A Review on Machine Learning for EEG Signal Processing in Bioengineering

Musical Emotions Recognition Using Entropy Features and Channel Optimization Based on EEG

The dynamic of music is an important factor to arouse emotional experience, but current research mainly uses short-term artificial stimulus materials, which cannot effectively awaken complex emotions and reflect their dynamic brain response. In this paper, we used three long-term stimulus materials with many dynamic emotions inside: the "Waltz No. 2" containing pleasure and excitement, the "No. 14 Couplets" containing excitement, briskness, and nervousness, and the first movement of "Symphony No. 5 in C minor" containing passion, relaxation, cheerfulness, and nervousness. Approximate entropy (ApEn) and sample entropy (SampEn) were applied to extract the non-linear features of electroencephalogram (EEG) signals under long-term dynamic stimulation, and the K-Nearest Neighbor (KNN) method was used to recognize emotions. Further, a supervised feature vector dimensionality reduction method was proposed. Firstly, the optimal channel set for each subject was obtained by using a particle swarm optimization (PSO) algorithm, and then the number of times to select each channel in the optimal channel set of all subjects was counted. If the number was greater than or equal to the threshold, it was a common channel suitable for all subjects. The recognition results based on the optimal channel set demonstrated that each accuracy of two categories of emotions based on "Waltz No. 2" and three categories of emotions based on "No. 14 Couplets" was generally above 80%, respectively, and the recognition accuracy of four categories based on the first movement of "Symphony No. 5 in C minor" was about 70%. The recognition accuracy based on the common channel set was about 10% lower than that based on the optimal channel set, but not much different from that based on the whole channel set. This result suggested that the common channel could basically reflect the universal features of the whole subjects while realizing feature dimension reduction. The common channels were mainly distributed in the frontal lobe, central region, parietal lobe, occipital lobe, and temporal lobe. The channel number distributed in the frontal lobe was greater than the ones in other regions, indicating that the frontal lobe was the main emotional response region. Brain region topographic map based on the common channel set showed that there were differences in entropy intensity between different brain regions of the same emotion and the same brain region of different emotions. The number of times to select each channel in the optimal channel set of all 30 subjects showed that the principal component channels representing five brain regions were Fp1/F3 in the frontal lobe, CP5 in the central region, Pz in the parietal lobe, O2 in the occipital lobe, and T8 in the temporal lobe, respectively.

The sound stimulation method and EEG change analysis for development of digital therapeutics that can stimulate the nervous system: Cortical activation and drug substitution potential

INTRODUCTION

The purpose of this study is to propose a treatment method and the effect on the nervous system of digital therapeutics, which is a new treatment method to replace surgery and drug prescription for the treatment and prevention of diseases.

METHODS

The 20 subjects who participated in the experiment, including men and women, had an average age of 26 ± 2.40 years. The proposed treatment method used three types of sound stimulation and air or bone conduction sound transmission methods to induce total of 6-time EEG electroencephalogram(EEG) changes. EEG was measured with 200 sampling rate each in the P4, Cz, F8 and T7 channel located in the parietal, central, frontal and temporal lobes, respectively, according to the 10/10 system. A total of 2 min of data were created by extracting EEG signals with less noise from the measured data and the extracted data were applied with a 1-40 Hz Butterworth filter and a 50 Hz notch filter with a quality factor of 30. After that, EEG are subdivided into delta (0.5-4 Hz), theta (4-8 Hz), alpha (8-13 Hz), beta (13-30 Hz), and gamma (30-45 Hz) bands. Finally, EEG changes in response to sound stimuli were analyzed using power spectral density and T-test validation in the frequency band.

RESULTS

When a sound stimulus of less than 1 KHz was stimulated by air conduction, brainstem activation was induced and the reticular activation system was activated. In addition, a great potential for replacing drugs was confirmed by inducing changes in the nervous system similar to drugs used for sedation.

CONCLUSION

These results will be able to expand the concept of digital therapeutics, and it is expected that it will be developed as a safer treatment method that can replace surgery and drugs.

Electronic nose and wireless sensor network for environmental monitoring application in pulp and paper industry: a review

Pulp and paper industries emit various odorous gases during the pulp production and paper-making phase, which are unpleasant and have harmful effects on the human body. The working staffs are continuously exposed to these gases and develop various health issues. Hence, regular monitoring and analysis of such gases are necessary to avoid any sudden high concentration exposure and to prevent adverse health effects on the staff. An electronic nose (EN) has an array of gas sensors with an alert system for early detection of gases. Various ENs have been developed for varying applications till date. The detailed knowledge of the sensors used, their sensitivity and technology is helpful in development of any EN. The objective of this study is to comprehensively review various developed ENs with respect to their gas sensing and pattern recognition (PR) technologies. The information on gases released from pulp and paper industries is also compiled. The evolution of EN technology, its various applications, challenges in developing EN and its utility in safeguarding the industrial workers' life have been described. Further, gap analysis among previously developed EN, contemporary EN and wireless sensor network (WSN) is elaborated. It will facilitate future researchers for better selection of sensors and PR technologies while developing EN. The commonly used sensing technologies are described with their advantages, disadvantages and working principles. Metal oxide semiconductor (MOS) gas sensor and ANN algorithm show better result and hence recommended in the development of EN, whereas ZigBee protocol has been widely used for WSN.

Traumatic Brain Injury (TBI) Detection: Past, Present, and Future

Traumatic brain injury (TBI) can produce temporary biochemical imbalance due to leaks through cell membranes or disruption of the axoplasmic flow due to the misalignment of intracellular neurofilaments. If untreated, TBI can lead to Alzheimer's, Parkinson's, or total disability. Mild TBI (mTBI) accounts for about about 90 percent of all TBI cases. The detection of TBI as soon as it happens is crucial for successful treatment management. Neuroimaging-based tests provide only a structural and functional mapping of the brain with poor temporal resolution. Such tests may not detect mTBI. On the other hand, the electroencephalogram (EEG) provides good spatial resolution and excellent temporal resolution of the brain activities beside its portability and low cost. The objective of this paper is to provide clinicians and scientists with a one-stop source of information to quickly learn about the different technologies used for TBI detection, their advantages and limitations. Our research led us to conclude that even though EEG-based TBI detection is potentially a powerful technology, it is currently not able to detect the presence of a mTBI with high confidence. The focus of the paper is to review existing approaches and provide the reason for the unsuccessful state of EEG-based detection of mTBI.

Electroencephalogram signals emotion recognition based on convolutional neural network-recurrent neural network framework with channel-temporal attention mechanism for older adults

Reminiscence and conversation between older adults and younger volunteers using past photographs are very effective in improving the emotional state of older adults and alleviating depression. However, we need to evaluate the emotional state of the older adult while conversing on the past photographs. While electroencephalogram (EEG) has a significantly stronger association with emotion than other physiological signals, the challenge is to eliminate muscle artifacts in the EEG during speech as well as to reduce the number of dry electrodes to improve user comfort while maintaining high emotion recognition accuracy. Therefore, we proposed the CTA-CNN-Bi-LSTM emotion recognition framework. EEG signals of eight channels (P3, P4, F3, F4, F7, F8, T7, and T8) were first implemented in the MEMD-CCA method on three brain regions separately (Frontal, Temporal, Parietal) to remove the muscle artifacts then were fed into the Channel-Temporal attention module to get the weights of channels and temporal points most relevant to the positive, negative and neutral emotions to recode the EEG data. A Convolutional Neural Networks (CNNs) module then extracted the spatial information in the new EEG data to obtain the spatial feature maps which were then sequentially inputted into a Bi-LSTM module to learn the bi-directional temporal information for emotion recognition. Finally, we designed four group experiments to demonstrate that the proposed CTA-CNN-Bi-LSTM framework outperforms the previous works. And the highest average recognition accuracy of the positive, negative, and neutral emotions achieved 98.75%.

A data-driven machine learning approach for brain-computer interfaces targeting lower limb neuroprosthetics

Prosthetic devices that replace a lost limb have become increasingly performant in recent years. Recent advances in both software and hardware allow for the decoding of electroencephalogram (EEG) signals to improve the control of active prostheses with brain-computer interfaces (BCI). Most BCI research is focused on the upper body. Although BCI research for the lower extremities has increased in recent years, there are still gaps in our knowledge of the neural patterns associated with lower limb movement. Therefore, the main objective of this study is to show the feasibility of decoding lower limb movements from EEG data recordings. The second aim is to investigate whether well-known neuroplastic adaptations in individuals with an amputation have an influence on decoding performance. To address this, we collected data from multiple individuals with lower limb amputation and a matched able-bodied control group. Using these data, we trained and evaluated common BCI methods that have already been proven effective for upper limb BCI. With an average test decoding accuracy of 84% for both groups, our results show that it is possible to discriminate different lower extremity movements using EEG data with good accuracy. There are no significant differences (p = 0.99) in the decoding performance of these movements between healthy subjects and subjects with lower extremity amputation. These results show the feasibility of using BCI for lower limb prosthesis control and indicate that decoding performance is not influenced by neuroplasticity-induced differences between the two groups.

Evaluation of Machine Learning Algorithms for Classification of EEG Signals

In brain–computer interfaces (BCIs), it is crucial to process brain signals to improve the accuracy of the classification of motor movements. Machine learning (ML) algorithms such as artificial neural networks (ANNs), linear discriminant analysis (LDA), decision tree (D.T.), K-nearest neighbor (KNN), naive Bayes (N.B.), and support vector machine (SVM) have made significant progress in classification issues. This paper aims to present a signal processing analysis of electroencephalographic (EEG) signals among different feature extraction techniques to train selected classification algorithms to classify signals related to motor movements. The motor movements considered are related to the left hand, right hand, both fists, feet, and relaxation, making this a multiclass problem. In this study, nine ML algorithms were trained with a dataset created by the feature extraction of EEG signals.The EEG signals of 30 Physionet subjects were used to create a dataset related to movement. We used electrodes C3, C1, CZ, C2, and C4 according to the standard 10-10 placement. Then, we extracted the epochs of the EEG signals and applied tone, amplitude levels, and statistical techniques to obtain the set of features. LabVIEW™2015 version custom applications were used for reading the EEG signals; for channel selection, noise filtering, band selection, and feature extraction operations; and for creating the dataset. MATLAB 2021a was used for training, testing, and evaluating the performance metrics of the ML algorithms. In this study, the model of Medium-ANN achieved the best performance, with an AUC average of 0.9998, Cohen’s Kappa coefficient of 0.9552, a Matthews correlation coefficient of 0.9819, and a loss of 0.0147. These findings suggest the applicability of our approach to different scenarios, such as implementing robotic prostheses, where the use of superficial features is an acceptable option when resources are limited, as in embedded systems or edge computing devices.

EMD-Based Method for Supervised Classification of Parkinson’s Disease Patients Using Balance Control Data

There has recently been increasing interest in postural stability aimed at gaining a better understanding of the human postural system. This system controls human balance in quiet standing and during locomotion. Parkinson’s disease (PD) is the most common degenerative movement disorder that affects human stability and causes falls and injuries. This paper proposes a novel methodology to differentiate between healthy individuals and those with PD through the empirical mode decomposition (EMD) method. EMD enables the breaking down of a complex signal into several elementary signals called intrinsic mode functions (IMFs). Three temporal parameters and three spectral parameters are extracted from each stabilometric signal as well as from its IMFs. Next, the best five features are selected using the feature selection method. The classification task is carried out using four known machine-learning methods, KNN, decision tree, Random Forest and SVM classifiers, over 10-fold cross validation. The used dataset consists of 28 healthy subjects (14 young adults and 14 old adults) and 32 PD patients (12 young adults and 20 old adults). The SVM method has a performance of 92% and the Dempster–Sahfer formalism method has an accuracy of 96.51%.

Vowel speech recognition from rat electroencephalography using long short-term memory neural network

Over the years, considerable research has been conducted to investigate the mechanisms of speech perception and recognition. Electroencephalography (EEG) is a powerful tool for identifying brain activity; therefore, it has been widely used to determine the neural basis of speech recognition. In particular, for the classification of speech recognition, deep learning-based approaches are in the spotlight because they can automatically learn and extract representative features through end-to-end learning. This study aimed to identify particular components that are potentially related to phoneme representation in the rat brain and to discriminate brain activity for each vowel stimulus on a single-trial basis using a bidirectional long short-term memory (BiLSTM) network and classical machine learning methods. Nineteen male Sprague-Dawley rats subjected to microelectrode implantation surgery to record EEG signals from the bilateral anterior auditory fields were used. Five different vowel speech stimuli were chosen, /a/, /e/, /i/, /o/, and /u/, which have highly different formant frequencies. EEG recorded under randomly given vowel stimuli was minimally preprocessed and normalized by a z-score transformation to be used as input for the classification of speech recognition. The BiLSTM network showed the best performance among the classifiers by achieving an overall accuracy, f1-score, and Cohen’s κ values of 75.18%, 0.75, and 0.68, respectively, using a 10-fold cross-validation approach. These results indicate that LSTM layers can effectively model sequential data, such as EEG; hence, informative features can be derived through BiLSTM trained with end-to-end learning without any additional hand-crafted feature extraction methods.

Using LinkedIn Endorsements to Reinforce an Ontology and Machine Learning-Based Recommender System to Improve Professional Skills

Nowadays, social networks have become highly relevant in the professional field, in terms of the possibility of sharing profiles, skills and jobs. LinkedIn has become the social network par excellence, owing to its content in professional and training information and where there are also endorsements, which are validations of the skills of users that can be taken into account in the recruitment process, as well as in the recommender system. In order to determine how endorsements influence Lifelong Learning course recommendations for professional skills development and enhancement, a new version of our Lifelong Learning course recommendation system is proposed. The recommender system is based on ontology, which allows modelling the data of knowledge areas and job performance sectors to represent professional skills of users obtained from social networks. Machine learning techniques are applied to group entities in the ontology and make predictions of new data. The recommender system has a semantic core, content-based filtering, and heuristics to perform the formative suggestion. In order to validate the data model and test the recommender system, information was obtained from web-based lifelong learning courses and information was collected from LinkedIn professional profiles, incorporating the skills endorsements into the user profile. All possible settings of the system were tested. The best result was obtained in the setting based on the spatial clustering algorithm based on the density of noisy applications. An accuracy of 94% and 80% recall was obtained.

In sync with your child: The potential of parent-child electroencephalography in developmental research

Healthy interaction between parent and child is foundational for the child's socioemotional development. Recently, an innovative paradigm shift in electroencephalography (EEG) research has enabled the simultaneous measurement of neural activity in caregiver and child. This dual-EEG or hyperscanning approach, termed parent-child dual-EEG, combines the strength of both behavioral observations and EEG methods. In this review, we aim to inform on the potential of dual-EEG in parents and children (0-6 years) for developmental researchers. We first provide a general overview of the dual-EEG technique and continue by reviewing the first empirical work on the emerging field of parent-child dual-EEG, discussing the limited but fascinating findings on parent-child brain-to-behavior and brain-to-brain synchrony. We then continue by providing an overview of dual-EEG analysis techniques, including the technical challenges and solutions one may encounter. We finish by discussing the potential of parent-child dual-EEG for the future of developmental research. The analysis of multiple EEG data is technical and challenging, but when performed well, parent-child EEG may transform the way we understand how caregiver and child connect on a neurobiological level. Importantly, studying objective physiological measures of parent-child interactions could lead to the identification of novel brain-to-brain synchrony markers of interaction quality.

Validation of a Light EEG-Based Measure for Real-Time Stress Monitoring during Realistic Driving

Driver’s stress affects decision-making and the probability of risk occurrence, and it is therefore a key factor in road safety. This suggests the need for continuous stress monitoring. This work aims at validating a stress neurophysiological measure—a Neurometric—for out-of-the-lab use obtained from lightweight EEG relying on two wet sensors, in real-time, and without calibration. The Neurometric was tested during a multitasking experiment and validated with a realistic driving simulator. Twenty subjects participated in the experiment, and the resulting stress Neurometric was compared with the Random Forest (RF) model, calibrated by using EEG features and both intra-subject and cross-task approaches. The Neurometric was also compared with a measure based on skin conductance level (SCL), representing one of the physiological parameters investigated in the literature mostly correlated with stress variations. We found that during both multitasking and realistic driving experiments, the Neurometric was able to discriminate between low and high levels of stress with an average Area Under Curve (AUC) value higher than 0.9. Furthermore, the stress Neurometric showed higher AUC and stability than both the SCL measure and the RF calibrated with a cross-task approach. In conclusion, the Neurometric proposed in this work proved to be suitable for out-of-the-lab monitoring of stress levels.

A neurophysiologically interpretable deep neural network predicts complex movement components from brain activity

The effective decoding of movement from non-invasive electroencephalography (EEG) is essential for informing several therapeutic interventions, from neurorehabilitation robots to neural prosthetics. Deep neural networks are most suitable for decoding real-time data but their use in EEG is hindered by the gross classes of motor tasks in the currently available datasets, which are solvable even with network architectures that do not require specialized design considerations. Moreover, the weak association with the underlying neurophysiology limits the generalizability of modern networks for EEG inference. Here, we present a neurophysiologically interpretable 3-dimensional convolutional neural network (3D-CNN) that captured the spatiotemporal dependencies in brain areas that get co-activated during movement. The 3D-CNN received topography-preserving EEG inputs, and predicted complex components of hand movements performed on a plane using a back-drivable rehabilitation robot, namely (a) the reaction time (RT) for responding to stimulus (slow or fast), (b) the mode of movement (active or passive, depending on whether there was an assistive force provided by the apparatus), and (c) the orthogonal directions of the movement (left, right, up, or down). We validated the 3D-CNN on a new dataset that we acquired from an in-house motor experiment, where it achieved average leave-one-subject-out test accuracies of 79.81%, 81.23%, and 82.00% for RT, active vs. passive, and direction classifications, respectively. Our proposed method outperformed the modern 2D-CNN architecture by a range of 1.1% to 6.74% depending on the classification task. Further, we identified the EEG sensors and time segments crucial to the classification decisions of the network, which aligned well with the current neurophysiological knowledge on brain activity in motor planning and execution tasks. Our results demonstrate the importance of biological relevance in networks for an accurate decoding of EEG, suggesting that the real-time classification of other complex brain activities may now be within our reach.

Deep Learning for Autism Diagnosis and Facial Analysis in Children

In this paper, we introduce a deep learning model to classify children as either healthy or potentially having autism with 94.6% accuracy using Deep Learning. Patients with autism struggle with social skills, repetitive behaviors, and communication, both verbal and non-verbal. Although the disease is considered to be genetic, the highest rates of accurate diagnosis occur when the child is tested on behavioral characteristics and facial features. Patients have a common pattern of distinct facial deformities, allowing researchers to analyze only an image of the child to determine if the child has the disease. While there are other techniques and models used for facial analysis and autism classification on their own, our proposal bridges these two ideas allowing classification in a cheaper, more efficient method. Our deep learning model uses MobileNet and two dense layers to perform feature extraction and image classification. The model is trained and tested using 3,014 images, evenly split between children with autism and children without it; 90% of the data is used for training and 10% is used for testing. Based on our accuracy, we propose that the diagnosis of autism can be done effectively using only a picture. Additionally, there may be other diseases that are similarly diagnosable.

A BCI-Based Study on the Relationship Between the SSVEP and Retinal Eccentricity in Overt and Covert Attention

Covert attention aids us in monitoring the environment and optimizing performance in visual tasks. Past behavioral studies have shown that covert attention can enhance spatial resolution. However, electroencephalography (EEG) activity related to neural processing between central and peripheral vision has not been systematically investigated. Here, we conducted an EEG study with 25 subjects who performed covert attentional tasks at different retinal eccentricities ranging from 0.75° to 13.90°, as well as tasks involving overt attention and no attention. EEG signals were recorded with a single stimulus frequency to evoke steady-state visual evoked potentials (SSVEPs) for attention evaluation. We found that the SSVEP response in fixating at the attended location was generally negatively correlated with stimulus eccentricity as characterized by Euclidean distance or horizontal and vertical distance. Moreover, more pronounced characteristics of SSVEP analysis were also acquired in overt attention than in covert attention. Furthermore, offline classification of overt attention, covert attention, and no attention yielded an average accuracy of 91.42%. This work contributes to our understanding of the SSVEP representation of attention in humans and may also lead to brain-computer interfaces (BCIs) that allow people to communicate with choices simply by shifting their attention to them.

Pilot Study on Analysis of Electroencephalography Signals from Children with FASD with the Implementation of Naive Bayesian Classifiers

In this paper Naive Bayesian classifiers were applied for the purpose of differentiation between the EEG signals recorded from children with Fetal Alcohol Syndrome Disorders (FASD) and healthy ones. This work also provides a brief introduction to the FASD itself, explaining the social, economic and genetic reasons for the FASD occurrence. The obtained results were good and promising and indicate that EEG recordings can be a helpful tool for potential diagnostics of FASDs children affected with it, in particular those with invisible physical signs of these spectrum disorders.

Foundations of Time Series Analysis

For almost a century, classical statistical methods including exponential smoothing and autoregression integrated moving averages (ARIMA) have been predominant in the analysis of time series (TS) and in the pursuit of forecasting future events from historical data. TS are chronological sequences of observations, and TS data are therefore prevalent in many aspects of clinical medicine and academic neuroscience. With the rise of highly complex and nonlinear datasets, machine learning (ML) methods have become increasingly popular for prediction or pattern detection and within neurosciences, including neurosurgery. ML methods regularly outperform classical methods and have been successfully applied to, inter alia, predict physiological responses in intracranial pressure monitoring or to identify seizures in EEGs. Implementing nonparametric methods for TS analysis in clinical practice can benefit clinical decision making and sharpen our diagnostic armory.

The emergence of machine learning in auditory neural impairment: A systematic review

Hearing loss is a common neurodegenerative disease that can start at any stage of life. Misalignment of the auditory neural impairment may impose challenges in processing incoming auditory stimulus that can be measured using electroencephalography (EEG). The electrophysiological behaviour response emanated from EEG auditory evoked potential (AEP) requires highly trained professionals for analysis and interpretation. Reliable automated methods using techniques of machine learning would assist the auditory assessment process for informed treatment and practice. It is thus highly required to develop models that are more efficient and precise by considering the characteristics of brain signals. This study aims to provide a comprehensive review of several state-of-the-art techniques of machine learning that adopt EEG evoked response for the auditory assessment within the last 13 years. Out of 161 initially screened articles, 11 were retained for synthesis. The outcome of the review presented that the Support Vector Machine (SVM) classifier outperformed with over 80% accuracy metric and was recognized as the best suited model within the field of auditory research. This paper discussed the comprehensive iterative properties of the proposed computed algorithms and the feasible future direction in hearing impaired rehabilitation.

Neural Decoding of EEG Signals with Machine Learning: A Systematic Review

Electroencephalography (EEG) is a non-invasive technique used to record the brain's evoked and induced electrical activity from the scalp. Artificial intelligence, particularly machine learning (ML) and deep learning (DL) algorithms, are increasingly being applied to EEG data for pattern analysis, group membership classification, and brain-computer interface purposes. This study aimed to systematically review recent advances in ML and DL supervised models for decoding and classifying EEG signals. Moreover, this article provides a comprehensive review of the state-of-the-art techniques used for EEG signal preprocessing and feature extraction. To this end, several academic databases were searched to explore relevant studies from the year 2000 to the present. Our results showed that the application of ML and DL in both mental workload and motor imagery tasks has received substantial attention in recent years. A total of 75% of DL studies applied convolutional neural networks with various learning algorithms, and 36% of ML studies achieved competitive accuracy by using a support vector machine algorithm. Wavelet transform was found to be the most common feature extraction method used for all types of tasks. We further examined the specific feature extraction methods and end classifier recommendations discovered in this systematic review.

Recognition of EEG Signals from Imagined Vowels Using Deep Learning Methods

The use of imagined speech with electroencephalographic (EEG) signals is a promising field of brain-computer interfaces (BCI) that seeks communication between areas of the cerebral cortex related to language and devices or machines. However, the complexity of this brain process makes the analysis and classification of this type of signals a relevant topic of research. The goals of this study were: to develop a new algorithm based on Deep Learning (DL), referred to as CNNeeg1-1, to recognize EEG signals in imagined vowel tasks; to create an imagined speech database with 50 subjects specialized in imagined vowels from the Spanish language (/a/,/e/,/i/,/o/,/u/); and to contrast the performance of the CNNeeg1-1 algorithm with the DL Shallow CNN and EEGNet benchmark algorithms using an open access database (BD1) and the newly developed database (BD2). In this study, a mixed variance analysis of variance was conducted to assess the intra-subject and inter-subject training of the proposed algorithms. The results show that for intra-subject training analysis, the best performance among the Shallow CNN, EEGNet, and CNNeeg1-1 methods in classifying imagined vowels (/a/,/e/,/i/,/o/,/u/) was exhibited by CNNeeg1-1, with an accuracy of 65.62% for BD1 database and 85.66% for BD2 database.

Heart Rate Variability and Clinical Features as Predictors of Atrial Fibrillation Recurrence After Catheter Ablation: A Pilot Study

Single-procedure catheter ablation success rate is as low as 52% in atrial fibrillation (AF) patients. This study evaluated the feasibility of using clinical data and heart rate variability (HRV) features extracted from an implantable cardiac monitor (ICM) to predict recurrences in patients prior to undergoing catheter ablation for AF. HRV-derived features were extracted from the 500 beats preceding the AF onset and from the first 2 min of the last AF episode recorded by an ICM of 74 patients (67% male; 57 ± 12 years; 26% non-paroxysmal AF; 57% AF recurrence) before undergoing their first AF catheter ablation. Two types of classification algorithm were studied to predict AF recurrence: single classifiers including support vector machines, classification and regression trees, and K-nearest neighbor classifiers as well as ensemble classifiers. The sequential forward floating search algorithm was used to select the optimum feature set for each classification method. The optimum weighted voting method, which used an optimum combination of the single classifiers, was the best overall classifier (accuracy = 0.82, sensitivity = 0.76, and specificity = 0.87). Clinical and HRV features can be used to predict rhythm outcome using an ensemble classifier which would enable a more effective pre-ablation patient triage that could reduce the economic and personal burden of the procedure by increasing the success rate of first catheter ablation.