector machine (SVM) classifier. We also compared the performance of SVM to K-nearest neighbors (K-NN). The results show that K-NN achieves a better classification rate by 94% accuracy. The findings of this work are expected to pave the way to a new horizon in neuroscience by proving the point that only single-channel EEG data carry enough information for emotion classification.
Frequency Estimation of Unbalanced Three-Phase Power Systems Using the Modified Adaptive Filtering
(Scientific & Academic Publishing, 2015-08) Rastegarnia, Amir; Khalili, Azam; Vahidpour, Vahid; Islam, Md Kafiul
In this paper, the problem of frequency estimation using adaptive filters is addressed based on the augmented complex normalized least mean squares (ACNLMS) technique. In other words, motivated from ACLMS technique, a new version called ACNLMS that outperforms the previous techniques is proposed. The proposed method makes use of two weight coefficients, like ACLMS, together with normalized time variant step size, thus its promising in increasing and enhancing both convergence rate and accuracy. Then, using derived weight coefficient, the frequency is estimated at each step. The performance and convergence analysis of the proposed method along with simulation results comparing with two existing techniques, CLMS and ACLMS, are provided in critical cases such as various unbalanced conditions and presence of harmonic distortion. Simulation results indicate that the proposed technique achieves a better performance in terms of convergence rate and frequency estimation accuracy as compared with CLMS and ACLMS techniques. Moreover, ACLMS achieves a smaller error variance than the other mentioned cases.
Intelligent fuzzy system for automatic artifact detection and removal from EEG signals
(Elsevier, 2022-10-05) Agounad, Said; Hamou, Soukaina; Tarahi, Ousama; Moufassih, Mustapha; Islam, Md Kafiul
The EEG signals were used in many medical and technological applications such as diagnosis of diseases, rehabilitation of disabled peoples, preventive healthcare, BCI (brain computer interface) systems. EEG signal is prone to the physiological and non-physiological artifacts which severely affect them and lead to its misinterpretation. An automatic method and/or algorithm; for handling EEG artifacts; is proposed. The proposed method is based on three statistical parameters (entropy, kurtosis and skewness), fuzzy inference system (FIS) and stationary wavelet transform (SWT). Each incoming EEG epoch is described using these three statistical parameters. Based on the extracted statistical parameters, the designed FIS decides if an epoch is artifactual or not. Then SWT is used to decompose the EEG epoch into detail and approximation coefficients. To reduce the effect of artifact removal, we propose to use other fuzzy inference systems, which allow to select the contaminated wavelet coefficients. The universal thresholding method is then applied to the corrupted coefficients. Finally, the inverse SWT applies to the thresholded and non-corrupted coefficients to restore the cleaned EEG signal. The performance of the proposed method in terms of amount of artifact removal and signal distortion is evaluated in three scenarios: fully simulated, semi-simulated, and real artifactual EEG data. The comparison of our method with some existing state-of-the-art methods shows the superiority of our method over others in terms of performance and computational time.
Machine Learning Model for Computer-Aided Depression Screening among Young Adults Using Wireless EEG Headset
(Hindawi, 2023-05-31) Islam, Md Kafiul; Faruk, Tasnuva; Sakib, Nazmus
Depression is a disorder that if not treated can hamper the quality of life. EEG has shown great promise in detecting depressed individuals from depression control individuals. It overcomes the limitations of traditional questionnaire-based methods. In this study, a machine learning-based method for detecting depression among young adults using EEG data recorded by the wireless headset is proposed. For this reason, EEG data has been recorded using an Emotiv Epoc+ headset. A total of 32 young adults participated and the PHQ9 screening tool was used to identify depressed participants. Features such as skewness, kurtosis, variance, Hjorth parameters, Shannon entropy, and Log energy entropy from 1 to 5 sec data filtered at different band frequencies were applied to KNN and SVM classifiers with different kernels. At AB band (8–30 Hz) frequency, 98.43 ± 0.15% accuracy was achieved by extracting Hjorth parameters, Shannon entropy, and Log energy entropy from 5 sec samples with a 5-fold CV using a KNN classifier. And with the same features and classifier overall accuracy = 98.10 ± 0.11, NPV = 0.977, precision = 0.984, sensitivity = 0.984, specificity = 0.976, and F1 score = 0.984 was achieved after splitting the data to 70/30 ratio for training and testing with 5-fold CV. From the findings, it can be concluded that EEG data from an Emotiv headset can be used to detect depression with the proposed method.
Methods for Artifact Detection and Removal from Scalp EEG: A Review
(Elsevier, 2016-10-15) Islam, Md Kafiul; Rastegarnia, Amir; Yang, Zhi
Electroencephalography (EEG) is the most popular brain activity recording technique used in wide range of applications. One of the commonly faced problems in EEG recordings is the presence of artifacts that come from sources other than brain and contaminate the acquired signals significantly. Therefore, much research over the past 15 years has focused on identifying ways for handling such artifacts in the preprocessing stage. However, this is still an active area of research as no single existing artifact detection/removal method is complete or universal. This article presents an extensive review of the existing state-of-the-art artifact detection and removal methods from scalp EEG for all potential EEG-based applications and analyses the pros and cons of each method. First, a general overview of the different artifact types that are found in scalp EEG and their effect on particular applications are presented. In addition, the methods are compared based on their ability to remove certain types of artifacts and their suitability in relevant applications (only functional comparison is provided not performance evaluation of methods). Finally, the future direction and expected challenges of current research is discussed. Therefore, this review is expected to be helpful for interested researchers who will develop and/or apply artifact handling algorithm/technique in future for their applications as well as for those willing to improve the existing algorithms or propose a new solution in this particular area of research.
Probability Mapping Based Artifact Detection and Wavelet Denoising based Artifact Removal from Scalp EEG for BCI Applications
(IEEE, 2019-02-25) Islam, Md Kafiul; Rastegarnia, Amir
In EEG-based Brain-Computer Interface (BCI) applications, the EEG recording is often contaminated by different types of artifacts that can misinterpret the BCI output. Automatic detection and removal of such offending artifacts from EEG for online processing pose a great challenge. In this paper, we present a novel method that can map the artifact probability of an EEG epoch based on four statistical measures: entropy, kurtosis, skewness and Periodic Waveform Index PWI). Then a removal method is adopted based on stationary wavelet transform that can be applied to the epochs by setting a particular probability threshold from the user. This epoch by epoch preprocessing would allow the user to tune the threshold parameters after some initial training with the same EEG recordings and eventually can be applied to both offline and online processing. Experimental results with both simulated and real EEG data prove the efficacy of the method that it can reliably trace the artifactual epoch with reasonable accuracy and eventually reduces the artifacts from EEG with very little distortion to the signal of interest. Further testing with EEG datasets for BCI experiments also shows that artifact removal can significantly enhance the BCI performance in both motor-imagery (MI) and event related potential (ERP) based BCI applications.
Steady-State Tracking Analysis of Adaptive Filter With Maximum Correntropy Criterion
(Springer, 2016-08-03) Khalili, Azam; Rastegarnia, Amir; Islam, Md Kafiul; Rezaii, Tohid Yousefi
This letter studies the tracking performance of a stochastic gradient-based adaptive algorithm, namely the maximum correntropy criterion algorithm, where a random walk is used to model the non-stationarity. In our analysis, we use the energy conservation argument to derive expressions for the steady-state excess mean square error (EMSE). We consider two different cases for measurement of noise distribution including the Gaussian noise and general non-Gaussian noise. For the Gaussian case, we derive a fixed-point equation that can be solved numerically to find steady-state EMSE value. For the general non-Gaussian case, we derive an approximate closed-form expression for EMSE. For both cases, unlike the stationary environment, the EMSE curves are not increasing functions of step size parameter. We use this observation to find the optimum step size learning parameter for general non-Gaussian case. The validity of the theoretical results are justified via simulation results.
The Journey of Elastography: Background, Current Status and Future Possibilities in Breast Cancer Diagnosis
(Elsevier, 2015) Faruk, Tasnuva; Islam, Md Kafiul; Arefin, Sams; Haq, Md Zahurul
Elastography is a promising way to assess tissue differences regarding stiffness or elasticity for what was historically assessed manually by palpation. Combined with conventional imaging modalities (eg, ultrasonography [US]), elastography can potentially evaluate the stiffness of a breast lesion and consequently help to detect malignant breast tumor from benign ones. Recent studies show that ultrasonographic elastography (USE) provides higher image quality compared with conventional B-mode US or mammography during breast cancer diagnosis, which eventually helps to reduce false-positive results (ie, increased specificity) and therefore is useful in avoiding breast biopsy. This article reviews the basics of elastography technique, classifications, diagnosis results obtained from clinical studies to date for differentiating malignant breast tumors from benign lesions, and its future possibilities. In addition, this article generalizes different elastography methods, modes, and associated imaging modalities in a simpler way and attempts to identify misconceptions and confusion related to existing elastography techniques. It also makes an effort to identify the gaps of information that need to be filled so that interested researchers can get an overall idea of elastography-based methods in a convenient way to carry out their research on breast elastography for prospective future applications, e.g. breast cancer diagnosis or even in intraoperative breast tumor localization.