Biomedical Signal Processing / Medical Signal Processing / Biosignal Processing
Safa Rafieivand; Mohammad Hasan Moradi; Hosein Asl Soleimani; Zahra Momayez Sanat
Volume 17, Issue 2 , September 2023, , Pages 120-130
Abstract
Esophageal mobility disorders are a type of digestive system problem characterized by abnormal bolus movement in the esophagus. The standard diagnostic method for these kinds of disorders is High-Resolution Manometry (HRM). Despite the availability of guidelines like “Chicago” for the analysis ...
Read More
Esophageal mobility disorders are a type of digestive system problem characterized by abnormal bolus movement in the esophagus. The standard diagnostic method for these kinds of disorders is High-Resolution Manometry (HRM). Despite the availability of guidelines like “Chicago” for the analysis of HRM results, diagnosis is still a challenging task that relies on the physician's skills or requires additional assessment modalities. Additionally, it is typical for esophageal mobility disorders to co-occur in one person, leading to a more complex situation for problem identification.The current study focuses on cases who suffering from more than one disorder simultaneously. Then the problem of disorder identification can be interpreted as a multi-label classification problem. Consequently, the fuzzy classifier architecture that was previously introduced for automatic single-disorder diagnosis by the authors is modified. The presented classifier in this paper not only learns the input space from the samples but also utilizes the co-morbidity of disorders to enhance the prediction results. The outcomes show that adding this information to the learning procedure of the base classifier enhances its performance and generates a new fuzzy classifier that overcomes other multi-label classifiers. The presented method is able to differentiate esophageal mobility disorders with an average Hamming loss of 0.18 ± 0.08 which is better than other competitor methods.
Biomedical Signal Processing / Medical Signal Processing / Biosignal Processing
Marziyeh Ghanavaty; Seyyedeh Fatemeh Molaeezadeh; Mojtaba Navidi
Volume 17, Issue 2 , September 2023, , Pages 150-160
Abstract
Hypertension is the leading cause of death worldwide. Continuous blood pressure (BP) measurement is crucial for the elderly and people with myocardial infarction, cardiovascular disease, kidney disease and gestational hypertension. Cuff-based blood pressure Holters are the most common method for continuous ...
Read More
Hypertension is the leading cause of death worldwide. Continuous blood pressure (BP) measurement is crucial for the elderly and people with myocardial infarction, cardiovascular disease, kidney disease and gestational hypertension. Cuff-based blood pressure Holters are the most common method for continuous blood pressure measurement, but due to the use of an inflatable cuff, they often cause discomfort, particularly during sleep. A solution to such problems is the optical measurement of blood pressure using the photoplethysmogram (PPG) signal. This paper introduces a transfer deep learning framework for estimating systolic BP (SBP) and diastolic BP (DBP) using a single PPG signal. The proposed framework consists of three main parts: 1) downsampling by a factor of 4 aimed at reducing model complexity, 2) designing a pre-trained model including CNN and BiLSTM layers, and 3) personalizing the pre-trained model for each patient through transfer learning. We carry out Bland-Altman and correlation analysis to compare our method to the invasive arterial catheter (the gold-standard BP measurement method). Our model was validated on a wide range of BP signals acquired from 100 patients in MIMIC-III database. Results showed that the error and Pearson correlation coefficient of our model are 0.14±7.38 mmHg (mean ± standard deviation) and 0.95 for SBP, and 0.00±4.67 mmHg and 0.92 for DBP. The proposed method satisfies the requirements the AAMI and IEEE-1708a standard and receives a grade A according to the BHS standard. This research has shed light on long-term BP monitoring and the prevention of cardiovascular events.
Biomedical Signal Processing / Medical Signal Processing / Biosignal Processing
Javad Delavar Matanaq; Hamed Danandeh Hesar; Mohammad Hadi Ahmadi fam
Volume 17, Issue 1 , May 2023, , Pages 11-20
Abstract
In recent years, model-based ECG processing algorithms have been successfully developed in various fileds of ECG processing. The calculation of ECG dynamic model (EDM) is a crucial step for these methods. The EDM parameters can be calculated using optimization algorithms. One of the popular optimization ...
Read More
In recent years, model-based ECG processing algorithms have been successfully developed in various fileds of ECG processing. The calculation of ECG dynamic model (EDM) is a crucial step for these methods. The EDM parameters can be calculated using optimization algorithms. One of the popular optimization methods in this field is an offline nonlinear method in which users have to manually select points on ECG signal in order to calculate EDM parameters. The objective function used in this algorithm is a complex function which is hard to optimize. In this paper an automatic optimization algorithm is proposed which uses meta-heuristic optimization algorithms to calculate EDM parameters. In this algorithm, we don’t need to select points manually. In addition, the objective function in this algorithm is broken in to several simple objective functions which makes the optimization more accurate. Meta-heuristic optimization algorithms may perform successfully on some optimization problems while failing on others. As a result, a specific algorithm cannot be considered the best optimizer for all optimization problems. For this reason, in this paper, the performances of nine popular meta-heuristic algorithms such as particle swarm optimization, artificial bee colony, cucko search, etc are investigated. In this paper, 200 ECG segments from different records of the MIT-BIH Normal Sinus Rhythm Database (NSRDB) have been selected for evaluation. The duration of each segment was 30 seconds. The EDM parameters for each segment were calculated using the aforemetinoned optimization algorithms. For evaluation, the similarities between the original signals and the synthetic ECG signals were inspected for each optimization algorithm. These synthetic signals were created using the calculated EDM parameters. The similarity results showed that the water evaporation optimization (WEO), teaching learning-based optimization (TLBO), and cucko’s search (CS) algorithms achived better results compared with other methods.
Biomedical Signal Processing / Medical Signal Processing / Biosignal Processing
davoud saadati; Sattar Mirzakuchaki
Volume 16, Issue 4 , March 2023, , Pages 61-70
Abstract
Analysis and examination of sound of organs can be utilized in order to diagnose various diseases and abnormal conditions. Diagnostic methods based on audio signal processing are non-invasive and inexpensive and can be especially useful in under-developed countries, where inadequate medical specialists ...
Read More
Analysis and examination of sound of organs can be utilized in order to diagnose various diseases and abnormal conditions. Diagnostic methods based on audio signal processing are non-invasive and inexpensive and can be especially useful in under-developed countries, where inadequate medical specialists and equipment has led to high fatality rates. Development of accessible methods based on machine learning can aid with early diagnosis. we used a convolutional network to attain the advantages of transfer learning. In previous studies, models have been proposed that feed spectrograms with frequency characteristics as inputs to the convolutional network. In this article, we propose a model which additionally employs a recurrent representation (Recurrence plot) that reflects the temporal characteristics of the sound. The audio data sequence is investigated by adding the temporal attention mechanism and the bi-directional recurrent gates for weighting data according to its informational value. Data used in this article is from the ICBHI lung sound database. The presented model was able to classify lung sounds into three categories: healthy, chronic obstructive pulmonary disease (COPD), and other diseases with an accuracy of 97%, which shows the superiority of the proposed method compared to results obtained from previous methods on the same database.
Biomedical Signal Processing / Medical Signal Processing / Biosignal Processing
Amir Reza Asadi; Aboozar Ghaffari
Volume 16, Issue 4 , March 2023, , Pages 41-50
Abstract
One of the procedures for estimating fetal heart rate (FHR) is the use of an electrocardiogram (ECG). The ECG is a safe, inexpensive, and convenient method that can be used for remote monitoring, so maternal abdominal ECG recording (AECG) is used. The AECG signal, in addition to the fetal ECG (FECG), ...
Read More
One of the procedures for estimating fetal heart rate (FHR) is the use of an electrocardiogram (ECG). The ECG is a safe, inexpensive, and convenient method that can be used for remote monitoring, so maternal abdominal ECG recording (AECG) is used. The AECG signal, in addition to the fetal ECG (FECG), includes the maternal ECG (MECG), maternal or fetal muscle activity, fetal brain activity, and noise, making it difficult to estimate the fetal heart rate based on the abdominal signal. In this study, the fetal heart rate is estimated from the single-channel AECG signal utilizing non-negative matrix factorization (NMF). In this method, the short-time Fourier transform (STFT) is used to obtain time-frequency information of the abdominal signal. Next, the NMF utilizes the STFT matrix as input. The rows of the non-negative matrix resulting from the NMF contain the content of maternal, fetal, and noise, which are used to detect R-peak and FHR. It performs well when MECG and FECG amplitudes are close together, which is one of the advantages of this method. The robustness and performance of the proposed algorithm have been compared with other state-of-the-art single-channel approaches, including deep learning models, on two databases, ADFECGDB and PCDB. Statistical analysis demonstrates that the proposed method is capable of estimating FHR and R-peak accurately. As a result, the proposed method is suitable for long-term fetal monitoring.
Biomedical Signal Processing / Medical Signal Processing / Biosignal Processing
Parastoo Sadeghinia; Hamed Danandeh Hesar
Volume 16, Issue 3 , December 2022, , Pages 271-287
Abstract
Phonocardiography (PCG) signals provide valuable information about the heart valves .These auditory signals can be useful in the early diagnosis of heart diseases. Automatic heart sound classification has a promising potential in the field of heart pathology. In this research, a new method based on machine ...
Read More
Phonocardiography (PCG) signals provide valuable information about the heart valves .These auditory signals can be useful in the early diagnosis of heart diseases. Automatic heart sound classification has a promising potential in the field of heart pathology. In this research, a new method based on machine learning techniques is proposed for discriminating normal and abnormal heart sounds. In this method, first, the heart sounds are segmented into 4 main parts: S1, S2, systole and diastole segments. From these segments, statistical and time-frequency features are extracted for classification. Before classification, the distinctive features are selected using two approaches. In the first approach, the feature selection is accomplished using particle swarm optimization algorithm (PSO). In the second approach, we use Sequential Forward Feature Selection (SFFS) method. The proposed method was evaluated on the Physionet 2016 Challenge database using 10-fold cross-validation method. In this database, the number of normal and abnormal PCG signals are not balanced. Therefore, in this paper, the synthetic minority over-sampling technique (SMOTE) is applied to produce balanced data. The evaluation results showed that the proposed method can distinguish the normal heart sounds from abnormal ones with accuracy of 98/03% and sensitivity and specificity of 97.64%, 98.43%respectively.
Biomedical Signal Processing / Medical Signal Processing / Biosignal Processing
Faezeh Daneshmand-Bahman; Ateke Goshvarpour
Volume 16, Issue 2 , September 2022, , Pages 115-131
Abstract
Anxiety disorders are one of the most common and debilitating mental disorders worldwide. On the other hand, since 2019, with the outbreak of Covid-19, anxiety has increased among people, especially the medical staff. Currently, anxiety is diagnosed (when the symptoms are severe enough) using a questionnaire ...
Read More
Anxiety disorders are one of the most common and debilitating mental disorders worldwide. On the other hand, since 2019, with the outbreak of Covid-19, anxiety has increased among people, especially the medical staff. Currently, anxiety is diagnosed (when the symptoms are severe enough) using a questionnaire by a specialist. To resolve this shortcoming, researchers have recently paid attention to the use of brain signals. Consequently, the present study aimed to diagnose anxiety using brain signals. The novelty of this study is the use of the Chebyshev chaotic map for the first time in biological signal analysis. It used the DASPS database, which includes a 14-channel electroencephalogram (EEG) of 23 people (10 men and 13 women, with a mean age of 30 years). The self-assessment manikin scores were used to divide anxiety into two and four levels. First, the data were normalized. Then, the chaotic map was reconstructed and divided into 128 strips. The density of points in each of the strips was calculated. Two indicators were considered as features, (1) maximum density and (2) its corresponding sample. Finally, features were applied to Support Vector Machines (SVM) and k-Nearest Neighbors (K-NN) in 5 ways, (1) feature 1 of all channels, (2) feature1 mapping of all channels using principal component analysis (PCA), (3) feature 2 of all channels, (4) feature 2 mapping of all channels using PCA and (5) each feature - each channel separately. The results show a maximum accuracy of 93.75% for diagnosing two levels of anxiety and 96.15% for diagnosing four levels of anxiety. In addition, K-NN outperformed SVM. Accordingly, the proposed algorithm can be introduced as a suitable approach for diagnosing anxiety.
Biomedical Signal Processing / Medical Signal Processing / Biosignal Processing
Hamid Shafaatfar; Mehdi Taghizadeh; Morteza Valizadeh; Mohamad Hossein Fatehi
Volume 16, Issue 2 , September 2022, , Pages 147-158
Abstract
Automatic detection of cardiac arrhythmias is very important for the successful treatment of heart disease and machine learning is used for this purpose. To correctly classify arrhythmic classes, it is important to extract the appropriate features to distinguish between different classes. In this paper, ...
Read More
Automatic detection of cardiac arrhythmias is very important for the successful treatment of heart disease and machine learning is used for this purpose. To correctly classify arrhythmic classes, it is important to extract the appropriate features to distinguish between different classes. In this paper, a deep convolutional neural network is used to extract the feature. Due to the fact that the heart rates of different patients are very different, arrhythmia classes will have many intra-class changes. To reduce intra-class changes, each patient’s heart rate is mapped with a dedicated function to increase its resemblance to the heart rate of one of the training patient data’s. The proposed specific mapping reduces intra-class changes and significantly increases the classification accuracy of cardiac arrhythmias. To prove the effectiveness of the proposed method, its results were compared with several new studies based on three criteria for accuracy, sensitivity and specificity and on the same data set. The accuracy obtained is about 96.24%, which shows the better performance of the proposed method compared to other works.
Biomedical Signal Processing / Medical Signal Processing / Biosignal Processing
Masoud Moradi; Sina Shamekhi
Volume 16, Issue 2 , September 2022, , Pages 167-182
Abstract
In recent years, the fabrication of devices that can facilitate the difficulty of communication between deaf people and the general public and translate sign language has attracted interest from researchers. But problems such as low accuracy and calculation speed and the high cost of tools have hindered ...
Read More
In recent years, the fabrication of devices that can facilitate the difficulty of communication between deaf people and the general public and translate sign language has attracted interest from researchers. But problems such as low accuracy and calculation speed and the high cost of tools have hindered the commercialization of research. Another challenge in making a practical tool is the necessity of good performance of the methods in the perspective of training by leave-one-subject-out or in other words classifying the data of a new person. Therefore, in this article, an efficient method for detecting hand gestures with the purpose of sign language translation has been presented, so that while using a method with lower dimensions, better performance can be obtained in all kinds of training methods. In the proposed method, the features consisting of the mean absolute value, variance, root mean square, waveform length, kurtosis, and skewness have been extracted from the empirical wavelet transformation of the electromyogram and inertial signals. Then, by the ReliefF method, effective features have been selected and for the classification of hand gestures, a support vector machine classifier has been used. The accuracy percentages of the proposed method on the PSL database and DB2, DB3, DB5, and DB7 datasets of the NinaPro database, have been respectively obtained as follows: 99.31%, 97.11%, 96.58%, 96.12%, and 97.32% in the word-subject training approach, 99.78%, 97.22%, 95.46%, 97.23%, and 97.72% in the word-all-subject training approach, and 97.43%, 94.68%, 89.66%, 91.55%, and 94.81% in the leave-one-subject-out method.
Biomedical Signal Processing / Medical Signal Processing / Biosignal Processing
Mohammad Davood Khalili; Vahid Abootalebi; Hamid Saeedi-Sourck
Volume 16, Issue 1 , May 2022, , Pages 75-94
Abstract
The human brain is one of the most complex and heterogeneous networks, and brain signals contain a lot of information, so researchers in this field are always looking for proper solutions to select meaningful features and reduce the dimension of this information appropriately to lead to better classification. ...
Read More
The human brain is one of the most complex and heterogeneous networks, and brain signals contain a lot of information, so researchers in this field are always looking for proper solutions to select meaningful features and reduce the dimension of this information appropriately to lead to better classification. Two of the new tools for brain signal processing are Graph Signal Processing (GSP) and Meta-heuristic and Evolutionary methods. In this paper, a geometric structure and a mixed structure are considered for the brain graph and the weights of the edges in the mixed structure are calculated by a combination of two measures: geometric distance and correlation. To reduce the graph dimension, the weighted degree metric and a combination of the Kron reduction method and Graph Fourier Transform (KG) are used to properly preserve the information of all vertices of the graph into the selected vertices. Feature extraction is performed by Ledoit-Wolf shrinkage estimation and Tangent Space Mapping (TSM) method. For dimension reduction of extracted features, Principal Component Analysis (PCA) method and feature selection based on Differential Evolution (DE) are used. The selected features are given to several well-known machine learning classifiers. To evaluate the performance of the proposed method, dataset IVa from BCI Competition III has been used. The results show that the average classification accuracy of the proposed KG-PCA method with SVM-RBF and DT classifiers, in the structural graph and the functional-structural graph, is higher than the TSM-GFT method expressed in previous studies, and the DT classifier has achieved an average accuracy of 91.15±1.17. Also, according to the obtained results, the performance of the proposed KG-DE method has been better compared to KG-PCA and in the best case, the average accuracy of the SVM-RBF classifier is equal to 95.50±1.27.
Biomedical Signal Processing / Medical Signal Processing / Biosignal Processing
Tahereh Taleei; Ali Motie Nasrabadi
Volume 15, Issue 4 , March 2022, , Pages 341-353
Abstract
To interact with such an ever-changing environment in which we live, our brain requires to continuously generate and update expectations about relevant upcoming events and their estimation for the corresponding sensory and motor responses. The goal of this study is to investigate the connectivity in ...
Read More
To interact with such an ever-changing environment in which we live, our brain requires to continuously generate and update expectations about relevant upcoming events and their estimation for the corresponding sensory and motor responses. The goal of this study is to investigate the connectivity in time perception in the two predictable and unpredictable conditions. The data needed for the study from EEG signals recorded from the existing database that included an experiment was conducted on 29 healthy subjects in the two predictable and unpredictable conditions and in 4 delays of 83, 150, 400, 800 ms for each person was done. To estimate the functional connectivity between brain regions, we used the phase lag index method. This method is used to detect time perception in two conditions, predictable and unpredictable events. Initially, by comparing the two conditions in 4 delays was shown that more of the differences were in the gamma, beta, and theta bands. Also, the significant difference between the delays in the predictable condition was greater than the unpredictable condition. Then, the difference between the two conditions in each delay was discussed. The results showed a significant difference in all delays. The alpha band in the unpredictable condition in 400-ms delay, the number of connectivity between occipital and temporal regions was increased and stronger, and also the mean of the unpredictable connectivity was higher than predictable condition. In the delta band for 150, 400 and 800-ms delays, there was connectivity between the central and frontal regions, whereas in 83-ms-delay there was stronger connectivity between the central and prefrontal regions. The right hemisphere of the prefrontal is important in time perception. At the longest delay (800 ms), in three bands, delta, theta, and beta, connectivity decreased in both conditions compared to the other delays.
Biomedical Signal Processing / Medical Signal Processing / Biosignal Processing
Hamed Danandeh Hesar; Amin Danandeh Hesar
Volume 15, Issue 3 , December 2021, , Pages 221-234
Abstract
Extended Kalman filter (EKF) is a well-known nonlinear Bayesian framework that has been deployed in various fields of ECG processing. However, it’s not very effective in removing non-stationary noises such as muscle artifacts (MA) which are common in ECG recordings. This paper addresses this issue ...
Read More
Extended Kalman filter (EKF) is a well-known nonlinear Bayesian framework that has been deployed in various fields of ECG processing. However, it’s not very effective in removing non-stationary noises such as muscle artifacts (MA) which are common in ECG recordings. This paper addresses this issue by proposing a new ECG dynamic model (EDM) and a novel formulation for EKF which improves its performance in non-stationary environments. In the new EDM, the measurement model is modified to include non-Gaussian, non-stationary additive noises as well as stationary ones. The proposed formulation for EKF algorithm in this paper enables it to perform better than standard EKF in removing non-stationary contaminants. The proposed filter also preserves the clinical characteristics of ECG signals better than standard EKF. In order to show the effectiveness of the proposed EKF algorithm, its denoising performance was evaluated on MIT-BIH Normal Sinus Rhythm database (NSRDB) in the presence of two different types of non-stationary contaminants; synthetic pink noise and real muscle artifact noise. The results showed that the proposed EKF framework in this paper has a significant outperformance over the standard EKF framework in non-stationary environments from both SNR improvement and MSEWPRD viewpoints.
Biomedical Signal Processing / Medical Signal Processing / Biosignal Processing
Hadi Grailu
Volume 15, Issue 3 , December 2021, , Pages 247-262
Abstract
Today, auscultation is one of the most effective methods in monitoring heart disease. With the advancement of technology and the facilitation of telecare on the one hand, and the increasing need for high quality and long-term recording of cardiac audio signals on the other hand, the amount of data generated ...
Read More
Today, auscultation is one of the most effective methods in monitoring heart disease. With the advancement of technology and the facilitation of telecare on the one hand, and the increasing need for high quality and long-term recording of cardiac audio signals on the other hand, the amount of data generated has increased and therefore, the storage and transmission of these signals has become a challenge. This, in turn, demonstrates the importance and necessity of using efficient methods for compression of these types of signals. In this paper, a lossy compression method is proposed for PCG signals recorded at a relatively high sampling rate so that it can control the quality of the compressed signal. This method is based on two techniques: "two-stage downsampling" and "pattern matching". The proposed two-stage downsampling technique increases the amount of compression ratio and at the same time reduces the computational complexity. The pattern matching technique is able to reduce the inter-period redundancy and therefore, increase the compression ratio. The simulation results of the proposed method on the two databases of the University of Michigan and the University of Washington showed that the two-stage downsampling and pattern matching techniques have a large contribution in increasing the compression ratio. The performance of the proposed method was evaluated according to the PRD and CR criteria and compared with that of some existing methods. In this evaluation, for the PRD range of 5%, the CR value was between 2500 and 3900 for the University of Michigan database and between 2500 and 4125 for the University of Washington database. Also, the results of applying the proposed method on the Pascal database showed that the efficiency of the proposed method depends to a large extent on the quality and regularity of the input PCG signals.
Biomedical Signal Processing / Medical Signal Processing / Biosignal Processing
Behnaz Sheikholeslami; Ghasem Sadeghi Bajestani; Reza Yaghoobi Karimui; Reyhaneh Zarifiyan
Volume 15, Issue 1 , May 2021, , Pages 29-46
Abstract
Attention Deficit/Hyperactivity Disorder (ADHD) is a neurodevelopmental disorder that can affect people of all ages in the community, especially children, and cause changes in their behavior. Previous studies have often focused on frequency domain processing or the nonlinear dynamic aspects of EEG signals ...
Read More
Attention Deficit/Hyperactivity Disorder (ADHD) is a neurodevelopmental disorder that can affect people of all ages in the community, especially children, and cause changes in their behavior. Previous studies have often focused on frequency domain processing or the nonlinear dynamic aspects of EEG signals such as correlation dimension, fractal dimension, Lyapunov exponent, entropy, and recurrence rate of brain processes to differentiate individuals with ADHD. In this study, we evaluate the volume of the EEG signal oscillation basin using Poincare sections in the phase space of EEG signals of people with ADHD and healthy people and sort this space as well as extract various geometric features. We present a different perspective of complexity of brain activity and the level of dynamism of people with ADHD compared to healthy individuals. Finally, by evaluating the extracted features and using the SFS algorithm based on the RBF-SVM classifier, we were able to separate people with ADHD from healthy people in the groups of children and adults, with accuracy of 93.20±2.04 and 95.60±1.13. The results of this study showed that the volume of the EEG signal oscillation basin in people with ADHD was significantly higher than healthy people, which indicates an increase in the degree of dynamism and thus a decrease in the complexity of brain activity in these people. It was also identified in this study that the increase in the volume of the EEG signal oscillation basin in children is more than adults, which indicates an increase in the level of dynamism of children compared to adults. Therefore, ADHD and age can be introduced as two important factors in changing the volume of the EEG signal oscillation basin.
Biomedical Signal Processing / Medical Signal Processing / Biosignal Processing
Aref Einizade; Sepide Hajipour Sardouie
Volume 14, Issue 3 , October 2020, , Pages 221-233
Abstract
The brain electrical signal has been widely used in clinical and academic research, due to its ease of recording, non-invasiveness, and precision. One of the applications can be emotion recognition from the brain's electrical signal. Generally, two types of parameters (Valence and Arousal) are used to ...
Read More
The brain electrical signal has been widely used in clinical and academic research, due to its ease of recording, non-invasiveness, and precision. One of the applications can be emotion recognition from the brain's electrical signal. Generally, two types of parameters (Valence and Arousal) are used to determine the type of emotion which in turn indicate "positive or negative" and "level of extroversion or excitement" for a specific emotion. The significance of emotion is determined by the effects of this phenomenon on daily tasks, especially in cases where the person is confronted with activities that require careful attention and concentration. In the emotion recognition problem, firstly, using proper emotion stimuli, different emotions are created for the subjects under study and the brain signals corresponding to each stimulus are recorded. The two main steps for solving the emotion recognition problem are extracting suitable features and using appropriate classification or regression methods. In previous studies, different visual and auditory have been used and various linear and nonlinear features and classifiers have been investigated. In this paper, the main goal was the improvement of linear regression algorithms to estimate the criteria for recognizing human emotions more efficiently. For this purpose we proposed a new algorithm that uses the sparseness of the mixing vector along with the linear regression cost function. The effectiveness of the proposed algorithm on simulated data has been investigated and its superiority to linear regression algorithms such as PLS, LASSO, SOPLS and Ridge was shown. Also, to apply the proposed algorithm on EEG data corresponding to emotion recognition, the DEAP dataset was used and the AR coefficients were extracted from the EEG signals. The results obtained from the proposed algorithm were compared with those of the other linear regression algorithms, which in total showed the relative superiority of the proposed method.
Biomedical Signal Processing / Medical Signal Processing / Biosignal Processing
Hessam Ahmadi; Emad Fatemizadeh; Alimotie Nasrabadi
Volume 14, Issue 3 , October 2020, , Pages 235-249
Abstract
Functional Magnetic Resonance Imaging (fMRI) is a non-invasive neuroimaging technique for analyzing the brain functions through low-frequency fluctuations called the Blood-Oxygen-Level-Dependent (BOLD) signals. Measurement of the functional connectivity in brain networks is usually done by the fMRI time-series ...
Read More
Functional Magnetic Resonance Imaging (fMRI) is a non-invasive neuroimaging technique for analyzing the brain functions through low-frequency fluctuations called the Blood-Oxygen-Level-Dependent (BOLD) signals. Measurement of the functional connectivity in brain networks is usually done by the fMRI time-series through Pearson Correlation Coefficients (PCC). As the PCC shows linear dependencies, in this study, non-linear relationships in the fMRI signals of the patients with Alzheimer's Disease (AD) were investigated using the kernel trick method. Kernel trick approach maps the input information into a higher dimension space and implements the linear calculations in a new space that is proportionate to the non-linear relationships in the primary space. After generating the weighted undirected brain graphs based on the Automated Anatomical Labeling (AAL) atlas, different kernel functions with different parameters were applied. Then the graph global measures including degree, strength, small-worldness, modularity, and efficiencies features were computed and the non-parametric permutation test was performed. According to the results, the kernel trick method showed more significant differences with AD and healthy subjects in comparison with the simple PCC and it could be because of the non-linear correlations that are not captured by the PCC. Among different kernel functions, the Polynomial function had the best performance. Applying this kernel, the classification was done by the Support Vector Machine (SVM) classifier. The achieved accuracy was equal to 98.68±0.79%. The Occipital and Temporal lobes and also the Default Mode Network (DMN) were analyzed and the kernel trick method showed more significant differences in all of them. It is worthwhile to mention that the right and left Angular areas of DMN showed no significant changes in none of the methods and it could be concluded that the AD does not affect this areas effectively.
Biomedical Signal Processing / Medical Signal Processing / Biosignal Processing
Maryam Dorvashi; Neda Behzadfar; Ghazanfar Shahgholian
Volume 14, Issue 2 , July 2020, , Pages 109-119
Abstract
Consumption of alcohol contributes to disorders in brain. In this study, in order to detect the consumption of alcohol, electroencephalogram (EEG) signal of 20 participants (10 alcoholic and 10 control subjects) recorded by 64 channels was investigated. Frequency and non-frequency features of EEG signal ...
Read More
Consumption of alcohol contributes to disorders in brain. In this study, in order to detect the consumption of alcohol, electroencephalogram (EEG) signal of 20 participants (10 alcoholic and 10 control subjects) recorded by 64 channels was investigated. Frequency and non-frequency features of EEG signal including power spectrum of signal, permutation entropy, approximate entropy, Katz fractal dimension and Petrosion fractal dimension were extracted to analyses the EEG signal. Statistical analysis was used to investigate the significant differences between the alcohol and control groups. The Davis-Bouldin (DB) criterion was used to select the best channel distinguishing between the alcoholic and non-alcoholic EEG signal. Results showed that between frequency features, power of lower2 alpha frequency decreased in alcoholic individuals and regarding the DB criterion, the CP3 channel (DB=1.7638) showed the best discrimination between the alcohol and control groups. Also, among the non-frequency features, the Katz fractal dimension increased in the control group and FP2 channel (DB = 0.862) had the best discrimination. Eventually, power of Lower2-alpha frequency band and Katz fractal dimension fed into the nearest neighbor classifier (KNN), 71% and 93% accuracy were achieved, respectively. According to the results, it can be concluded that the best feature and channel discriminating between alcohol and control groups is the Katz fractal dimension and FP2 channel.
Biomedical Signal Processing / Medical Signal Processing / Biosignal Processing
Alireza Talesh Jafadideh; Babak Mohammadzadeh Asl
Volume 14, Issue 2 , July 2020, , Pages 143-157
Abstract
Autism spectrum disorder (ASD) is a highly prevalent neurodevelopmental disorder characterized by impaired social communication and restricted and repetitive behaviors. Comparison study between ASD and typically control (TC) subjects through magnetic resonance imaging (MRI) provides valuable understanding ...
Read More
Autism spectrum disorder (ASD) is a highly prevalent neurodevelopmental disorder characterized by impaired social communication and restricted and repetitive behaviors. Comparison study between ASD and typically control (TC) subjects through magnetic resonance imaging (MRI) provides valuable understanding for differences in brain function. Recently, through dynamic functional connectivity (DFC) analysis, it is found that brain functional connectivity possesses dynamic nature and shows transient connectivity patterns (“states”) repeating over time. In this comparison study between ASD and TC, we employed the rest functional MRI (rfMRI) data of San Diego State University (SDSU) of ABIDE II database to examine the brain intra and inter network connectivity and also to investigate the relations of age and social responsiveness scale (SRS) score (score measuring autistic traits) to brain inter regions connectivity strength. These aims were implemented in all DFC states. The ASD subjects experienced more the state with less intra and inter network connections. Further, the DMN segregation reduction from other functional networks emerged as a common them. Furthermore, in ASD, the connection strength between auditory and visual networks was decreased by increasing the age. In ASD, the SRS had more positive relation to connectivity strength existing between cerebellar, auditory, visual networks and cognitive control network in comparison to TC. All these results demonstrate that some differences exist in brain network connection of ASD in comparison to the TC subjects and these differences can be more distinctively revealed by employing DFC analysis.
Biomedical Signal Processing / Medical Signal Processing / Biosignal Processing
Nasrin Sho'ouri
Volume 14, Issue 2 , July 2020, , Pages 159-168
Abstract
Previous research has shown that eye movements in people with Attention Deficit Hyperactivity Disorder (ADHD) and healthy people were different, and it is possible that there is a difference between the two groups' EOG signals. Therefore, in the present study, the recorded EOG signals of 30 children ...
Read More
Previous research has shown that eye movements in people with Attention Deficit Hyperactivity Disorder (ADHD) and healthy people were different, and it is possible that there is a difference between the two groups' EOG signals. Therefore, in the present study, the recorded EOG signals of 30 children with ADHD and 30 healthy children were examined during performing an attentional related task. For this purpose, the scaling exponents of the two groups' EOG signals were calculated and the differences between the two groups were examined using statistical tests. The EOG signals were then classified using a Growing Neural Gas network. The results show that the scaling exponents of the EOG signals in children with ADHD were significantly higher than that of healthy children (p < 0.001). This result shows that the decay slope of power spectrum in ADHD children is more as compared to healthy children. In addition, the EOG signals were classified into two groups with a detection accuracy of 72.22±2.8%. The results of this study could be used to design a course of treatment with EOG biofeedback to treat or reduce the symptoms of people with ADHD.
Biomedical Signal Processing / Medical Signal Processing / Biosignal Processing
Farzaneh Dasar; Majid Ghoshuni; Ghasem Sadeghi Bajestani
Volume 14, Issue 1 , May 2020, , Pages 13-22
Abstract
Autism spectrum disorder is a developmental disorder that involves disorders in social interaction and communication and repetitive or stereotypical behavior. In some children with autism, the sensitivity to acoustic stimuli is much higher than normal (hypersensitive) versus in some other children, this ...
Read More
Autism spectrum disorder is a developmental disorder that involves disorders in social interaction and communication and repetitive or stereotypical behavior. In some children with autism, the sensitivity to acoustic stimuli is much higher than normal (hypersensitive) versus in some other children, this sensitivity is less than normal (hyposensitive). In this study a method for evaluation of auditory system of hypersensitive and hyposensitive autism children using event related potentials (ERPs) was presented. The EEG signal was recorded from 10 autism children (2 girls) with average age of 7.7±2.31 years. In order to record ERPs, 2000 audio stimulation based on the MissMatch Negetivity (MMN) Pattern was presented to participants. These stimulus include 1600 standard sounds with a frequency of 1000 Hz, deviant at 1300 Hz, and noise at frequencies of 1500-1000, 500 and 2000 Hz. In order to analyze ERP data, 18 time domain features have been extracted from the ERP components in all three types of stimulation (standard, deviation, noise). Based on the results, in the deviant stimuli, total positive area of the Pz channel in the hypersensitive group was significantly increased (p=0.028) compared to the hyposensetive group. Also, in the noise stimuli, total positive area in C4 and Pz channels has significantly increased (p=0.028, p=0.009) in the hyposensitive group compared to hypersensetive group. In conclusion, when hypersensitive children were exposed to deviant stimulue, neural activity was increased in parietal lobe, wheras in hyposensitive children neural activity increased in central and parietal lobe during noise stimulue. Therefore, this method can be useful in assessing children's autism spectrum in terms of hearing loss sensitivity.
Biomedical Signal Processing / Medical Signal Processing / Biosignal Processing
Somayeh Raiesdana; Samaneh Safari
Volume 13, Issue 2 , August 2019, , Pages 117-134
Abstract
In this study, a neuromarketing project was conducted via EEG signal processing in which the individuals’ interest for buying a relatively luxurious decorative product (which has a relative advantage in exports based on commonly evaluated criteria and indicators in economic) was evaluated. EEG ...
Read More
In this study, a neuromarketing project was conducted via EEG signal processing in which the individuals’ interest for buying a relatively luxurious decorative product (which has a relative advantage in exports based on commonly evaluated criteria and indicators in economic) was evaluated. EEG signals of 24 participants during observing and selecting gemstone images were recorded and processed in order to analyze statistical significance of brain activity variations involved in the emotional (liking) and the decision making (choosing) processes. The recorded signals during the stimulation and selection phases were pre-processed in several steps to remove the existing noises and artifacts. Then, the 19-channel EEGs were processed via multiple tools to indicate active brain regions while watching gemstones. Brain mapping and regional analysis indicated that the occipital>frontal>limbic regions were more activated than other regions. Moreover, the left hemisphere has been more active than the right hemisphere. At the next step, nonlinear entropy feature of each signal segment was extracted to be used for training a neurofuzzy system which is an automatic classifier that learns to classify the individuals’ choices. The classification has resulted in 86.25% precision and 87.4% accuracy in a three-class classification task (including two pleasant selections and one unpleasant selection). At the final step, using a questionnaire filled by participants following the recording session, a number of statistical analyses were performed over the self-conscious and unconscious by means of statistical tools including t-test, analysis of variance and regression. The results of statistical tests indicated that there are significant differences for the cognition of liking or preferring among different choices and based on the selections made by women and men. Furthermore, the lack of existence of a significant difference between conscious and unconscious choices were rejected.
Bioelectrics
Amir Soleymankhani; Vahid Shalchyan
Volume 12, Issue 2 , September 2018, , Pages 85-96
Abstract
The extracellular recording from the brain's single neurons is known as a popular method in neuroscience and neuro-rehabilitation engineering. These recordings include the activity of all neurons around the electrode, for better use of which, spike sorting methods should be utilized to obtain the activity ...
Read More
The extracellular recording from the brain's single neurons is known as a popular method in neuroscience and neuro-rehabilitation engineering. These recordings include the activity of all neurons around the electrode, for better use of which, spike sorting methods should be utilized to obtain the activity of single neurons. Based on the structural properties of the neuron, such as its dendritic tree, and the distance and direction of it relative to the electrode, it can be claimed that the form of its spike waveform is unique and constant. However, spike sorting under low signal-to-noise ratio (SNR) conditions is always accompanied with challenges. A spike sorting algorithm usually consists of three sections including the spike detection, feature extraction, and classification. In this paper, a method based on optimization of continuous wavelet coefficients is presented which is effective in low SNR values. In the proposed method, after the calculation of the parameterized wavelet coefficients, using the Euclidean distance and the area under the receiver operator characteristic curve, the best parameters were chosen to increase the separation of the features, so that a suitable scale was first found with the Euclidean distance criterion and then the translation parameter was obtained with the second criterion. In this research k-means algorithm was used for the clustering as a simple but efficient method. For evaluation, three simulated data sets were made in 9 different SNRs with a modeled background noise. The obtained results from simulated data showed that the optimization of parameters in continuous wavelet transform using the proposed algorithm could effectively improve the spike sorting performance compared to principal component analysis method.
Biomedical Signal Processing / Medical Signal Processing / Biosignal Processing
Raheleh Davoodi; Mohammad Hasan Moradi
Volume 12, Issue 1 , June 2018, , Pages 25-39
Abstract
Depression is one of the most common mental disorders in the current century where early diagnosis can result in better treatment. One of the depression diagnostic methods is the analysis of the brain electrical signals. In this paper, we are seeking for a method to distinguish among the levels of the ...
Read More
Depression is one of the most common mental disorders in the current century where early diagnosis can result in better treatment. One of the depression diagnostic methods is the analysis of the brain electrical signals. In this paper, we are seeking for a method to distinguish among the levels of the depression. The proposed model is a deep rule-based system based on the stacked principle and focuses on the interpretability of the rules alongside high accuracy. Fuzzy systems have the proper capability in the classification of medical data with various levels of uncertainty. Moreover, in the recent years, deep learning has been taken considerable attention in the field of Artificial Intelligence. In this paper, we aim to benefit from capabilities of both fields. The proposed architecture employs a robust fuzzy clustering approach that can determine an appropriate number of clusters in each layer, unsupervised and a hierarchical stacked structure to transfer the interpretable trained rules from the previous layers with the same linguistic labels to the next layer. The interpretability is due to the presence of the input space into the consequent ones. The presence of the output of the previous layer’s rules at the input space of the next parts equals to a fuzzy system with non-linear consequent or the certainty factor in a fuzzy system with linear consequent. EEG data were preprocessed and time, frequency and nonlinear features such as recurrent plot were extracted and selected and after that were employed in the proposed system. The proposed system was compared with common classifiers like Neural Net, Support Vector Machine, Naive Bayes, Decision Tree and Linear Discriminant Analysis. Accuracy results for the test data in 30 folds (49.01% in comparison to 41.42%, 40.47%, 40.01%, 38.35% and 40.28% respectively) demonstrate the considerable performance of the proposed system.
Biomedical Signal Processing / Medical Signal Processing / Biosignal Processing
Hamed Danandeh Hesar; Maryam Mohebbi
Volume 11, Issue 4 , February 2018, , Pages 275-289
Abstract
Marginalized particle extended Kalman filter (MP-EKF) takes advantage of both extended Kalman filter and particle filter frameworks to estimate nonlinear ECG dynamic models (EDMs) with reduced number of calculations in comparison to typical particle filters. However, due to existence of Kalman filter ...
Read More
Marginalized particle extended Kalman filter (MP-EKF) takes advantage of both extended Kalman filter and particle filter frameworks to estimate nonlinear ECG dynamic models (EDMs) with reduced number of calculations in comparison to typical particle filters. However, due to existence of Kalman filter framework inside MP-EKF, some limitations are introduced in implementation of MP-EKF especially in embedded systems with finite numerical accuracies. In this paper, for the first time, we propose a square root filtering strategy for MP-EKF which alleviates these restrictions using factorization. Typical or other square-root Kalman filters cannot be employed inside MP-EKF due to presence of minus operations in some equations of MP-EKF. However, our method can be implemented in MP-EKF structure. The proposed method can be used in any EDM previously used by EKF based frameworks in the field of ECG processing.
Biomedical Signal Processing / Medical Signal Processing / Biosignal Processing
Marjan Mozaffarilegha; Seyed Mohammad Sadegh Movahed
Volume 11, Issue 3 , September 2017, , Pages 255-264
Abstract
The complexities and the effects of inter-subject variations on the encoding of sounds are features of the brainstem processing. Examining such data based on linear analysis is not reliable, encouraging to take into account non-linear methods which are effective ways of explaining such non-stationary ...
Read More
The complexities and the effects of inter-subject variations on the encoding of sounds are features of the brainstem processing. Examining such data based on linear analysis is not reliable, encouraging to take into account non-linear methods which are effective ways of explaining such non-stationary signals. The purpose of this study is to explore the behavior of the brainstem in response to complex auditory stimuli /da/ using Multifractal Detrended Fluctuation Analysis modified by Singular Value Decomposition (SVD), Adaptive Detrending (AD) and Empirical Mode Decomposition (EMD). Auditory brainstem responses to synthetic /da/ stimuli were recorded for 40 normal subjects with a mean age of 22.7 years. MFDFA is carried out on the s-ABR time series data to evaluate the variation of their complexity and multiscaling. To utilize optimal Detrending of s-ABR time series, AD, SVD and EMD algorithms are applied on time series. By computing the fluctuation function and evaluating scaling behavior, scaling exponents such as generalized Hurst exponent and multifractal spectrum are determined. Given results in this method indicate that underlying signal has non-stationary nature in small scales, but property of system is controlled by trend in large scales. There is a crossover at msec on the behavior of fluctuation function corresponding to dominant sinusoidal trend in all samples. The average of Hurst exponent is at 68% confidence interval in small scales msec. The -dependency of demonstrate that underlying data sets have multifractality nature and are almost due to long-range correlations. The width of singularity spectrum which is a measure of the signal complexity of underlying data in average equates to at confidence interval.