On the role of generative artificial intelligence in the development of brain-computer interfaces

Since their inception more than 50 years ago, Brain-Computer Interfaces (BCIs) have held promise to compensate for functions lost by people with disabilities through allowing direct communication between the brain and external devices. While research throughout the past decades has demonstrated the feasibility of BCI to act as a successful assistive technology, the widespread use of BCI outside the lab is still beyond reach. This can be attributed to a number of challenges that need to be addressed for BCI to be of practical use including limited data availability, limited temporal and spatial resolutions of brain signals recorded non-invasively and inter-subject variability. In addition, for a very long time, BCI development has been mainly confined to specific simple brain patterns, while developing other BCI applications relying on complex brain patterns has been proven infeasible. Generative Artificial Intelligence (GAI) has recently emerged as an artificial intelligence domain in which trained models can be used to generate new data with properties resembling that of available data. Given the enhancements observed in other domains that possess similar challenges to BCI development, GAI has been recently employed in a multitude of BCI development applications to generate synthetic brain activity; thereby, augmenting the recorded brain activity. Here, a brief review of the recent adoption of GAI techniques to overcome the aforementioned BCI challenges is provided demonstrating the enhancements achieved using GAI techniques in augmenting limited EEG data, enhancing the spatiotemporal resolution of recorded EEG data, enhancing cross-subject performance of BCI systems and implementing end-to-end BCI applications. GAI could represent the means by which BCI would be transformed into a prevalent assistive technology, thereby improving the quality of life of people with disabilities, and helping in adopting BCI as an emerging human-computer interaction technology for general use.

A review of signal processing and machine learning techniques for interictal epileptiform discharge detection

Brain interictal epileptiform discharges (IEDs), as one of the hallmarks of epileptic brain, are transient events captured by electroencephalogram (EEG). IEDs are generated by seizure networks, and they occur between seizures (interictal periods). The development of a robust method for IED detection could be highly informative for clinical treatment procedures and epileptic patient management. Since 1972, different machine learning techniques, from template matching to deep learning, have been developed to automatically detect IEDs from scalp EEG (scEEG) and intracranial EEG (iEEG). While the scEEG signals suffer from low information details and high attenuation of IEDs due to the high skull electrical impedance, the iEEG signals recorded using implanted electrodes enjoy higher details and are more suitable for identifying the IEDs. In this review paper, we group IED detection techniques into six categories: (1) template matching, (2) feature representation (mimetic, time-frequency, and nonlinear features), (3) matrix decomposition, (4) tensor factorization, (5) neural networks, and (6) estimation of the iEEG from the concurrent scEEG followed by detection and classification. The methods are compared quantitatively (e.g., in terms of accuracy, sensitivity, and specificity), and their general advantages and limitations are described. Finally, current limitations and possible future research paths related to this field are mentioned.

Status of deep learning for EEG-based brain-computer interface applications

In the previous decade, breakthroughs in the central nervous system bioinformatics and computational innovation have prompted significant developments in brain-computer interface (BCI), elevating it to the forefront of applied science and research. BCI revitalization enables neurorehabilitation strategies for physically disabled patients (e.g., disabled patients and hemiplegia) and patients with brain injury (e.g., patients with stroke). Different methods have been developed for electroencephalogram (EEG)-based BCI applications. Due to the lack of a large set of EEG data, methods using matrix factorization and machine learning were the most popular. However, things have changed recently because a number of large, high-quality EEG datasets are now being made public and used in deep learning-based BCI applications. On the other hand, deep learning is demonstrating great prospects for solving complex relevant tasks such as motor imagery classification, epileptic seizure detection, and driver attention recognition using EEG data. Researchers are doing a lot of work on deep learning-based approaches in the BCI field right now. Moreover, there is a great demand for a study that emphasizes only deep learning models for EEG-based BCI applications. Therefore, we introduce this study to the recent proposed deep learning-based approaches in BCI using EEG data (from 2017 to 2022). The main differences, such as merits, drawbacks, and applications are introduced. Furthermore, we point out current challenges and the directions for future studies. We argue that this review study will help the EEG research community in their future research.

E2SGAN: EEG-to-SEEG translation with generative adversarial networks

High-quality brain signal data recorded by Stereoelectroencephalography (SEEG) electrodes provide clinicians with clear guidance for presurgical assessments for epilepsy surgeries. SEEG, however, is limited to selected patients with epilepsy due to its invasive procedure. In this work, a brain signal synthesis framework is presented to synthesize SEEG signals from non-invasive EEG signals. First, a strategy to determine the matching relation between EEG and SEEG channels is presented by considering both signal correlation and spatial distance. Second, the EEG-to-SEEG generative adversarial network (E2SGAN) is proposed to precisely synthesize SEEG data from the simultaneous EEG data. Although the widely adopted magnitude spectra has proved to be informative in EEG tasks, it leaves much to be desired in the setting of signal synthesis. To this end, instantaneous frequency spectra is introduced to further represent the alignment of the signal. Correlative spectral attention (CSA) is proposed to enhance the discriminator of E2SGAN by capturing the correlation between each pair of EEG and SEEG frequencies. The weighted patch prediction (WPP) technique is devised to ensure robust temporal results. Comparison experiments on real-patient data demonstrate that E2SGAN outperforms baseline methods in both temporal and frequency domains. The perturbation experiment reveals that the synthesized results have the potential to capture abnormal discharges in epileptic patients before seizures.

Phase-amplitude coupling between low-frequency scalp EEG and high-frequency intracranial EEG during working memory task

Objective. Revealing the relationship between simultaneous scalp electroencephalography (EEG) and intracranial electroencephalography (iEEG) is of great importance for both neuroscientific research and translational applications. However, whether prominent iEEG features in the high-gamma band can be reflected by scalp EEG is largely unknown. To address this, we investigated the phase-amplitude coupling (PAC) phenomenon between the low-frequency band of scalp EEG and the high-gamma band of iEEG.Approach. We analyzed a simultaneous iEEG and scalp EEG dataset acquired under a verbal working memory paradigm from nine epilepsy subjects. The PAC values between pairs of scalp EEG channel and identified iEEG channel were explored. After identifying the frequency combinations and electrode locations that generated the most significant PAC values, we compared the PAC values of different task periods (encoding, maintenance, and retrieval) and memory loads.Main results. We demonstrated that the amplitude of high-gamma activities in the entorhinal cortex, hippocampus, and amygdala was correlated to the delta or theta phase at scalp locations such as Cz and Pz. In particular, the frequency bin that generated the maximum PAC value centered at 3.16-3.84 Hz for the phase and 50-85 Hz for the amplitude. Moreover, our results showed that PAC values for the retrieval period were significantly higher than those of the encoding and maintenance periods, and the PAC was also influenced by the memory load.Significance. This is the first human simultaneous iEEG and scalp EEG study demonstrating that the amplitude of iEEG high-gamma components is associated with the phase of low-frequency components in scalp EEG. These findings enhance our understanding of multiscale neural interactions during working memory, and meanwhile, provide a new perspective to estimate intracranial high-frequency features with non-invasive neural recordings.

Higher-order tensor decomposition based scalp-to-intracranial EEG projection for detection of interictal epileptiform discharges

Objective.Interictal epileptiform discharges (IEDs) occur between two seizures onsets. IEDs are mainly captured by intracranial recordings and are often invisible over the scalp. This study proposes a model based on tensor factorization to map the time-frequency (TF) features of scalp EEG (sEEG) to the TF features of intracranial EEG (iEEG) in order to detect IEDs from over the scalp with high sensitivity.Approach.Continuous wavelet transform is employed to extract the TF features. Time, frequency, and channel modes of IED segments from iEEG recordings are concatenated into a four-way tensor. Tucker and CANDECOMP/PARAFAC decomposition techniques are employed to decompose the tensor into temporal, spectral, spatial, and segmental factors. Finally, TF features of both IED and non-IED segments from scalp recordings are projected onto the temporal components for classification.Main results.The model performance is obtained in two different approaches: within- and between-subject classification approaches. Our proposed method is compared with four other methods, namely a tensor-based spatial component analysis method, TF-based method, linear regression mapping model, and asymmetric-symmetric autoencoder mapping model followed by convolutional neural networks. Our proposed method outperforms all these methods in both within- and between-subject classification approaches by respectively achieving 84.2% and 72.6% accuracy values.Significance.The findings show that mapping sEEG to iEEG improves the performance of the scalp-based IED detection model. Furthermore, the tensor-based mapping model outperforms the autoencoder- and regression-based mapping models.

A Multi-Scale Activity Transition Network for Data Translation in EEG Signals Decoding

Electroencephalogram (EEG) is a non-invasive collection method for brain signals. It has broad prospects in brain-computer interface (BCI) applications. Recent advances have shown the effectiveness of the widely used convolutional neural network (CNN) in EEG decoding. However, some studies reveal that a slight disturbance to the inputs, e.g., data translation, can change CNN's outputs. Such instability is dangerous for EEG-based BCI applications because signals in practice are different from training data. In this study, we propose a multi-scale activity transition network (MSATNet) to alleviate the influence of the translation problem in convolution-based models. MSATNet provides an activity state pyramid consisting of multi-scale recurrent neural networks to capture the relationship between brain activities, which is a translation-invariant feature. In the experiment, Kullback-Leibler divergence is applied to measure the degree of translation. The comprehensive results demonstrate that our method surpasses the AUC of 0.0080, 0.0254, 0.0393 in 1, 5, and 10 KL divergence compared to competitors with various convolution structures.

Time-Frequency Decomposition of Scalp Electroencephalograms Improves Deep Learning-Based Epilepsy Diagnosis

Epilepsy diagnosis based on Interictal Epileptiform Discharges (IEDs) in scalp electroencephalograms (EEGs) is laborious and often subjective. Therefore, it is necessary to build an effective IED detector and an automatic method to classify IED-free versus IED EEGs. In this study, we evaluate features that may provide reliable IED detection and EEG classification. Specifically, we investigate the IED detector based on convolutional neural network (ConvNet) with different input features (temporal, spectral, and wavelet features). We explore different ConvNet architectures and types, including 1D (one-dimensional) ConvNet, 2D (two-dimensional) ConvNet, and noise injection at various layers. We evaluate the EEG classification performance on five independent datasets. The 1D ConvNet with preprocessed full-frequency EEG signal and frequency bands (delta, theta, alpha, beta) with Gaussian additive noise at the output layer achieved the best IED detection results with a false detection rate of 0.23/min at 90% sensitivity. The EEG classification system obtained a mean EEG classification Leave-One-Institution-Out (LOIO) cross-validation (CV) balanced accuracy (BAC) of 78.1% (area under the curve (AUC) of 0.839) and Leave-One-Subject-Out (LOSO) CV BAC of 79.5% (AUC of 0.856). Since the proposed classification system only takes a few seconds to analyze a 30-min routine EEG, it may help in reducing the human effort required for epilepsy diagnosis.

A deep learning algorithm based on 1D CNN-LSTM for automatic sleep staging

BACKGROUND

Sleep staging is an important part of sleep research. Traditional automatic sleep staging based on machine learning requires extensive feature extraction and selection.

OBJECTIVE

This paper proposed a deep learning algorithm without feature extraction based on one-dimensional convolutional neural network and long short-term memory.

METHODS

The algorithm can automatically divide sleep into 5 phases including awake period, non-rapid eye movement sleep period (N1 ∼ N3) and rapid eye movement using the electroencephalogram signals. The raw signal was processed by the wavelet transform. Then, the processed signal was directly input into the deep learning algorithm to obtain the staging result.

RESULTS

The accuracy of staging is 93.47% using the Fpz-Cz electroencephalogram signal. When using the Fpz-Cz and electroencephalogram signal, the algorithm can obtain the highest accuracy of 94.15%.

CONCLUSION

These results show that this algorithm is suitable for different physiological signals and can realize end-to-end automatic sleep staging without any manual feature extraction.

Epileptic Seizures Detection Using Deep Learning Techniques: A Review

A variety of screening approaches have been proposed to diagnose epileptic seizures, using electroencephalography (EEG) and magnetic resonance imaging (MRI) modalities. Artificial intelligence encompasses a variety of areas, and one of its branches is deep learning (DL). Before the rise of DL, conventional machine learning algorithms involving feature extraction were performed. This limited their performance to the ability of those handcrafting the features. However, in DL, the extraction of features and classification are entirely automated. The advent of these techniques in many areas of medicine, such as in the diagnosis of epileptic seizures, has made significant advances. In this study, a comprehensive overview of works focused on automated epileptic seizure detection using DL techniques and neuroimaging modalities is presented. Various methods proposed to diagnose epileptic seizures automatically using EEG and MRI modalities are described. In addition, rehabilitation systems developed for epileptic seizures using DL have been analyzed, and a summary is provided. The rehabilitation tools include cloud computing techniques and hardware required for implementation of DL algorithms. The important challenges in accurate detection of automated epileptic seizures using DL with EEG and MRI modalities are discussed. The advantages and limitations in employing DL-based techniques for epileptic seizures diagnosis are presented. Finally, the most promising DL models proposed and possible future works on automated epileptic seizure detection are delineated.

Expert-Level Intracranial Electroencephalogram Ictal Pattern Detection by a Deep Learning Neural Network

Background: Decision-making in epilepsy surgery is strongly connected to the interpretation of the intracranial EEG (iEEG). Although deep learning approaches have demonstrated efficiency in processing extracranial EEG, few studies have addressed iEEG seizure detection, in part due to the small number of seizures per patient typically available from intracranial investigations. This study aims to evaluate the efficiency of deep learning methodology in detecting iEEG seizures using a large dataset of ictal patterns collected from epilepsy patients implanted with a responsive neurostimulation system (RNS). Methods: Five thousand two hundred and twenty-six ictal events were collected from 22 patients implanted with RNS. A convolutional neural network (CNN) architecture was created to provide personalized seizure annotations for each patient. Accuracy of seizure identification was tested in two scenarios: patients with seizures occurring following a period of chronic recording (scenario 1) and patients with seizures occurring immediately following implantation (scenario 2). The accuracy of the CNN in identifying RNS-recorded iEEG ictal patterns was evaluated against human neurophysiology expertise. Statistical performance was assessed via the area-under-precision-recall curve (AUPRC). Results: In scenario 1, the CNN achieved a maximum mean binary classification AUPRC of 0.84 ± 0.19 (95%CI, 0.72-0.93) and mean regression accuracy of 6.3 ± 1.0 s (95%CI, 4.3-8.5 s) at 30 seed samples. In scenario 2, maximum mean AUPRC was 0.80 ± 0.19 (95%CI, 0.68-0.91) and mean regression accuracy was 6.3 ± 0.9 s (95%CI, 4.8-8.3 s) at 20 seed samples. We obtained near-maximum accuracies at seed size of 10 in both scenarios. CNN classification failures can be explained by ictal electro-decrements, brief seizures, single-channel ictal patterns, highly concentrated interictal activity, changes in the sleep-wake cycle, and progressive modulation of electrographic ictal features. Conclusions: We developed a deep learning neural network that performs personalized detection of RNS-derived ictal patterns with expert-level accuracy. These results suggest the potential for automated techniques to significantly improve the management of closed-loop brain stimulation, including during the initial period of recording when the device is otherwise naïve to a given patient's seizures.

Incorporating Uncertainty in Data Labeling into Automatic Detection of Interictal Epileptiform Discharges from Concurrent Scalp-EEG via Multi-way Analysis

Interictal epileptiform discharges (IEDs) are elicited from an epileptic brain, whereas they can also be due to other neurological abnormalities. The diversity in their morphologies, their strengths, and their sources within the brain cause a great deal of uncertainty in their labeling by clinicians. The aim of this study is therefore to exploit and incorporate this uncertainty (the probability of the waveform being an IED) in the IED detection system which combines spatial component analysis (SCA) with the IED probabilities referred to as SCA-IEDP-based method. For comparison, we also propose and study SCA-based method in which probability of the waveform being an IED is ignored. The proposed models are employed to detect IEDs in two different classification approaches: (1) subject-dependent and (2) subject-independent classification approaches. The proposed methods are compared with two other state-of-the-art methods namely, time-frequency features and tensor factorization methods. The proposed SCA-IEDP model has achieved superior performance in comparison with the traditional SCA and other competing methods. It achieved 79.9% and 63.4% accuracy values in subject-dependent and subject-independent classification approaches, respectively. This shows that considering the IED probabilities in designing an IED detection system can boost its performance.

A Review on Machine Learning for EEG Signal Processing in Bioengineering

Electroencephalography (EEG) has been a staple method for identifying certain health conditions in patients since its discovery. Due to the many different types of classifiers available to use, the analysis methods are also equally numerous. In this review, we will be examining specifically machine learning methods that have been developed for EEG analysis with bioengineering applications. We reviewed literature from 1988 to 2018 to capture previous and current classification methods for EEG in multiple applications. From this information, we are able to determine the overall effectiveness of each machine learning method as well as the key characteristics. We have found that all the primary methods used in machine learning have been applied in some form in EEG classification. This ranges from Naive-Bayes to Decision Tree/Random Forest, to Support Vector Machine (SVM). Supervised learning methods are on average of higher accuracy than their unsupervised counterparts. This includes SVM and KNN. While each of the methods individually is limited in their accuracy in their respective applications, there is hope that the combination of methods when implemented properly has a higher overall classification accuracy. This paper provides a comprehensive overview of Machine Learning applications used in EEG analysis. It also gives an overview of each of the methods and general applications that each is best suited to.

A survey on deep learning-based non-invasive brain signals: recent advances and new frontiers

Brain signals refer to the biometric information collected from the human brain. The research on brain signals aims to discover the underlying neurological or physical status of the individuals by signal decoding. The emerging deep learning techniques have improved the study of brain signals significantly in recent years. In this work, we first present a taxonomy of non-invasive brain signals and the basics of deep learning algorithms. Then, we provide a comprehensive survey of the frontiers of applying deep learning for non-invasive brain signals analysis, by summarizing a large number of recent publications. Moreover, upon the deep learning-powered brain signal studies, we report the potential real-world applications which benefit not only disabled people but also normal individuals. Finally, we discuss the opening challenges and future directions.

A Neural Network for Image Anomaly Detection with Deep Pyramidal Representations and Dynamic Routing

Image anomaly detection is an application-driven problem where the aim is to identify novel samples, which differ significantly from the normal ones. We here propose Pyramidal Image Anomaly DEtector (PIADE), a deep reconstruction-based pyramidal approach, in which image features are extracted at different scale levels to better catch the peculiarities that could help to discriminate between normal and anomalous data. The features are dynamically routed to a reconstruction layer and anomalies can be identified by comparing the input image with its reconstruction. Unlike similar approaches, the comparison is done by using structural similarity and perceptual loss rather than trivial pixel-by-pixel comparison. The proposed method performed at par or better than the state-of-the-art methods when tested on publicly available datasets such as CIFAR10, COIL-100 and MVTec.

A Swallowing Decoder Based on Deep Transfer Learning: AlexNet Classification of the Intracranial Electrocorticogram

To realize a brain-machine interface to assist swallowing, neural signal decoding is indispensable. Eight participants with temporal-lobe intracranial electrode implants for epilepsy were asked to swallow during electrocorticogram (ECoG) recording. Raw ECoG signals or certain frequency bands of the ECoG power were converted into images whose vertical axis was electrode number and whose horizontal axis was time in milliseconds, which were used as training data. These data were classified with four labels (Rest, Mouth open, Water injection, and Swallowing). Deep transfer learning was carried out using AlexNet, and power in the high-[Formula: see text] band (75-150[Formula: see text]Hz) was the training set. Accuracy reached 74.01%, sensitivity reached 82.51%, and specificity reached 95.38%. However, using the raw ECoG signals, the accuracy obtained was 76.95%, comparable to that of the high-[Formula: see text] power. We demonstrated that a version of AlexNet pre-trained with visually meaningful images can be used for transfer learning of visually meaningless images made up of ECoG signals. Moreover, we could achieve high decoding accuracy using the raw ECoG signals, allowing us to dispense with the conventional extraction of high-[Formula: see text] power. Thus, the images derived from the raw ECoG signals were equivalent to those derived from the high-[Formula: see text] band for transfer deep learning.

Machine learning (ML) for the diagnosis of autism spectrum disorder (ASD) using brain imaging

Autism spectrum disorder (ASD) is a neurodevelopmental incurable disorder with a long diagnostic period encountered in the early years of life. If diagnosed early, the negative effects of this disease can be reduced by starting special education early. Machine learning (ML), an increasingly ubiquitous technology, can be applied for the early diagnosis of ASD. The aim of this study is to examine and provide a comprehensive state-of-the-art review of ML research for the diagnosis of ASD based on (a) structural magnetic resonance image (MRI), (b) functional MRI and (c) hybrid imaging techniques over the past decade. The accuracy of the studies with a large number of participants is in general lower than those with fewer participants leading to the conclusion that further large-scale studies are needed. An examination of the age of the participants shows that the accuracy of the automated diagnosis of ASD is higher at a younger age range. ML technology is expected to contribute significantly to the early and rapid diagnosis of ASD in the coming years and become available to clinicians in the near future. This review is aimed to facilitate that.

Determination of Antiepileptic Drugs Withdrawal Through EEG Hjorth Parameter Analysis

The decision to continue or to stop antiepileptic drug (AED) treatment in patients with prolonged seizure remission is a critical issue. Previous studies have used certain risk factors or electroencephalogram (EEG) findings to predict seizure recurrence after the withdrawal of AEDs. However, validated biomarkers to guide the withdrawal of AEDs are lacking. In this study, we used quantitative EEG analysis to establish a method for predicting seizure recurrence after the withdrawal of AEDs. A total of 34 patients with epilepsy were divided into two groups, 17 patients in the recurrence group and the other 17 patients in the nonrecurrence group. All patients were seizure free for at least two years. Before AED withdrawal, an EEG was performed for each patient that showed no epileptiform discharges. These EEG recordings were classified using Hjorth parameter-based EEG features. We found that the Hjorth complexity values were higher in patients in the recurrence group than in the nonrecurrence group. The extreme gradient boosting classification method achieved the highest performance in terms of accuracy, area under the curve, sensitivity, and specificity (84.76%, 88.77%, 89.67%, and 80.47%, respectively). Our proposed method is a promising tool to help physicians determine AED withdrawal for seizure-free patients.

Morphological Characterization of Functional Brain Imaging by Isosurface Analysis in Parkinson’s Disease

Finding new biomarkers to model Parkinson’s Disease (PD) is a challenge not only to help discerning between Healthy Control (HC) subjects and patients with potential PD but also as a way to measure quantitatively the loss of dopaminergic neurons mainly concentrated at substantia nigra. Within this context, this work presented here tries to provide a set of imaging features based on morphological characteristics extracted from I[Formula: see text]-Ioflupane SPECT scans to discern between HC and PD participants in a balanced set of [Formula: see text] scans from Parkinson’s Progression Markers Initiative (PPMI) database. These features, obtained from isosurfaces of each scan at different intensity levels, have been classified through the use of classical Machine Learning classifiers such as Support-Vector-Machines (SVM) or Naïve Bayesian and compared with the results obtained using a Multi-Layer Perceptron (MLP).

The proposed system, based on a Mann–Whitney–Wilcoxon U-Test for feature selection and the SVM approach, yielded a [Formula: see text] balanced accuracy when the performance was evaluated using a [Formula: see text]-fold cross-validation. This proves the reliability of these biomarkers, especially those related to sphericity, center of mass, number of vertices, 2D-projected perimeter or the 2D-projected eccentricity, among others, but including both internal and external isosurfaces.

Middle-Level Features for the Explanation of Classification Systems by Sparse Dictionary Methods

Machine learning (ML) systems are affected by a pervasive lack of transparency. The eXplainable Artificial Intelligence (XAI) research area addresses this problem and the related issue of explaining the behavior of ML systems in terms that are understandable to human beings. In many explanation of XAI approaches, the output of ML systems are explained in terms of low-level features of their inputs. However, these approaches leave a substantive explanatory burden with human users, insofar as the latter are required to map low-level properties into more salient and readily understandable parts of the input. To alleviate this cognitive burden, an alternative model-agnostic framework is proposed here. This framework is instantiated to address explanation problems in the context of ML image classification systems, without relying on pixel relevance maps and other low-level features of the input. More specifically, one obtains sets of middle-level properties of classification inputs that are perceptually salient by applying sparse dictionary learning techniques. These middle-level properties are used as building blocks for explanations of image classifications. The achieved explanations are parsimonious, for their reliance on a limited set of middle-level image properties. And they can be contrastive, because the set of middle-level image properties can be used to explain why the system advanced the proposed classification over other antagonist classifications. In view of its model-agnostic character, the proposed framework is adaptable to a variety of other ML systems and explanation problems.

Ensemble Deep Learning on Large, Mixed-Site fMRI Datasets in Autism and Other Tasks

Deep learning models for MRI classification face two recurring problems: they are typically limited by low sample size, and are abstracted by their own complexity (the “black box problem”). In this paper, we train a convolutional neural network (CNN) with the largest multi-source, functional MRI (fMRI) connectomic dataset ever compiled, consisting of 43,858 datapoints. We apply this model to a cross-sectional comparison of autism spectrum disorder (ASD) versus typically developing (TD) controls that has proved difficult to characterize with inferential statistics. To contextualize these findings, we additionally perform classifications of gender and task versus rest. Employing class-balancing to build a training set, we trained [Formula: see text] modified CNNs in an ensemble model to classify fMRI connectivity matrices with overall AUROCs of 0.6774, 0.7680, and 0.9222 for ASD versus TD, gender, and task versus rest, respectively. Additionally, we aim to address the black box problem in this context using two visualization methods. First, class activation maps show which functional connections of the brain our models focus on when performing classification. Second, by analyzing maximal activations of the hidden layers, we were also able to explore how the model organizes a large and mixed-center dataset, finding that it dedicates specific areas of its hidden layers to processing different covariates of data (depending on the independent variable analyzed), and other areas to mix data from different sources. Our study finds that deep learning models that distinguish ASD from TD controls focus broadly on temporal and cerebellar connections, with a particularly high focus on the right caudate nucleus and paracentral sulcus.

Epileptic Seizure Detection with EEG Textural Features and Imbalanced Classification Based on EasyEnsemble Learning

Imbalance data classification is a challenging task in automatic seizure detection from electroencephalogram (EEG) recordings when the durations of non-seizure periods are much longer than those of seizure activities. An imbalanced learning model is proposed in this paper to improve the identification of seizure events in long-term EEG signals. To better represent the underlying microstructure distributions of EEG signals while preserving the non-stationary nature, discrete wavelet transform (DWT) and uniform 1D-LBP feature extraction procedure are introduced. A learning framework is then designed by the ensemble of weakly trained support vector machines (SVMs). Under-sampling is employed to split the imbalanced seizure and non-seizure samples into multiple balanced subsets where each of them is utilized to train an individual SVM classifier. The weak SVMs are incorporated to build a strong classifier which emphasizes seizure samples and in the meantime analyzing the imbalanced class distribution of EEG data. Final seizure detection results are obtained in a multi-level decision fusion process by considering temporal and frequency factors. The model was validated over two long-term and one short-term public EEG databases. The model achieved a [Formula: see text]-mean of 97.14% with respect to epoch-level assessment, an event-level sensitivity of 96.67%, and a false detection rate of 0.86/h on the long-term intracranial database. An epoch-level [Formula: see text]-mean of 95.28% and event-level false detection rate of 0.81/h were yielded over the long-term scalp database. The comparisons with 14 published methods demonstrated the improved detection performance for imbalanced EEG signals and the generalizability of the proposed model.

Performing Multi-Target Regression via a Parameter Sharing-Based Deep Network

Multi-target regression (MTR) comprises the prediction of multiple continuous target variables from a common set of input variables. There are two major challenges when addressing the MTR problem: the exploration of the inter-target dependencies and the modeling of complex input–output relationships. This paper proposes a neural network model that is able to simultaneously address these two challenges in a flexible way. A deep architecture well suited for learning multiple continuous outputs is designed, providing some flexibility to model the inter-target relationships by sharing network parameters as well as the possibility to exploit target-specific patterns by learning a set of nonshared parameters for each target. The effectiveness of the proposal is analyzed through an extensive experimental study on 18 datasets, demonstrating the benefits of using a shared representation that exploits the commonalities between target variables. According to the experimental results, the proposed model is competitive with respect to the state-of-the-art in MTR.

Descriptor Selection Improvements for Quantitative Structure-Activity Relationships

Molecular descriptor selection is an essential procedure to improve a predictive quantitative structure–activity relationship (QSAR) model. However, within the QSAR model, there are a number of redundant, noisy and irrelevant descriptors. In this study, we propose a novel descriptor selection framework using self-paced learning (SPL) via sparse logistic regression (LR) with Logsum penalty (SPL-Logsum), which can simultaneously adaptively identify the simple and complex samples and avoid over-fitting. SPL is inspired by the learning process of humans or animals gradually learned from simple and complex samples to train models, and the Logsum penalized LR helps to select a small subset of significant molecular descriptors for improving the QSAR models. Experimental results on some simulations and three public QSAR datasets show that our proposed SPL-Logsum framework outperforms other existing sparse methods regarding the area under the curve, sensitivity, specificity, accuracy, and [Formula: see text]-values.

A Machine Learning Approach to Reveal the NeuroPhenotypes of Autisms

Although much research has been undertaken, the spatial patterns, developmental course, and sexual dimorphism of brain structure associated with autism remains enigmatic. One of the difficulties in investigating differences between the sexes in autism is the small sample sizes of available imaging datasets with mixed sex. Thus, the majority of the investigations have involved male samples, with females somewhat overlooked. This paper deploys machine learning on partial least squares feature extraction to reveal differences in regional brain structure between individuals with autism and typically developing participants. A four-class classification problem (sex and condition) is specified, with theoretical restrictions based on the evaluation of a novel upper bound in the resubstitution estimate. These conditions were imposed on the classifier complexity and feature space dimension to assure generalizable results from the training set to test samples. Accuracies above [Formula: see text] on gray and white matter tissues estimated from voxel-based morphometry (VBM) features are obtained in a sample of equal-sized high-functioning male and female adults with and without autism ([Formula: see text], [Formula: see text]/group). The proposed learning machine revealed how autism is modulated by biological sex using a low-dimensional feature space extracted from VBM. In addition, a spatial overlap analysis on reference maps partially corroborated predictions of the “extreme male brain” theory of autism, in sexual dimorphic areas.

A Machine Learning Approach to Characterize the Modulation of the Hippocampal Rhythms Via Optogenetic Stimulation of the Medial Septum

The medial septum (MS) is a potential target for modulating hippocampal activity. However, given the multiple cell types involved, the changes in hippocampal neural activity induced by MS stimulation have not yet been fully characterized. We combined MS optogenetic stimulation with local field potential (LFP) recordings from the hippocampus and leveraged machine learning techniques to explore how activating or inhibiting multiple MS neuronal subpopulations using different optical stimulation parameters affects hippocampal LFP biomarkers. First, of the seven different optogenetic viral vectors used for modulating different neuronal subpopulations, only two induced a substantial change in hippocampal LFP. Second, we found hippocampal low-gamma band to be most effectively modulated by the stimulation. Third, the hippocampal biomarkers were sensitive to the optogenetic virus type and the stimulation frequency, establishing those parameters as the critical ones for the regulation of hippocampal biomarker activity. Last, we built a Gaussian process regression model to describe the relationship between stimulation parameters and activity of the biomarker as well as to identify the optimal parameters for biomarker modulation. This new machine learning approach can further our understanding of the effects of neural stimulation and guide the selection of optimal parameters for neural control.

Latent Phase Detection of Hypoxic-Ischemic Spike Transients in the EEG of Preterm Fetal Sheep Using Reverse Biorthogonal Wavelets & Fuzzy Classifier

Hypoxic-ischemic (HI) studies in preterms lack reliable prognostic biomarkers for diagnostic tests of HI encephalopathy (HIE). Our group's observations from in utero fetal sheep models suggest that potential biomarkers of HIE in the form of developing HI micro-scale epileptiform transients emerge along suppressed EEG/ECoG background during a latent phase of 6-7h post-insult. However, having to observe for the whole of the latent phase disqualifies any chance of clinical intervention. A precise automatic identification of these transients can help for a well-timed diagnosis of the HIE and to stop the spread of the injury before it becomes irreversible. This paper reports fusion of Reverse-Biorthogonal Wavelets with Type-1 Fuzzy classifiers, for the accurate real-time automatic identification and quantification of high-frequency HI spike transients in the latent phase, tested over seven in utero preterm sheep. Considerable high performance of 99.78 ± 0.10% was obtained from the Rbio-Wavelet Type-1 Fuzzy classifier for automatic identification of HI spikes tested over 42h of high-resolution recordings (sampling-freq:1024Hz). Data from post-insult automatic time-localization of high-frequency HI spikes reveals a promising trend in the average rate of the HI spikes, even in the animals with shorter occlusion periods, which highlights considerable higher number of transients within the first 2h post-insult.

A Deep Learning Model for Automated Sleep Stages Classification Using PSG Signals

Sleep disorder is a symptom of many neurological diseases that may significantly affect the quality of daily life. Traditional methods are time-consuming and involve the manual scoring of polysomnogram (PSG) signals obtained in a laboratory environment. However, the automated monitoring of sleep stages can help detect neurological disorders accurately as well. In this study, a flexible deep learning model is proposed using raw PSG signals. A one-dimensional convolutional neural network (1D-CNN) is developed using electroencephalogram (EEG) and electrooculogram (EOG) signals for the classification of sleep stages. The performance of the system is evaluated using two public databases (sleep-edf and sleep-edfx). The developed model yielded the highest accuracies of 98.06%, 94.64%, 92.36%, 91.22%, and 91.00% for two to six sleep classes, respectively, using the sleep-edf database. Further, the proposed model obtained the highest accuracies of 97.62%, 94.34%, 92.33%, 90.98%, and 89.54%, respectively for the same two to six sleep classes using the sleep-edfx dataset. The developed deep learning model is ready for clinical usage, and can be tested with big PSG data.

Perceptual Generalization and Context in a Network Memory Inspired Long-Term Memory for Artificial Cognition

In the framework of open-ended learning cognitive architectures for robots, this paper deals with the design of a Long-Term Memory (LTM) structure that can accommodate the progressive acquisition of experience-based decision capabilities, or what different authors call "automation" of what is learnt, as a complementary system to more common prospective functions. The LTM proposed here provides for a relational storage of knowledge nuggets given the form of artificial neural networks (ANNs) that is representative of the contexts in which they are relevant in a configural associative structure. It also addresses the problem of continuous perceptual spaces and the task- and context-related generalization or categorization of perceptions in an autonomous manner within the embodied sensorimotor apparatus of the robot. These issues are analyzed and a solution is proposed through the introduction of two new types of knowledge nuggets: P-nodes representing perceptual classes and C-nodes representing contexts. The approach is studied and its performance evaluated through its implementation and application to a real robotic experiment.