Novel machine learning methods for ERP analysis: a validation from research on infants at risk for autism

Contributions of Artificial Intelligence to Analysis of Gut Microbiota in Autism Spectrum Disorder: A Systematic Review

BACKGROUND

Autism spectrum disorder (ASD) is a highly heterogeneous neurodevelopmental disorder whose etiology is not known today, but everything indicates that it is multifactorial. For example, genetic and epigenetic factors seem to be involved in the etiology of ASD. In recent years, there has been an increase in studies on the implications of gut microbiota (GM) on the behavior of children with ASD given that dysbiosis in GM may trigger the onset, development and progression of ASD through the microbiota-gut-brain axis. At the same time, significant progress has occurred in the development of artificial intelligence (AI).

METHODS

The aim of the present study was to perform a systematic review of articles using AI to analyze GM in individuals with ASD. In line with the PRISMA model, 12 articles using AI to analyze GM in ASD were selected.

RESULTS

Outcomes reveal that the majority of relevant studies on this topic have been conducted in China (33.3%) and Italy (25%), followed by the Netherlands (16.6%), Mexico (16.6%) and South Korea (8.3%).

CONCLUSIONS

The bacteria Bifidobacterium is the most relevant biomarker with regard to ASD. Although AI provides a very promising approach to data analysis, caution is needed to avoid the over-interpretation of preliminary findings. A first step must be taken to analyze GM in a representative general population and ASD samples in order to obtain a GM standard according to age, sex and country. Thus, more work is required to bridge the gap between AI in mental health research and clinical care in ASD.

Accurately predicting hit songs using neurophysiology and machine learning

Identifying hit songs is notoriously difficult. Traditionally, song elements have been measured from large databases to identify the lyrical aspects of hits. We took a different methodological approach, measuring neurophysiologic responses to a set of songs provided by a streaming music service that identified hits and flops. We compared several statistical approaches to examine the predictive accuracy of each technique. A linear statistical model using two neural measures identified hits with 69% accuracy. Then, we created a synthetic set data and applied ensemble machine learning to capture inherent non-linearities in neural data. This model classified hit songs with 97% accuracy. Applying machine learning to the neural response to 1st min of songs accurately classified hits 82% of the time showing that the brain rapidly identifies hit music. Our results demonstrate that applying machine learning to neural data can substantially increase classification accuracy for difficult to predict market outcomes.

Machine Learning and Prediction in Fetal, Infant, and Toddler Neuroimaging: A Review and Primer

Predictive models in neuroimaging are increasingly designed with the intent to improve risk stratification and support interventional efforts in psychiatry. Many of these models have been developed in samples of children school-aged or older. Nevertheless, despite growing evidence that altered brain maturation during the fetal, infant, and toddler (FIT) period modulates risk for poor mental health outcomes in childhood, these models are rarely implemented in FIT samples. Applications of predictive modeling in children of these ages provide an opportunity to develop powerful tools for improved characterization of the neural mechanisms underlying development. To facilitate the broader use of predictive models in FIT neuroimaging, we present a brief primer and systematic review on the methods used in current predictive modeling FIT studies. Reflecting on current practices in more than 100 studies conducted over the past decade, we provide an overview of topics, modalities, and methods commonly used in the field and under-researched areas. We then outline ethical and future considerations for neuroimaging researchers interested in predicting health outcomes in early life, including researchers who may be relatively new to either advanced machine learning methods or using FIT data. Altogether, the last decade of FIT research in machine learning has provided a foundation for accelerating the prediction of early-life trajectories across the full spectrum of illness and health.

Identifying Visual Attention Features Accurately Discerning Between Autism and Typically Developing: a Deep Learning Framework

Atypical visual attention is a hallmark of autism spectrum disorder (ASD). Identifying the attention features accurately discerning between people with ASD and typically developing (TD) at the individual level remains a challenge. In this study, we developed a new systematic framework combining high accuracy deep learning classification, deep learning segmentation, image ablation and a direct measurement of classification ability to identify the discriminative features for autism identification. Our two-stream model achieved the state-of-the-art performance with a classification accuracy of 0.95. Using this framework, two new categories of features, Food & drink and Outdoor-objects, were identified as discriminative attention features, in addition to the previously reported features including Center-object and Human-faces, etc. Altered attention to the new categories helps to understand related atypical behaviors in ASD. Importantly, the area under curve (AUC) based on the combined top-9 features identified in this study was 0.92, allowing an accurate classification at the individual level. We also obtained a small but informative dataset of 12 images with an AUC of 0.86, suggesting a potentially efficient approach for the clinical diagnosis of ASD. Together, our deep learning framework based on VGG-16 provides a novel and powerful tool to recognize and understand abnormal visual attention in ASD, which will, in turn, facilitate the identification of biomarkers for ASD.

A practical guide to applying machine learning to infant EEG data

Electroencephalography (EEG) has been widely adopted by the developmental cognitive neuroscience community, but the application of machine learning (ML) in this domain lags behind adult EEG studies. Applying ML to infant data is particularly challenging due to the low number of trials, low signal-to-noise ratio, high inter-subject variability, and high inter-trial variability. Here, we provide a step-by-step tutorial on how to apply ML to classify cognitive states in infants. We describe the type of brain attributes that are widely used for EEG classification and also introduce a Riemannian geometry based approach for deriving connectivity estimates that account for inter-trial and inter-subject variability. We present pipelines for learning classifiers using trials from a single infant and from multiple infants, and demonstrate the application of these pipelines on a standard infant EEG dataset of forty 12-month-old infants collected under an auditory oddball paradigm. While we classify perceptual states induced by frequent versus rare stimuli, the presented pipelines can be easily adapted for other experimental designs and stimuli using the associated code that we have made publicly available.

The Value of Brain Imaging and Electrophysiological Testing for Early Screening of Autism Spectrum Disorder: A Systematic Review

Given the significance of validating reliable tests for the early detection of autism spectrum disorder (ASD), this systematic review aims to summarize available evidence of neuroimaging and neurophysiological changes in high-risk infants to improve ASD early diagnosis. We included peer-reviewed, primary research in English published before May 21, 2021, involving the use of magnetic resonance imaging (MRI), electroencephalogram (EEG), or functional near-infrared spectroscopy (fNIRS) in children with high risk for ASD under 24 months of age. The main exclusion criteria includes diagnosis of a genetic disorder and gestation age of less the 36 weeks. Online research was performed on PubMed, Web of Science, PsycINFO, and CINAHL. Article selection was conducted by two reviewers to minimize bias. This research was funded by Massachusetts General Hospital Sundry funding. IRB approval was not submitted as it was deemed unnecessary. We included 75 primary research articles. Studies showed that high-risk infants had divergent developmental trajectories for fractional anisotropy and regional brain volumes, increased CSF volume, and global connectivity abnormalities on MRI, decreased sensitivity for familiar faces, atypical lateralization during facial and auditory processing, and different spectral powers across multiple band frequencies on EEG, and distinct developmental trajectories in functional connectivity and regional oxyhemoglobin concentrations in fNIRS. These findings in infants were found to be correlated with the core ASD symptoms and diagnosis at toddler age. Despite the lack of quantitative analysis of the research database, neuroimaging and electrophysiological biomarkers have promising value for the screening of ASD as early as infancy with high accuracy, which warrants further investigation.

Internal consistency reliability of the P300 to novelty in infants: The influence of trial number and data loss due to artifacts

The P300 is an event-related potential component that reflects attention to motivationally salient stimuli and may be a promising tool to examine individual differences in cognitive-affective processing very early in development. However, the psychometric properties of the P300 in infancy are unknown, a fact that limits the component's utility as an individual difference measure in developmental research. To address this gap, 38 infants completed an auditory three-stimulus oddball task that included frequent standard, infrequent deviant, and novel stimuli. We quantified the P300 at a single electrode site and at region of interest (ROI) and examined the internal consistency reliability of the component, both via split-half reliability and as a function of trial number. Results indicated that the P300 to standard, deviant, and novel stimuli fell within moderate to high internal consistency reliability thresholds, and that scoring the component at an ROI led to slightly higher estimates of reliability. However, the percentage of data loss due to artifacts increased across the course of the task, suggesting that including more trials will not necessarily improve the reliability of the P300. Together, these results suggest that robust and reliable measurement of the P300 will require designing tasks that minimize trial number and maximize infant tolerability.

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.

Discriminative Dictionary Learning for Autism Spectrum Disorder Identification

Autism Spectrum Disorder (ASD) is a group of lifelong neurodevelopmental disorders with complicated causes. A key symptom of ASD patients is their impaired interpersonal communication ability. Recent study shows that face scanning patterns of individuals with ASD are often different from those of typical developing (TD) ones. Such abnormality motivates us to study the feasibility of identifying ASD children based on their face scanning patterns with machine learning methods. In this paper, we consider using the bag-of-words (BoW) model to encode the face scanning patterns, and propose a novel dictionary learning method based on dual mode seeking for better BoW representation. Unlike k-means which is broadly used in conventional BoW models to learn dictionaries, the proposed method captures discriminative information by finding atoms which maximizes both the purity and coverage of belonging samples within one class. Compared to the rich literature of ASD studies from psychology and neural science, our work marks one of the relatively few attempts to directly identify high-functioning ASD children with machine learning methods. Experiments demonstrate the superior performance of our method with considerable gain over several baselines. Although the proposed work is yet too preliminary to directly replace existing autism diagnostic observation schedules in the clinical practice, it shed light on future applications of machine learning methods in early screening of ASD.

A single-session behavioral protocol for successful event-related potential recording in children with neurodevelopmental disorders

Event-related potentials (ERPs) are an ideal tool for measuring neural responses in a wide range of participants, including children diagnosed with neurodevelopmental disorders (NDDs). However, due to perceived barriers regarding participant compliance, much of this work has excluded children with low IQ and/or reduced adaptive functioning, significant anxiety symptoms, and/or sensory processing difficulties, including heterogeneous samples of children with autism spectrum disorder (ASD) and children with fragile X syndrome (FXS). We have developed a behavioral support protocol designed to obtain high-quality ERP data from children in a single session. Using this approach, ERP data were successfully collected from participants with ASD, FXS, and typical development (TD). Higher success rates were observed for children with ASD and TD than children with FXS. Unique clinical-behavioral characteristics were associated with successful data collection across these groups. Higher chronological age, nonverbal mental age, and receptive language skills were associated with a greater number of valid trials completed in children with ASD. In contrast, higher language ability, lower autism severity, increased anxiety, and increased sensory hyperresponsivity were associated with a greater number of valid trials completed in children with FXS. This work indicates that a "one-size-fits-all" approach cannot be taken to ERP research on children with NDDs, but that a single-session paradigm is feasible and is intended to promote increased representation of children with NDDs in neuroscience research through development of ERP methods that support inclusion of diverse and representative samples.

A neural network predicting the amplitude of the N2pc in individual EEG datasets

Objective.The N2pc is a small amplitude transient interhemispheric voltage asymmetry used in cognitive neuroscience to investigate subject's allocation of selective visuo-spatial attention. N2pc is typically estimated by averaging the sweeps of the electroencephalographic (EEG) signal but, in absence of explicit normative indications, the number of sweeps is often based on arbitrariness or personal experience. With the final aim of reducing duration and cost of experimental protocols, here we developed a new approach to reliably predict N2pc amplitude from a minimal EEG dataset.Approach.First, features predictive of N2pc amplitude were identified in the time-frequency domain. Then, an artificial neural network (NN) was trained to predict N2pc mean amplitude at the individual level. By resorting to simulated data, accuracy of the NN was assessed by computing the mean squared error (MSE) and the amplitude discretization error (ADE) and compared to the standard time averaging (TA) technique. The NN was then tested against two real datasets consisting of 14 and 12 subjects, respectively.Main result.In simulated scenarios entailing different number of sweeps (between 10 and 100), the MSE obtained with the proposed method resulted, on average, 1/5 of that obtained with the TA technique. Implementation on real EEG datasets showed that N2pc amplitude could be reliably predicted with as few as 40 EEG sweeps per cell of the experimental design.Significance.The developed approach allows to reduce duration and cost of experiments involving the N2pc, for instance in studies investigating attention deficits in pathological subjects.

Automatic classification of children with autism spectrum disorder by using a computerized visual-orienting task

Early screening and diagnosis of autism spectrum disorder (ASD) primarily rely on behavioral observations by qualified clinicians whose decision process can benefit from the combination of machine learning algorithms and sensor data. We designed a computerized visual-orienting task with gaze-related or non-gaze-related directional cues, which triggered participants' gaze-following behavior. Based on their eye-movement data registered by an eye tracker, we applied the machine learning algorithms to classify high-functioning children with ASD (HFA), low-functioning children with ASD (LFA), and typically developing children (TD). We found that TD children had higher success rates in obtaining rewards than HFA children, and HFA children had higher rates than LFA children. Based on raw eye-tracking data, our machine learning algorithm could classify the three groups with an accuracy of 81.1% and relatively high sensitivity and specificity. Classification became worse if only data from the gaze or nongaze conditions were used, suggesting that "less-social" directional cues also carry useful information for distinguishing these groups. Our findings not only provide insights about visual-orienting deficits among children with ASD but also demonstrate the promise of combining classical behavioral paradigms with machine learning algorithms for aiding the screening for individuals with ASD.

Use of Empirical Mode Decomposition in ERP Analysis to Classify Familial Risk and Diagnostic Outcomes for Autism Spectrum Disorder

Event-related potentials (ERPs) activated by faces and gaze processing are found in individuals with autism spectrum disorder (ASD) in the early stages of their development and may serve as a putative biomarker to supplement behavioral diagnosis. We present a novel approach to the classification of visual ERPs collected from 6-month-old infants using intrinsic mode functions (IMFs) derived from empirical mode decomposition (EMD). Selected features were used as inputs to two machine learning methods (support vector machines and k-nearest neighbors (k-NN)) using nested cross validation. Different runs were executed for the modelling and classification of the participants in the control and high-risk (HR) groups and the classification of diagnosis outcome within the high-risk group: HR-ASD and HR-noASD. The highest accuracy in the classification of familial risk was 88.44%, achieved using a support vector machine (SVM). A maximum accuracy of 74.00% for classifying infants at risk who go on to develop ASD vs. those who do not was achieved through k-NN. IMF-based extracted features were highly effective in classifying infants by risk status, but less effective by diagnostic outcome. Advanced signal analysis of ERPs integrated with machine learning may be considered a first step toward the development of an early biomarker for ASD.

Virulence factor-related gut microbiota genes and immunoglobulin A levels as novel markers for machine learning-based classification of autism spectrum disorder

Autism spectrum disorder (ASD) is a neurodevelopmental condition for which early identification and intervention is crucial for optimum prognosis. Our previous work showed gut Immunoglobulin A (IgA) to be significantly elevated in the gut lumen of children with ASD compared to typically developing (TD) children. Gut microbiota variations have been reported in ASD, yet not much is known about virulence factor-related gut microbiota (VFGM) genes. Upon determining the VFGM genes distinguishing ASD from TD, this study is the first to utilize VFGM genes and IgA levels for a machine learning-based classification of ASD. Sequence comparisons were performed of metagenome datasets from children with ASD (n = 43) and TD children (n = 31) against genes in the virulence factor database. VFGM gene composition was associated with ASD phenotype. VFGM gene diversity was higher in children with ASD and positively correlated with IgA content. As Group B streptococcus (GBS) genes account for the highest proportion of 24 different VFGMs between ASD and TD and positively correlate with gut IgA, GBS genes were used in combination with IgA and VFGMs diversity to distinguish ASD from TD. Given that VFGM diversity, increases in IgA, and ASD-enriched VFGM genes were independent of sex and gastrointestinal symptoms, a classification method utilizing them will not pertain only to a specific subgroup of ASD. By introducing the classification value of VFGM genes and considering that VFs can be isolated in pregnant women and newborns, these findings provide a novel machine learning-based early risk identification method for ASD.

Automated identification of postural control for children with autism spectrum disorder using a machine learning approach

It is unclear whether postural sway characteristics could be used as diagnostic biomarkers for autism spectrum disorder (ASD). The purpose of this study was to develop and validate an automated identification of postural control patterns in children with ASD using a machine learning approach. 50 children aged 5-12 years old were recruited and assigned into two groups: ASD (n = 25) and typically developing groups (n = 25). Participants were instructed to stand barefoot on two feet and maintain a stationary stance for 20 s during two conditions: (1) eyes open and (2) eyes closed. The center of pressure (COP) data were collected using a force plate. COP variables were computed, including linear displacement, total distance, sway area, and complexity. Six supervised machine learning classifiers were trained to classify the ASD postural control based on these COP variables. All machine learning classifiers successfully identified ASD postural control patterns based on the COP features with high accuracy rates (>0.800). The naïve Bayes method was the optimal means to identify ASD postural control with the highest accuracy rate (0.900), specificity (1.000), precision (1.000), F1 score (0.898) and satisfactory sensitivity (0.826). By increasing the sample size and analyzing more data/features of postural control, a better classification performance would be expected. The use of computer-aided machine learning to assess COP data is efficient, accurate, with minimum human intervention and thus, could benefit the diagnosis of ASD.

An electrophysiological biomarker for the classification of cataract-reversal patients: A case-control study

BACKGROUND

Untreated congenital blindness through cataracts leads to lasting visual brain system changes, including substantial alterations of extrastriate visual areas. Consequently, late-treated individuals (> 5 months of age) with dense congenital bilateral cataracts (CC) exhibit poorer visual function recovery compared to individuals with bilateral developmental cataracts (DC). Reliable methods to differentiate between patients with congenital and developmental cataracts are often lacking, impeding efficient rehabilitation management and introducing confounds in clinical and basic research on recovery prognosis and optimal timing of surgery. A persistent reduction of the P1 wave of visual event-related potentials (VERPs), associated with extrastriate visual cortical activity, has been reported in CC but not in DC individuals. Using two experiments, this study developed and validated P1-based biomarkers for diagnosing a history of congenital blindness in cataract-reversal individuals.

METHODS

Congenital and developmental cataract-reversal individuals as well as typically sighted matched controls took part in a first experiment used for exploring an electrophysiological biomarker (N _CC = 13, N _DC = 13, N _Control = 26). Circular stimuli containing gratings were presented in one of the visual field quadrants while visual event-related potentials (VERPs) were recorded. Two biomarkers were derived from the P1 wave of the VERP: (1) The mean of the normalized P1 amplitude at posterior electrodes, and (2) a classifier obtained from a linear support vector machine (SVM). A second experiment with partially new CC/DC individuals and their matched controls (N _CC = 14, N _DC = 15, N _Control = 29) was consecutively used to validate the classification based on both biomarkers. Performance of the classifiers were evaluated using receiver operating characteristic (ROC) curve analyses. All cataract-reversal individuals were tested after at least one year of vision recovery.

FINDINGS

The normalized P1 amplitude over posterior electrodes allowed a successful classification of the CC from the DC individuals and typically sighted controls (area under ROC curve, AUC = 0.803 and 0.929 for the normalized P1 amplitude and the SVM-based biomarker, respectively). The validation for both biomarkers in experiment 2 again resulted in a high classification success (AUC = 0.800 and 0.883, respectively for the normalized P1 amplitude and the SVM-based biomarker). In the most conservative scenario involving classification of CC from DC individuals in a group of only cataract-reversal individuals, excluding typically sighted controls, the SVM-based biomarker was found to be superior to the mean P1 amplitude based biomarker (AUC = 0.852 compared to 0.757 for the mean P1 based biomarker in validation). Minimum specificity obtained was 80% across all biomarkers.

INTERPRETATION

A persistent reduction of the P1 wave provides a highly specific method for classifying cataract patients post-surgically as having suffered from bilateral congenital vs. bilateral developmental cataracts. We suggest that using the P1 based non-invasive electrophysiological biomarker will augment existing clinical classification criteria for individuals with a history of bilateral congenital cataracts, aiding clinical and basic research, recovery prognosis, and rehabilitation efforts.

FUNDING

German Research Foundation (DFG) and the European Research Council (ERC).

Towards a Multivariate Biomarker-Based Diagnosis of Autism Spectrum Disorder: Review and Discussion of Recent Advancements

An ever-evolving understanding of autism spectrum disorder (ASD) pathophysiology necessitates that diagnostic standards also evolve from being observation-based to include quantifiable clinical measurements. The multisystem nature of ASD motivates the use of multivariate methods of statistical analysis over common univariate approaches for discovering clinical biomarkers relevant to this goal. In addition to characterization of important behavioral patterns for improving current diagnostic instruments, multivariate analyses to date have allowed for thorough investigation of neuroimaging-based, genetic, and metabolic abnormalities in individuals with ASD. This review highlights current research using multivariate statistical analyses to quantify the value of these behavioral and physiological markers for ASD diagnosis. A detailed discussion of a blood-based diagnostic test for ASD using specific metabolite concentrations is also provided. The advancement of ASD biomarker research promises to provide earlier and more accurate diagnoses of the disorder.

Autism spectrum disorder risk prediction: A systematic review of behavioral and neural investigations

A reliable diagnosis of autism spectrum disorder (ASD) is difficult to make until after toddlerhood. Detection in an earlier age enables early intervention, which is typically more effective. Recent studies of the development of brain and behavior in infants and toddlers have provided important insights in the diagnosis of autism. This extensive review focuses on published studies of predicting the diagnosis of autism during infancy and toddlerhood younger than 3 years using behavioral and neuroimaging approaches. After screening a total of 782 papers, 17 neuroimaging and 43 behavioral studies were reviewed. The features for prediction consist of behavioral measures using screening tools, observational and experimental methods, brain volumetric measures, and neural functional activation and connectivity patterns. The classification approaches include logistic regression, linear discriminant function, decision trees, support vector machine, and deep learning based methods. Prediction performance has large variance across different studies. For behavioral studies, the sensitivity varies from 20% to 100%, and specificity ranges from 48% to 100%. The accuracy rates range from 61% to 94% in neuroimaging studies. Possible factors contributing to this inconsistency may be partially due to the heterogeneity of ASD, different targeted populations (i.e., high-risk group for ASD and general population), age when the features were collected, and validation procedures. The translation to clinical practice requires extensive further research including external validation with large sample size and optimized feature selection. The use of multi-modal features, e.g., combination of neuroimaging and behavior, is worth further investigation to improve the prediction accuracy.

Functional Connectivities Are More Informative Than Anatomical Variables in Diagnostic Classification of Autism

Machine learning techniques have been implemented to reveal brain features that distinguish people with autism spectrum disorders (ASDs) from typically developing (TD) peers. However, it remains unknown whether different neuroimaging modalities are equally informative for diagnostic classification. We combined anatomical magnetic resonance imaging (aMRI), diffusion weighted imaging (DWI), and functional connectivity MRI (fcMRI) using conditional random forest (CRF) for supervised learning to compare how informative each modality was in diagnostic classification. In-house data (N = 93) included 47 TD and 46 ASD participants, matched on age, motion, and nonverbal IQ. Four main analyses consistently indicated that fcMRI variables were significantly more informative than anatomical variables from aMRI and DWI. This was found (1) when the top 100 variables from CRF (run separately in each modality) were combined for multimodal CRF; (2) when only 19 top variables reaching >67% accuracy in each modality were combined in multimodal CRF; and (3) when the large number of initial variables (before dimension reduction) potentially biasing comparisons in favor of fcMRI was reduced using a less granular region of interest scheme. Consistent superiority of fcMRI was even found (4) when 100 variables per modality were randomly selected, removing any such potential bias. Greater informative value of functional than anatomical modalities may relate to the nature of fcMRI data, reflecting more closely behavioral condition, which is also the basis of diagnosis, whereas brain anatomy may be more reflective of neurodevelopmental history.

Event related potential analysis techniques for autism spectrum disorders: A review

Autism Spectrum Disorders (ASD) comprise all pervasive neurodevelopmental diseases marked by deficits in social and communication skills, delayed cognitive development, restricted and repetitive behaviors. The core symptoms begin in early childhood, may continue life-long resulting in poor performance in adult stage. Event-related potential (ERP) is basically a time-locked electroencephalogram signal elicited by various stimuli, related to sensory and cognitive processes. The various ERP based techniques used for the study of ASD are considered in this review. ERP based study offers the advantage of being a non-invasive technique to measure the brain activity precisely. The techniques are categorized into three based on the processing domain: time, frequency and time-frequency. Power spectral density, coherence, phase synchrony, multiscale entropy, modified multiscale entropy, sum of signed differences, synchrostates and variance are some of the measures that have been widely used to study the abnormalities in frequency bands and brain connectivity. Various signal processing techniques such as Fast Fourier Transform, Discrete Fourier Transform, Short-Time Fourier Transform, Principal Component Analysis, Wavelet Transform, Directed Transfer Function etc. have been used to analyze the recorded signals so as to unravel the distinctive event-related potential patterns in individuals with ASD. The review concludes that ERP proves to be an efficient tool in detecting the brain abnormalities and connectivity issues, indicating the heterogeneity of ASD. Many advanced techniques are utilized to decipher the underlying neural circuitry so as to aid in therapeutic interventions for improving the core areas of deficits.

Fearful but not happy expressions boost face detection in human infants

Human adults show an attentional bias towards fearful faces, an adaptive behaviour that relies on amygdala function. This attentional bias emerges in infancy between 5 and 7 months, but the underlying developmental mechanism is unknown. To examine possible precursors, we investigated whether 3.5-, 6- and 12-month-old infants show facilitated detection of fearful faces in noise, compared to happy faces. Happy or fearful faces, mixed with noise, were presented to infants (N = 192), paired with pure noise. We applied multivariate pattern analyses to several measures of infant looking behaviour to derive a criterion-free, continuous measure of face detection evidence in each trial. Analyses of the resulting psychometric curves supported the hypothesis of a detection advantage for fearful faces compared to happy faces, from 3.5 months of age and across all age groups. Overall, our data show a readiness to detect fearful faces (compared to happy faces) in younger infants that developmentally precedes the previously documented attentional bias to fearful faces in older infants and adults.

Applying machine learning to identify autistic adults using imitation: An exploratory study

Autism spectrum condition (ASC) is primarily diagnosed by behavioural symptoms including social, sensory and motor aspects. Although stereotyped, repetitive motor movements are considered during diagnosis, quantitative measures that identify kinematic characteristics in the movement patterns of autistic individuals are poorly studied, preventing advances in understanding the aetiology of motor impairment, or whether a wider range of motor characteristics could be used for diagnosis. The aim of this study was to investigate whether data-driven machine learning based methods could be used to address some fundamental problems with regard to identifying discriminative test conditions and kinematic parameters to classify between ASC and neurotypical controls. Data was based on a previous task where 16 ASC participants and 14 age, IQ matched controls observed then imitated a series of hand movements. 40 kinematic parameters extracted from eight imitation conditions were analysed using machine learning based methods. Two optimal imitation conditions and nine most significant kinematic parameters were identified and compared with some standard attribute evaluators. To our knowledge, this is the first attempt to apply machine learning to kinematic movement parameters measured during imitation of hand movements to investigate the identification of ASC. Although based on a small sample, the work demonstrates the feasibility of applying machine learning methods to analyse high-dimensional data and suggest the potential of machine learning for identifying kinematic biomarkers that could contribute to the diagnostic classification of autism.

A machine learning approach to identify functional biomarkers in human prefrontal cortex for individuals with traumatic brain injury using functional near-infrared spectroscopy

BACKGROUND

We have explored the potential prefrontal hemodynamic biomarkers to characterize subjects with Traumatic Brain Injury (TBI) by employing the multivariate machine learning approach and introducing a novel task-related hemodynamic response detection followed by a heuristic search for optimum set of hemodynamic features. To achieve this goal, the hemodynamic response from a group of 31 healthy controls and 30 chronic TBI subjects were recorded as they performed a complexity task.

METHODS

To determine the optimum hemodynamic features, we considered 11 features and their combinations in characterizing TBI subjects. We investigated the significance of the features by utilizing a machine learning classification algorithm to score all the possible combinations of features according to their predictive power.

RESULTS AND CONCLUSIONS

The identified optimum feature elements resulted in classification accuracy, sensitivity, and specificity of 85%, 85%, and 84%, respectively. Classification improvement was achieved for TBI subject classification through feature combination. It signified the major advantage of the multivariate analysis over the commonly used univariate analysis suggesting that the features that are individually irrelevant in characterizing the data may become relevant when used in combination. We also conducted a spatio-temporal classification to identify regions within the prefrontal cortex (PFC) that contribute in distinguishing between TBI and healthy subjects. As expected, Brodmann areas (BA) 10 within the PFC were isolated as the region that healthy subjects (unlike subjects with TBI), showed major hemodynamic activity in response to the High Complexity task. Overall, our results indicate that identified temporal and spatio-temporal features from PFC's hemodynamic activity are promising biomarkers in classifying subjects with TBI.

Autism research dynamic through ontology-based text mining

Purpose

– The increase of prevalence of autism spectrum disorders (ASD) has been accompanied by much new research. The amount and the speed of growth of scientific information available online have strongly influenced the way of work in the research community which calls for new methods and tools to support it. The purpose of this paper is to present ontology-based text mining in the field of autism trend analysis that may help to understand the broader picture of the disorder since its discovery.

Design/methodology/approach

– The data sets consisted of abstracts of more than 18,000 articles on ASD published from 1943 to the end of 2012 found in MEDLINE and of the documents’ titles for all those articles where the abstracts were not available.

Findings

– In this way, the authors demonstrated a steeper exponential curve of ASD publications compared with all publications in MEDLINE. In addition, the main research topics over time were identified using the “open discovery” approach. Finally, the relationship between a priori setting up research topics including communication, genetics, environmental risk factors, vaccination and adulthood were revealed.

Originality/value

– Using ontology-based text mining the authors were able to identify the main research topics in the field of autism during the time, as well as to show the dynamics of some research topics as a priori setting up. The computerised methodology that was used allowed the authors to analyse a much larger quantity of information, saving time and manual work.

Identifying children with autism spectrum disorder based on their face processing abnormality: A machine learning framework

The atypical face scanning patterns in individuals with Autism Spectrum Disorder (ASD) has been repeatedly discovered by previous research. The present study examined whether their face scanning patterns could be potentially useful to identify children with ASD by adopting the machine learning algorithm for the classification purpose. Particularly, we applied the machine learning method to analyze an eye movement dataset from a face recognition task [Yi et al., 2016], to classify children with and without ASD. We evaluated the performance of our model in terms of its accuracy, sensitivity, and specificity of classifying ASD. Results indicated promising evidence for applying the machine learning algorithm based on the face scanning patterns to identify children with ASD, with a maximum classification accuracy of 88.51%. Nevertheless, our study is still preliminary with some constraints that may apply in the clinical practice. Future research should shed light on further valuation of our method and contribute to the development of a multitask and multimodel approach to aid the process of early detection and diagnosis of ASD. Autism Res 2016, 9: 888-898. © 2016 International Society for Autism Research, Wiley Periodicals, Inc.

Knowledge-guided mutation in classification rules for autism treatment efficacy

Data mining methods in biomedical research might benefit by combining genetic algorithms with domain-specific knowledge. The objective of this research is to show how the evolution of treatment rules for autism might be guided. The semantic distance between two concepts in the taxonomy is measured by the number of relationships separating the concepts in the taxonomy. The hypothesis is that replacing a concept in a treatment rule will change the accuracy of the rule in direct proportion to the semantic distance between the concepts. The method uses a patient database and autism taxonomies. Treatment rules are developed with an algorithm that exploits the taxonomies. The results support the hypothesis. This research should both advance the understanding of autism data mining in particular and of knowledge-guided evolutionary search in biomedicine in general.

Detection of event-related potentials in individual subjects using support vector machines

Event-related potentials (ERPs) are tiny electrical brain responses in the human electroencephalogram that are typically not detectable until they are isolated by a process of signal averaging. Owing to the extremely smallsize of ERP components (ranging from less than 1 μV to tens of μV), compared to background brain rhythms, statistical analyses of ERPs are predominantly carried out in groups of subjects. This limitation is a barrier to the translation of ERP-based neuroscience to applications such as medical diagnostics. We show here that support vector machines (SVMs) are a useful method to detect ERP components in individual subjects with a small set of electrodes and a small number of trials for a mismatch negativity (MMN) ERP component. Such a reduced experiment setup is important for clinical applications. One hundred healthy individuals were presented with an auditory pattern containing pattern-violating deviants to evoke the MMN. Two-class SVMs were then trained to classify averaged ERP waveforms in response to the standard tone (tones that match the pattern) and deviant tone stimuli (tones that violate the pattern). The influence of kernel type, number of epochs, electrode selection, and temporal window size in the averaged waveform were explored. When using all electrodes, averages of all available epochs, and a temporal window from 0 to 900-ms post-stimulus, a linear SVM achieved 94.5 % accuracy. Further analyses using SVMs trained with narrower, sliding temporal windows confirmed the sensitivity of the SVM to data in the latency range associated with the MMN.

A rapid event-related potential (ERP) method for point-of-care evaluation of brain function: development of the Halifax Consciousness Scanner

BACKGROUND

Event-related potentials (ERPs) may provide a non-invasive index of brain function for a range of clinical applications. However, as a lab-based technique, ERPs are limited by technical challenges that prevent full integration into clinical settings.

NEW METHOD

To translate ERP capabilities from the lab to clinical applications, we have developed methods like the Halifax Consciousness Scanner (HCS). HCS is essentially a rapid, automated ERP evaluation of brain functional status. The present study describes the ERP components evoked from auditory tones and speech stimuli. ERP results were obtained using a 5-min test in 100 healthy individuals. The HCS sequence was designed to evoke the N100, the mismatch negativity (MMN), P300, the early negative enhancement (ENE), and the N400. These components reflected sensation, perception, attention, memory, and language perception, respectively. Component detection was examined at group and individual levels, and evaluated across both statistical and classification approaches.

RESULTS

All ERP components were robustly detected at the group level. At the individual level, nonparametric statistical analyses showed reduced accuracy relative to support vector (SVM) machine classification, particularly for speech-based ERPs. Optimized SVM results were MMN: 95.6%; P300: 99.0%; ENE: 91.8%; and N400: 92.3%.

CONCLUSIONS

A spectrum of individual-level ERPs can be obtained in a very short time. Machine learning classification improved detection accuracy across a large healthy control sample. Translating ERPs into clinical applications is increasingly possible at the individual level.

Guidelines and best practices for electrophysiological data collection, analysis and reporting in autism

The EEG reflects the activation of large populations of neurons that act in synchrony and propagate to the scalp surface. This activity reflects both the brain's background electrical activity and when the brain is being challenged by a task. Despite strong theoretical and methodological arguments for the use of EEG in understanding the neural correlates of autism, the practice of collecting, processing and evaluating EEG data is complex. Scientists should take into consideration both the nature of development in autism given the life-long, pervasive course of the disorder and the disability of altered or atypical social, communicative, and motor behaviors, all of which require accommodations to traditional EEG environments and paradigms. This paper presents guidelines for the recording, analyzing, and interpreting of EEG data with participants with autism. The goal is to articulate a set of scientific standards as well as methodological considerations that will increase the general field's understanding of EEG methods, provide support for collaborative projects, and contribute to the evaluation of results and conclusions.

Classification of autism spectrum disorder using supervised learning of brain connectivity measures extracted from synchrostates

OBJECTIVE

The paper investigates the presence of autism using the functional brain connectivity measures derived from electro-encephalogram (EEG) of children during face perception tasks.

APPROACH

Phase synchronized patterns from 128-channel EEG signals are obtained for typical children and children with autism spectrum disorder (ASD). The phase synchronized states or synchrostates temporally switch amongst themselves as an underlying process for the completion of a particular cognitive task. We used 12 subjects in each group (ASD and typical) for analyzing their EEG while processing fearful, happy and neutral faces. The minimal and maximally occurring synchrostates for each subject are chosen for extraction of brain connectivity features, which are used for classification between these two groups of subjects. Among different supervised learning techniques, we here explored the discriminant analysis and support vector machine both with polynomial kernels for the classification task.

MAIN RESULTS

The leave one out cross-validation of the classification algorithm gives 94.7% accuracy as the best performance with corresponding sensitivity and specificity values as 85.7% and 100% respectively.

SIGNIFICANCE

The proposed method gives high classification accuracies and outperforms other contemporary research results. The effectiveness of the proposed method for classification of autistic and typical children suggests the possibility of using it on a larger population to validate it for clinical practice.

In search of biomarkers in psychiatry: EEG-based measures of brain function

Current clinical parameters used for diagnosis and phenotypic definitions of psychopathology are both highly variable and subjective. Intensive research efforts for specific and sensitive biological markers, or biomarkers, for psychopathology as objective alternatives to the current paradigm are ongoing. While biomarker research in psychiatry has focused largely on functional neuroimaging methods for identifying the neural functions that associate with psychopathology, scalp electroencephalography (EEG) has been viewed, historically, as offering little specific brain source information, as scalp appearance is only loosely correlated to its brain source dynamics. However, ongoing advances in signal processing of EEG data can now deliver functional EEG brain-imaging with distinctly improved spatial, as well as fine temporal, resolution. One computational approach proving particularly useful for EEG cortical brain imaging is independent component analysis (ICA). ICA decomposition can be used to identify distinct cortical source activities that are sensitive and specific to the pathophysiology of psychiatric disorders. Given its practical research advantages, relatively low cost, and ease of use, EEG-imaging is now both feasible and attractive, in particular for studies involving the large samples required by genetically informative designs to characterize causal pathways to psychopathology. The completely non-invasive nature of EEG data acquisition, coupled with ongoing advances in dry, wireless, and wearable EEG technology, makes EEG-imaging increasingly attractive and appropriate for psychiatric research, including the study of developmentally young samples. Applied to large genetically and developmentally informative samples, EEG imaging can advance the search for robust diagnostic biomarkers and phenotypes in psychiatry.

Neurobiological abnormalities in the first few years of life in individuals later diagnosed with autism spectrum disorder: a review of recent data

Background. Despite the widely-held understanding that the biological changes that lead to autism usually occur during prenatal life, there has been relatively little research into the functional development of the brain during early infancy in individuals later diagnosed with autism spectrum disorder (ASD). Objective. This review explores the studies over the last three years which have investigated differences in various brain regions in individuals with ASD or who later go on to receive a diagnosis of ASD. Methods. We used PRISMA guidelines and selected published articles reporting any neurological abnormalities in very early childhood in individuals with or later diagnosed with ASD. Results. Various brain regions are discussed including the amygdala, cerebellum, frontal cortex, and lateralised abnormalities of the temporal cortex during language processing. This review discusses studies investigating head circumference, electrophysiological markers, and interhemispheric synchronisation. All of the recent findings from the beginning of 2009 across these different aspects of defining neurological abnormalities are discussed in light of earlier findings. Conclusions. The studies across these different areas reveal the existence of atypicalities in the first year of life, well before ASD is reliably diagnosed. Cross-disciplinary approaches are essential to elucidate the pathophysiological sequence of events that lead to ASD.

Diagnosing autism in neurobiological research studies

Autism spectrum disorder (ASD) is by definition a complex and heterogeneous disorder. Variation in factors such as developmental level, language ability and IQ further complicate the presentation of symptoms. Clinical research and basic science must continue to inform each other's questions to help address the heterogeneity inherent to the disorder. This review uses a clinical perspective to outline the common tools and best practices for diagnosing and characterizing ASD in a research setting. We discuss considerations for classifying research populations, including language ability and IQ and examine the advantages and disadvantages of different psychometric measurements. Ultimately, the contribution of multiple sources of data representing different perspectives is crucial for interpreting and understanding the ASD phenotype.

Recording infant ERP data for cognitive research

Researchers from different backgrounds have an increasing interest in investigating infant cognitive development using electroencephalogram (EEG) recordings. Although EEG measurements are suitable for infants, the method poses several challenges including setting up an infant-friendly, but interference-free lab environment and designing age-appropriate stimuli and paradigms. Certain specifics of infant EEG data have to be considered when deriving event-related potentials (ERPs) to investigate cognitive processes in the developing brain. The present article summarizes the practical aspects of conducting ERP research with infants and describes how researchers typically deal with the specific challenges entailed in this work.

A meta-analysis investigating factors underlying attrition rates in infant ERP studies

In this meta-analysis, we examined interrelationships between characteristics of infant event-related potential (ERP) studies and their attrition rates. One-hundred and forty-nine published studies provided information on 314 experimental groups of which 181 provided data on attrition. A random effects meta-analysis revealed a high average attrition rate of 49.2%. Additionally, we used meta-regression for 178 groups with attrition data to analyze which variables best explained attrition variance. Our main findings were that the nature of the stimuli-visual, auditory, or combined as well as if stimuli were animated-influenced exclusion rates from the final analysis and that infant age did not alter attrition rates.