EEG Analytics for Early Detection of Autism Spectrum Disorder: A data-driven approach

Coarse-graining and the Haar wavelet transform for multiscale analysis

BACKGROUND

Multiscale entropy (MSE) has become increasingly common as a quantitative tool for analysis of physiological signals. The MSE computation involves first decomposing a signal into multiple sub-signal 'scales' using a coarse-graining algorithm.

METHODS

The coarse-graining algorithm averages adjacent values in a time series to produce a coarser scale time series. The Haar wavelet transform convolutes a time series with a scaled square wave function to produce an approximation which is equivalent to averaging points.

RESULTS

Coarse-graining is mathematically identical to the Haar wavelet transform approximations. Thus, multiscale entropy is entropy computed on sub-signals derived from approximations of the Haar wavelet transform. By describing coarse-graining algorithms properly as Haar wavelet transforms, the meaning of 'scales' as wavelet approximations becomes transparent. The computed value of entropy is different with different wavelet basis functions, suggesting further research is needed to determine optimal methods for computing multiscale entropy.

CONCLUSION

Coarse-graining is mathematically identical to Haar wavelet approximations at power-of-two scales. Referring to coarse-graining as a Haar wavelet transform motivates research into the optimal approach to signal decomposition for entropy analysis.

Searching for a "Brain Signature" of Neurodevelopmental Disorders: Event-Related Potentials and the Quest for Biomarkers of Cognition

SUMMARY

This review summarizes main applications of event-related potentials (ERPs) to the study of cognitive processes in persons with neurodevelopmental disorders, for whom traditional behavioral assessments may not be suitable. A brief introduction to the ERPs is followed by a review of empirical studies using passive ERP paradigms to address three main questions: characterizing individual differences, predicting risk for poor developmental outcomes, and documenting treatment effects in persons with neurodevelopmental disorders. Evidence across studies reveals feasibility of ERP methodology in a wide range of clinical populations and notes consistently stronger brain-behavior associations involving ERP measures of higher-order cognition compared with sensory-perceptual processes. The final section describes the current limitations of ERP methodology that need to be addressed before it could be used as a clinical tool and highlights the needed steps toward translating ERPs from group-level research applications to individually interpretable clinical use.

Neonatal chemokine markers predict subsequent diagnosis of autism spectrum disorder and delayed development

Immune dysregulation has been found to be related to a diagnosis of autism spectrum disorder (ASD). However, investigations in very early childhood examining immunological abnormalities such as altered neonatal cytokine/chemokine profiles in association with an aberrant developmental trajectory, are sparse. We assessed neonatal blood spots from 398 children, including 171 with ASD, which were subdivided according to severity (121 severe, 50 mild/moderate) and cognitive/adaptive levels (144 low-functioning, 27 typical to high-functioning). The remainder were 69 children with developmental delay (DD), and 158 with typical development (TD), who served as controls in the Childhood Autism Risks from Genetics and the Environment (CHARGE) study. Exploratory analysis suggested that, in comparisons with TD and DD, CTACK (CCL27) and MPIF-1 (CCL23), respectively, were independently associated with ASD. Higher neonatal levels of CTACK were associated with decreased odds of ASD compared to TD (odds ratio [OR] = 0.40, 95% confidence interval [Cl] 0.21, 0.77), whereas higher levels of MPIF-1 were associated with increased odds of ASD (OR = 2.38, 95% Cl 1.42, 3.98) compared to DD but not to TD. MPIF-1 was positively associated with better scores in several developmental domains. Dysregulation of chemokine levels in early life can impede normal immune and neurobehavioral development, which can lead to diagnosis of ASD or DD. This study collectively suggests that certain peripheral chemokines at birth are associated with ASD progression during childhood and that children with ASD and DD have distinct neonatal chemokine profiles that can differentiate their diagnoses.

Abnormal brain state distribution and network connectivity in a SYNGAP1 rat model

Mutations in the SYNGAP1 gene are one of the common predictors of neurodevelopmental disorders, commonly resulting in individuals developing autism, intellectual disability, epilepsy, and sleep deficits. EEG recordings in neurodevelopmental disorders show potential to identify clinically translatable biomarkers to both diagnose and track the progress of novel therapeutic strategies, as well as providing insight into underlying pathological mechanisms. In a rat model of SYNGAP1 haploinsufficiency in which the exons encoding the calcium/lipid binding and GTPase-activating protein domains have been deleted (Syngap^+/Δ-GAP ), we analysed the duration and occurrence of wake, non-rapid eye movement and rapid eye movement brain states during 6 h multi-electrode EEG recordings. We find that although Syngap^+/Δ-GAP animals spend an equivalent percent time in wake and sleep states, they have an abnormal brain state distribution as the number of wake and non-rapid eye movement bouts are reduced and there is an increase in the average duration of both wake and non-rapid eye movement epochs. We perform connectivity analysis by calculating the average imaginary coherence between electrode pairs at varying distance thresholds during these states. In group averages from pairs of electrodes at short distances from each other, a clear reduction in connectivity during non-rapid eye movement is present between 11.5 Hz and 29.5 Hz, a frequency range that overlaps with sleep spindles, oscillatory phenomena thought to be important for normal brain function and memory consolidation. Sleep abnormalities were mostly uncorrelated to the electrophysiological signature of absence seizures, spike and wave discharges, as was the imaginary coherence deficit. Sleep spindles occurrence, amplitude, power and spread across multiple electrodes were not reduced in Syngap^+/Δ-GAP rats, with only a small decrease in duration detected. Nonetheless, by analysing the dynamic imaginary coherence during sleep spindles, we found a reduction in high-connectivity instances between short-distance electrode pairs. Finally comparing the dynamic imaginary coherence during sleep spindles between individual electrode pairs, we identified a group of channels over the right somatosensory, association and visual cortices that have a significant reduction in connectivity during sleep spindles in mutant animals. This matched a significant reduction in connectivity during spindles when averaged regional comparisons were made. These data suggest that Syngap^+/Δ-GAP rats have altered brain state dynamics and EEG connectivity, which may have clinical relevance for SYNGAP1 haploinsufficiency in humans.

Looking Back at the Next 40 Years of ASD Neuroscience Research

During the last 40 years, neuroscience has become one of the most central and most productive approaches to investigating autism. In this commentary, we assemble a group of established investigators and trainees to review key advances and anticipated developments in neuroscience research across five modalities most commonly employed in autism research: magnetic resonance imaging, functional near infrared spectroscopy, positron emission tomography, electroencephalography, and transcranial magnetic stimulation. Broadly, neuroscience research has provided important insights into brain systems involved in autism but not yet mechanistic understanding. Methodological advancements are expected to proffer deeper understanding of neural circuitry associated with function and dysfunction during the next 40 years.

Prediction of autism spectrum disorder diagnosis using nonlinear measures of language-related EEG at 6 and 12 months

BACKGROUND

Early identification of autism spectrum disorder (ASD) provides an opportunity for early intervention and improved developmental outcomes. The use of electroencephalography (EEG) in infancy has shown promise in predicting later ASD diagnoses and in identifying neural mechanisms underlying the disorder. Given the high co-morbidity with language impairment, we and others have speculated that infants who are later diagnosed with ASD have altered language learning, including phoneme discrimination. Phoneme learning occurs rapidly in infancy, so altered neural substrates during the first year of life may serve as early, accurate indicators of later autism diagnosis.

METHODS

Using EEG data collected at two different ages during a passive phoneme task in infants with high familial risk for ASD, we compared the predictive accuracy of a combination of feature selection and machine learning models at 6 months (during native phoneme learning) and 12 months (after native phoneme learning), and we identified a single model with strong predictive accuracy (100%) for both ages. Samples at both ages were matched in size and diagnoses (n = 14 with later ASD; n = 40 without ASD). Features included a combination of power and nonlinear measures across the 10‑20 montage electrodes and 6 frequency bands. Predictive features at each age were compared both by feature characteristics and EEG scalp location. Additional prediction analyses were performed on all EEGs collected at 12 months; this larger sample included 67 HR infants (27 HR-ASD, 40 HR-noASD).

RESULTS

Using a combination of Pearson correlation feature selection and support vector machine classifier, 100% predictive diagnostic accuracy was observed at both 6 and 12 months. Predictive features differed between the models trained on 6- versus 12-month data. At 6 months, predictive features were biased to measures from central electrodes, power measures, and frequencies in the alpha range. At 12 months, predictive features were more distributed between power and nonlinear measures, and biased toward frequencies in the beta range. However, diagnosis prediction accuracy substantially decreased in the larger, more behaviorally heterogeneous 12-month sample.

CONCLUSIONS

These results demonstrate that speech processing EEG measures can facilitate earlier identification of ASD but emphasize the need for age-specific predictive models with large sample sizes to develop clinically relevant classification algorithms.

Neural Correlates of Infant Face Processing and Later Emerging Autism Symptoms in Fragile X Syndrome

Fragile X syndrome (FXS) is the leading known genetic cause of autism spectrum disorder (ASD) with 60-74% of males with FXS meeting diagnostic criteria for ASD. Infants with FXS have demonstrated atypical neural responses during face processing that are unique from both typically developing, low-risk infants and infants at high familial risk for ASD (i.e., infants siblings of children with ASD). In the current study, event-related potential (ERP) responses during face processing measured at 12 months of age were examined in relation to ASD symptoms measured at ~48 months of age in participants with FXS, as well as siblings of children with ASD and low-risk control participants. Results revealed that greater amplitude N290 responses in infancy were associated with more severe ASD symptoms in childhood in FXS and in siblings of children with ASD. This pattern of results was not observed for low-risk control participants. Reduced Nc amplitude was associated with more severe ASD symptoms in participants with FXS but was not observed in the other groups. This is the first study to examine ASD symptoms in childhood in relation to infant ERP responses in FXS. Results indicate that infant ERP responses may be predictive of later symptoms of ASD in FXS and the presence of both common and unique pathways to ASD in etiologically-distinct high-risk groups is supported (i.e., syndromic risk vs. familial risk).

A versatile computational algorithm for time-series data analysis and machine-learning models

Here we introduce Local Topological Recurrence Analysis (LoTRA), a simple computational approach for analyzing time-series data. Its versatility is elucidated using simulated data, Parkinsonian gait, and in vivo brain dynamics. We also show that this algorithm can be used to build a remarkably simple machine-learning model capable of outperforming deep-learning models in detecting Parkinson’s disease from a single digital handwriting test.

Can Autism Be Diagnosed with Artificial Intelligence? A Narrative Review

Radiomics with deep learning models have become popular in computer-aided diagnosis and have outperformed human experts on many clinical tasks. Specifically, radiomic models based on artificial intelligence (AI) are using medical data (i.e., images, molecular data, clinical variables, etc.) for predicting clinical tasks such as autism spectrum disorder (ASD). In this review, we summarized and discussed the radiomic techniques used for ASD analysis. Currently, the limited radiomic work of ASD is related to the variation of morphological features of brain thickness that is different from texture analysis. These techniques are based on imaging shape features that can be used with predictive models for predicting ASD. This review explores the progress of ASD-based radiomics with a brief description of ASD and the current non-invasive technique used to classify between ASD and healthy control (HC) subjects. With AI, new radiomic models using the deep learning techniques will be also described. To consider the texture analysis with deep CNNs, more investigations are suggested to be integrated with additional validation steps on various MRI sites.

Digital Endpoints: Definition, Benefits, and Current Barriers in Accelerating Development and Adoption

The assessment of health and disease requires a set of criteria to define health status and progression. These health measures are referred to as "endpoints." A "digital endpoint" is defined by its use of sensor-generated data often collected outside of a clinical setting such as in a patient's free-living environment. Applicable sensors exist in an array of devices and can be applied in a diverse set of contexts. For example, a smartphone's microphone might be used to diagnose or predict mild cognitive impairment due to Alzheimer's disease or a wrist-worn activity monitor (such as those found in smartwatches) may be used to measure a drug's effect on the nocturnal activity of patients with sickle cell disease. Digital endpoints are generating considerable excitement because they permit a more authentic assessment of the patient's experience, reveal formerly untold realities of disease burden, and can cut drug discovery costs in half. However, before these benefits can be realized, effort must be applied not only to the technical creation of digital endpoints but also to the environment that allows for their development and application. The future of digital endpoints rests on meaningful interdisciplinary collaboration, sufficient evidence that digital endpoints can realize their promise, and the development of an ecosystem in which the vast quantities of data that digital endpoints generate can be analyzed. The fundamental nature of health care is changing. With coronavirus disease 2019 serving as a catalyst, there has been a rapid expansion of home care models, telehealth, and remote patient monitoring. The increasing adoption of these health-care innovations will expedite the requirement for a digital characterization of clinical status as current assessment tools often rely upon direct interaction with patients and thus are not fit for purpose to be administered remotely. With the ubiquity of relatively inexpensive sensors, digital endpoints are positioned to drive this consequential change. It is therefore not surprising that regulators, physicians, researchers, and consultants have each offered their assessment of these novel tools. However, as we further describe later, the broad adoption of digital endpoints will require a cooperative effort. In this article, we present an analysis of the current state of digital endpoints. We also attempt to unify the perspectives of the parties involved in the development and deployment of these tools. We conclude with an interdependent list of challenges that must be collaboratively addressed before these endpoints are widely adopted.

EEG/ERP as a pragmatic method to expand the reach of infant-toddler neuroimaging in HBCD: Promises and challenges

Though electrophysiological measures (EEG and ERP) offer complementary information to MRI and a variety of advantages for studying infants and young children, these measures have not yet been included in large cohort studies of neurodevelopment. This review summarizes the types of EEG and ERP measures that could be used in the HEALthy Brain and Cognitive Development (HBCD) study, and the promises and challenges in doing so. First, we provide brief overview of the use of EEG/ERP for studying the developing brain and discuss exemplar findings, using resting or baseline EEG measures as well as the ERP mismatch negativity (MMN) as exemplars. We then discuss the promises of EEG/ERP such as feasibility, while balancing challenges such as ensuring good signal quality in diverse children with different hair types. We then describe an ongoing multi-site EEG data harmonization from our groups. We discuss the process of alignment and provide preliminary usability data for both resting state EEG data and auditory ERP MMN in diverse samples including over 300 infants and toddlers. Finally, we provide recommendations and considerations for the HBCD study and other studies of neurodevelopment.

The Babytwins Study Sweden (BATSS): A Multi-Method Infant Twin Study of Genetic and Environmental Factors Influencing Infant Brain and Behavioral Development

Twin studies can help us understand the relative contributions of genes and environment to phenotypic trait variation, including attentional and brain activation measures. In terms of applying methodologies such as electroencephalography (EEG) and eye tracking, which are key methods in developmental neuroscience, infant twin studies are almost nonexistent. Here, we describe the Babytwins Study Sweden (BATSS), a multi-method longitudinal twin study of 177 MZ and 134 DZ twin pairs (i.e., 622 individual infants) covering the 5-36 month time period. The study includes EEG, eye tracking and genetics, together with more traditional measures based on in-person testing, direct observation and questionnaires. The results show that interest in participation in research among twin parents is high, despite the comprehensive protocol. DNA analysis from saliva samples was possible in virtually all participants, allowing for both zygosity confirmation and polygenic score analyses. Combining a longitudinal twin design with advanced technologies in developmental cognitive neuroscience and genomics, BATSS represents a new approach in infancy research, which we hope to have impact across multiple disciplines in the coming years.

Detection of an Autism EEG Signature From Only Two EEG Channels Through Features Extraction and Advanced Machine Learning Analysis

BACKGROUND AND OBJECTIVE

In 2 previous studies, we have shown the ability of special machine learning systems applied to standard EEG data in distinguishing children with autism spectrum disorder (ASD) from non-ASD children with an overall accuracy rate of 100% and 98.4%, respectively. Since the equipment routinely available in neonatology units employ few derivations, we were curious to check if just 2 derivations were enough to allow good performance in the same cases of the above-mentioned studies.

METHODS

A continuous segment of artifact-free EEG data lasting 1 minute in ASCCI format from C3 and C4 EEG channels present in 2 previous studies, was used for features extraction and subsequent analyses with advanced machine learning systems. A features extraction software package (Python tsfresh) applied to time-series raw data derived 1588 quantitative features. A special hybrid system called TWIST (Training with Input Selection and Testing), coupling an evolutionary algorithm named Gen-D and a backpropagation neural network, was used to subdivide the data set into training and testing sets as well as to select features yielding the maximum amount of information after a first variable selection performed with linear correlation index threshold.

RESULTS

After this intelligent preprocessing, 12 features were extracted from C3-C4 time-series of study 1 and 36 C3-C4 time-series of study 2 representing the EEG signature. Acting on these features the overall accuracy predictive capability of the best artificial neural network acting as a classifier in deciphering autistic cases from typicals (study 1) and other neuropsychiatric disorders (study 2) resulted in 100 % for study 1 and 94.95 % for study 2.

CONCLUSIONS

The results of this study suggest that also a minor part of EEG contains precious information useful to detect autism if treated with advanced computational algorithms. This could allow in the future to use standard EEG from newborns to check if the ASD signature is already present at birth.

A new neurofeedback training method based on feature space clustering to control EEG features within target clusters

BACKGROUND

Within the most commonly used neurofeedback training methods, a threshold has been defined for each EEG feature wherein subjects' status during training can be assessed according to the given value. In the present study, a neurofeedback training method based on feature-space clustering was proposed in order to assess subjects' status more accurately.

NEW METHOD

Neural gas algorithm was employed for feature space clustering. Then, the clusters were labeled as initial clusters (where the EEG features were placed prior to training) and target (where the EEG features should be shifted towards during training) ones. A scoring index was defined whose value was determined according to subjects' brain activity. This method was simulated in two versions: soft-boundary and hard-boundary based methods.

RESULTS

The results of the present simulation showed that the proposed hard-boundary based version could guide the subjects towards the boundaries of the target clusters and even their status would be stabilized in case of too many changes in subjects' EEG features. In the proposed soft-boundary based version, in case of too many changes in training features, the subjects would not be encouraged and they could be guided towards the target boundaries.

CONCLUSION

The proposed hard-boundary based version could be effective in guiding a subject towards being placed within the boundaries of target clusters and even beyond them if no specific limits exited for EEG features. As well, the soft-boundary based version could be useful when controlling EEG features within a limit.

Predicting Autism in Infancy

Autism spectrum disorder (ASD) is a neurodevelopmental disorder characterized by social communication and interaction deficits and restricted, repetitive patterns of interests and behavior that are evident in early childhood. Its prevalence has grown substantially over the past several decades, with current estimates ranging from 1.7% to 2.5% in the United States.^1,2 This represents more than 1.5 million children with ASD, the vast majority of whom receive or will receive specialized services.² Each year, approximately 100,000 (and growing) individuals with ASD reach adulthood, and many face myriad challenges related to employment, housing, mental health, and overburdened or insufficient support services.^3-5 A host of significant costs can be associated with ASD, from direct costs related to the provision of special education programs, housing, and medical care to indirect costs, such as loss of productivity affecting both individuals with ASD and their families.⁶ Currently, overall lifetime cost of care per person with ASD can exceed $3 million, totaling more than $265 billion annually in the United States and rising to an estimated $1 trillion by 2025.^7,8.

Association between spectral electroencephalography power and autism risk and diagnosis in early development

Autism spectrum disorder (ASD) has its origins in the atypical development of brain networks. Infants who are at high familial risk for, and later diagnosed with ASD, show atypical activity in multiple electroencephalography (EEG) oscillatory measures. However, infant-sibling studies are often constrained by small sample sizes. We used the International Infant EEG Data Integration Platform, a multi-site dataset with 432 participants, including 222 at high-risk for ASD, from whom repeated measurements of EEG were collected between the ages of 3-36 months. We applied a latent growth curve model to test whether familial risk status predicts developmental trajectories of spectral power across the first 3 years of life, and whether these trajectories predict ASD outcome. Change in spectral EEG power in all frequency bands occurred during the first 3 years of life. Familial risk, but not a later diagnosis of ASD, was associated with reduced power at 3 months, and a steeper developmental change between 3 and 36 months in nearly all absolute power bands. ASD outcome was not associated with absolute power intercept or slope. No associations were found between risk or outcome and relative power. This study applied an analytic approach not used in previous prospective biomarker studies of ASD, which was modeled to reflect the temporal relationship between genetic susceptibility, brain development, and ASD diagnosis. Trajectories of spectral power appear to be predicted by familial risk; however, spectral power does not predict diagnostic outcome above and beyond familial risk status. Discrepancies between current results and previous studies are discussed. LAY SUMMARY: Infants with an older sibling who is diagnosed with ASD are at increased risk of developing ASD themselves. This article tested whether EEG spectral power in the first year of life can predict whether these infants did or did not develop ASD.

A spectrogram image based intelligent technique for automatic detection of autism spectrum disorder from EEG

Autism spectrum disorder (ASD) is a developmental disability characterized by persistent impairments in social interaction, speech and nonverbal communication, and restricted or repetitive behaviors. Currently Electroencephalography (EEG) is the most popular tool to inspect the existence of neurological disorders like autism biomarkers due to its low setup cost, high temporal resolution and wide availability. Generally, EEG recordings produce vast amount of data with dynamic behavior, which are visually analyzed by professional clinician to detect autism. It is laborious, expensive, subjective, error prone and has reliability issue. Therefor this study intends to develop an efficient diagnostic framework based on time-frequency spectrogram images of EEG signals to automatically identify ASD. In the proposed system, primarily, the raw EEG signals are pre-processed using re-referencing, filtering and normalization. Then, Short-Time Fourier Transform is used to transform the pre-processed signals into two-dimensional spectrogram images. Afterward those images are evaluated by machine learning (ML) and deep learning (DL) models, separately. In the ML process, textural features are extracted, and significant features are selected using principal component analysis, and feed them to six different ML classifiers for classification. In the DL process, three different convolutional neural network models are tested. The proposed DL based model achieves higher accuracy (99.15%) compared to the ML based model (95.25%) on an ASD EEG dataset and also outperforms existing methods. The findings of this study suggest that the DL based structure could discover important biomarkers for efficient and automatic diagnosis of ASD from EEG and may assist to develop computer-aided diagnosis system.

Measuring the effects of sleep on epileptogenicity with multifrequency entropy

OBJECTIVE

We demonstrate that multifrequency entropy gives insight into the relationship between epileptogenicity and sleep, and forms the basis for an improved measure of medical assessment of sleep impairment in epilepsy patients.

METHODS

Multifrequency entropy was computed from electroencephalography measurements taken from 31 children with Benign Epilepsy with Centrotemporal Spikes and 31 non-epileptic controls while awake and during sleep. Values were compared in the epileptic zone and away from the epileptic zone in various sleep stages.

RESULTS

We find that (I) in lower frequencies, multifrequency entropy decreases during non-rapid eye movement sleep stages when compared with wakefulness in a general population of pediatric patients, (II) patients with Benign Epilepsy with Centrotemporal Spikes had lower multifrequency entropy across stages of sleep and wakefulness, and (III) the epileptic regions of the brain exhibit lower multifrequency entropy patterns than the rest of the brain in epilepsy patients.

CONCLUSIONS

Our results show that multifrequency entropy decreases during sleep, particularly sleep stage 2, confirming, in a pediatric population, an association between sleep, lower multifrequency entropy, and increased likelihood of seizure.

SIGNIFICANCE

We observed a correlation between lowered multifrequency entropy and increased epileptogenicity that lays preliminary groundwork for the detection of a digital biomarker for epileptogenicity.

Automated ASD detection using hybrid deep lightweight features extracted from EEG signals

BACKGROUND

Autism spectrum disorder is a common group of conditions affecting about one in 54 children. Electroencephalogram (EEG) signals from children with autism have a common morphological pattern which makes them distinguishable from normal EEG. We have used this type of signal to design and implement an automated autism detection model.

MATERIALS AND METHOD

We propose a hybrid lightweight deep feature extractor to obtain high classification performance. The system was designed and tested with a big EEG dataset that contained signals from autism patients and normal controls. (i) A new signal to image conversion model is presented in this paper. In this work, features are extracted from EEG signal using one-dimensional local binary pattern (1D_LBP) and the generated features are utilized as input of the short time Fourier transform (STFT) to generate spectrogram images. (ii) The deep features of the generated spectrogram images are extracted using a combination of pre-trained MobileNetV2, ShuffleNet, and SqueezeNet models. This method is named hybrid deep lightweight feature generator. (iii) A two-layered ReliefF algorithm is used for feature ranking and feature selection. (iv) The most discriminative features are fed to various shallow classifiers, developed using a 10-fold cross-validation strategy for automated autism detection.

RESULTS

A support vector machine (SVM) classifier reached 96.44% accuracy based on features from the proposed model.

CONCLUSIONS

The results strongly indicate that the proposed hybrid deep lightweight feature extractor is suitable for autism detection using EEG signals. The model is ready to serve as part of an adjunct tool that aids neurologists during autism diagnosis in medical centers.

Children ASD Evaluation Through Joint Analysis of EEG and Eye-Tracking Recordings With Graph Convolution Network

Recent advances in neuroscience indicate that analysis of bio-signals such as rest state electroencephalogram (EEG) and eye-tracking data can provide more reliable evaluation of children autism spectrum disorder (ASD) than traditional methods of behavior measurement relying on scales do. However, the effectiveness of the new approaches still lags behind the increasing requirement in clinical or educational practices as the “bio-marker” information carried by the bio-signal of a single-modality is likely insufficient or distorted. This study proposes an approach to joint analysis of EEG and eye-tracking for children ASD evaluation. The approach focuses on deep fusion of the features in two modalities as no explicit correlations between the original bio-signals are available, which also limits the performance of existing methods along this direction. First, the synchronization measures, information entropy, and time-frequency features of the multi-channel EEG are derived. Then a random forest applies to the eye-tracking recordings of the same subjects to single out the most significant features. A graph convolutional network (GCN) model then naturally fuses the two group of features to differentiate the children with ASD from the typically developed (TD) subjects. Experiments have been carried out on the two types of the bio-signals collected from 42 children (21 ASD and 21 TD subjects, 3–6 years old). The results indicate that (1) the proposed approach can achieve an accuracy of 95% in ASD detection, and (2) strong correlations exist between the two bio-signals collected even asynchronously, in particular the EEG synchronization against the face related/joint attentions in terms of covariance.

Clinical and Translational Implications of an Emerging Developmental Substructure for Autism

A vast share of the population-attributable risk for autism relates to inherited polygenic risk. A growing number of studies in the past five years have indicated that inherited susceptibility may operate through a finite number of early developmental liabilities that, in various permutations and combinations, jointly predict familial recurrence of the convergent syndrome of social communication disability that defines the condition. Here, we synthesize this body of research to derive evidence for a novel developmental substructure for autism, which has profound implications for ongoing discovery efforts to elucidate its neurobiological causes, and to inform future clinical and biomarker studies, early interventions, and personalized approaches to therapy.

Screening for Behavioral Signs of Autism Spectrum Disorder in 9-Month-Old Infant Siblings

Despite considerable progress in characterizing the early signs of autism spectrum disorder (ASD), more remains to be learned about how symptoms emerge in the first year of life. Parents with a new baby who already had at least one biological child diagnosed with ASD (high-risk) or no family history of ASD (low-risk) completed two measures when their baby was 9 months of age, the Autism Parent Screen for Infants (APSI) questionnaire and the interview-based Parent Concerns Form. Children underwent a blinded independent diagnostic assessment for ASD at age 3 years. Total scores on the APSI and the Parent Concerns Form were both able to independently differentiate high-risk children who were later diagnosed with ASD from other high-risk and low-risk children who were not. Using logistic regression, we found that the total score on the APSI predicted ASD outcomes at age 3 with 70% accuracy, but the Parent Concerns Form did not contribute any unique variance when the APSI was already in the model. The results suggest that the APSI identifies early features predictive of ASD in high-risk infants and can be used to flag them for targeted follow-up and screening.

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.

Cortical maturation from childhood to adolescence is reflected in resting state EEG signal complexity

Endogenous cortical fluctuations captured by electroencephalograms (EEGs) reflect activity in large-scale brain networks that exhibit dynamic patterns over multiple time scales. Developmental changes in the coordination and integration of brain function leads to greater complexity in population level neural dynamics. In this study we examined multiscale entropy, a measure of signal complexity, in resting-state EEGs in a large (N = 405) cross-sectional sample of children and adolescents (9–16 years). Our findings showed consistent age-dependent increases in EEG complexity that are distributed across multiple temporal scales and spatial regions. Developmental changes were most robust as the age gap between groups increased, particularly between late childhood and adolescence, and were most prominent over fronto-central scalp regions. These results suggest that the transition from late childhood to adolescence is characterized by age-dependent changes in the underlying complexity of endogenous brain networks.

Toward a Developmental Nosology for Disruptive Mood Dysregulation Disorder in Early Childhood

OBJECTIVE

Disruptive mood dysregulation disorder (DMDD) in DSM, characterized by severe, chronic irritability, currently excludes children <6 years of age. However, capitalizing on a burgeoning developmental science base to differentiate clinically salient irritability in young children may enable earlier identification. The objective of this study was to advance an empirically derived framework for early childhood DMDD (EC-DMDD) by modeling and validating DMDD patterns in early childhood and generating clinically informative, optimized behaviors with thresholds.

METHOD

Data (N = 425) were from 3 longitudinal assessments of the MAPS Study, spanning preschool (means = 4.7 and 5.5 years) to early school age (mean = 6.8 years). The Multidimensional Assessment Profile of Disruptive Behavior (MAP-DB) Temper Loss scale captured irritability, the Family Life Impairment Scale (FLIS) assessed cross-domain impairment at the preschool time points and the Schedule for Affective Disorders and Schizophrenia for School-Age Children (K-SADS) was used to assess clinical status at early school age. Latent transition analyses differentiated children with EC-DMDD from children with low, transient, or nonimpairing irritability.

RESULTS

Developmental patterning of irritability proved important for normal:abnormal differentiation. Of children, 27% had initially high irritability, but only two-thirds of these were persistently highly irritable. Thus, "false positives" based on a single screen would be substantial. Yet, "false negatives" are low, as <1% of children with baseline low irritability demonstrated later high irritability. Based on the sequential preschool-age time points, 6.7% of children were identified with EC-DMDD, characterized by persistent irritability with pervasive impairment, similar to prevalence at older ages. Specific behaviors included low frustration tolerance; dysregulated, developmentally unexpectable tantrums; and sustained irritable mood, all of which sensitively (0.85-0.96) and specifically (0.80-0.91) identified EC-DMDD. EC-DMDD predicted irritability-related syndromes (DMDD, oppositional defiant disorder) at early school age better than downward extension of DSM DMDD criteria to preschool age.

CONCLUSION

These findings provide empirical thresholds for preschool-age clinical identification of DMDD patterns. The results lay the foundation for validation of DMDD in early childhood and inform revision of DSM criteria.

Early infant behavioural correlates of social skills in adolescents

BACKGROUND

Autism Spectrum Disorder (ASD) is a neurodevelopmental disorder typically diagnosed after the second year of life; however, differences in brain structure and function associated with ASD have been ascertained in early infancy. Identifying behavioural markers of ASD risk in early infancy has the potential to facilitate early detection and intervention.

OBJECTIVES

We examined associations between infant behaviour and adolescent behaviours associated with ASD.

METHODS

Analyses leveraged data available on 370 participants from the New Bedford Cohort, a sociodemographically diverse prospective birth cohort of children born from 1993 to 1998 to mothers residing near the New Bedford Harbor Superfund site in Massachusetts. Longitudinal assessments were used to examine the associations between behaviours when children were approximately 2 weeks old (measured by the Neonatal Behavioral Assessment Scale [NBAS]), and subsequent maladaptive behaviours associated with ASD at approximately 15 years old [measured by the Behavior Assessment System for Children, 2nd Edition-Teacher Rating Scale (BASC-2 TRS) scores which are standardised to a mean (SD) of 50 (10)].

RESULTS

Poorer performance on select individual items and cluster scales of the NBAS was associated with an increase in behaviours associated with ASD in adolescents. Associations were strongest for neonatal measures of self-regulation, response to auditory input, and autonomic nervous system regulation. For example, in covariate-adjusted models, infants with Regulation of State NBAS cluster scores in the lowest tertile (poorest performance) compared to infants with scores in the higher two tertiles had adolescent BASC-2 TRS Developmental Social Disorders T-scores that were 2.9 points higher (95% CI: 0.8, 4.9), indicating more behaviours associated with ASD.

CONCLUSION

The NBAS is an established and accessible instrument that assesses a broad range of behaviours in very young infants, and may be a useful tool for newborn assessments of developmental risk, including risk of ASD-associated behaviours.

EEG signatures of cognitive and social development of preschool children-a systematic review

Background

Early identification of preschool children who are at risk of faltering in their development is essential to ensuring that all children attain their full potential. Electroencephalography (EEG) has been used to measure neural correlates of cognitive and social development in children for decades. Effective portable and low-cost EEG devices increase the potential of its use to assess neurodevelopment in children at scale and particularly in low-resource settings. We conducted a systematic review aimed to synthesise EEG measures of cognitive and social development in 2-5-year old children. Our secondary aim was to identify how these measures differ across a) the course of development within this age range, b) gender and c) socioeconomic status (SES).

Methods and findings

A systematic literature search identified 51 studies for inclusion in this review. Data relevant to the primary and secondary aims was extracted from these studies and an assessment for risk of bias was done, which highlighted the need for harmonisation of EEG data collection and analysis methods across research groups and more detailed reporting of participant characteristics. Studies reported on the domains of executive function (n = 22 papers), selective auditory attention (n = 9), learning and memory (n = 5), processing of faces (n = 7) and emotional stimuli (n = 8). For papers investigating executive function and selective auditory attention, the most commonly reported measures were alpha power and the amplitude and latency of positive (P1, P2, P3) and negative (N1, N2) deflections of event related potential (ERPs) components. The N170 and P1 ERP components were the most commonly reported neural responses to face and emotional faces stimuli. A mid-latency negative component and positive slow wave were used to index learning and memory, and late positive potential in response to emotional non-face stimuli. While almost half the studies described changes in EEG measures across age, only eight studies disaggregated results based on gender, and six included children from low income households to assess the impact of SES on neurodevelopment. No studies were conducted in low- and middle-income countries.

Conclusion

This review has identified power across the EEG spectrum and ERP components to be the measures most commonly reported in studies in which preschool children engage in tasks indexing cognitive and social development. It has also highlighted the need for additional research into their changes across age and based on gender and SES.

Theories of Autism and Autism Treatment from the DSM III Through the Present and Beyond: Impact on Research and Practice

The purely descriptive definition of autism introduced by the DSM III in 1980 marked a departure from previous DSM editions, which mixed phenomenological descriptions with psychoanalytic theories of etiology. This provided a blank slate upon which a variety of novel theories emerged to conceptualize autism and its treatment in the following four decades. In this article we examine the contribution of these different theoretical orientations with a focus on their impact on research and practice, areas of overlap and conflict between current theories, and their relevance in the context of the evolving landscape of scientific knowledge and societal views of autism.

Prevention in Autism Spectrum Disorder: A Lifelong Focused Approach

Autism Spectrum Disorder (ASD) is a complex highly heritable disorder, in which multiple environmental factors interact with the genes to increase its risk and lead to variable clinical presentations and outcomes. Furthermore, the inherent fundamental deficits of ASD in social attention and interaction critically diverge children from the typical pathways of learning, "creating" what we perceive as autism syndrome during the first three years of life. Later in life, training and education, the presence and management of comorbidities, as well as social and vocational support throughout the lifespan, will define the quality of life and the adaptation of an individual with ASD. Given the overall burden of ASD, prevention strategies seem like a cost-effective endeavour that we have to explore. In this paper, we take a life course approach to prevention. We will review the possibilities of the management of risk factors from preconception until the perinatal period, that of early intervention in the first three years of life and that of effective training and support from childhood until adulthood.

Machine Learning Methods for Diagnosing Autism Spectrum Disorder and Attention- Deficit/Hyperactivity Disorder Using Functional and Structural MRI: A Survey

Here we summarize recent progress in machine learning model for diagnosis of Autism Spectrum Disorder (ASD) and Attention-deficit/Hyperactivity Disorder (ADHD). We outline and describe the machine-learning, especially deep-learning, techniques that are suitable for addressing research questions in this domain, pitfalls of the available methods, as well as future directions for the field. We envision a future where the diagnosis of ASD, ADHD, and other mental disorders is accomplished, and quantified using imaging techniques, such as MRI, and machine-learning models.

Cell Type-Specific Predictive Models Perform Prioritization of Genes and Gene Sets Associated With Autism

Bulk transcriptomic analyses of autism spectrum disorder (ASD) have revealed dysregulated pathways, while the brain cell type-specific molecular pathology of ASD still needs to be studied. Machine learning-based studies can be conducted for ASD, prioritizing high-confidence gene candidates and promoting the design of effective interventions. Using human brain nucleus gene expression of ASD and controls, we construct cell type-specific predictive models for ASD based on individual genes and gene sets, respectively, to screen cell type-specific ASD-associated genes and gene sets. These two kinds of predictive models can predict the diagnosis of a nucleus with known cell type. Then, we construct a multi-label predictive model for predicting the cell type and diagnosis of a nucleus at the same time. Our findings suggest that layer 2/3 and layer 4 excitatory neurons, layer 5/6 cortico-cortical projection neurons, parvalbumin interneurons, and protoplasmic astrocytes are preferentially affected in ASD. The functions of genes with predictive power for ASD are different and the top important genes are distinct across different cells, highlighting the cell-type heterogeneity of ASD. The constructed predictive models can promote the diagnosis of ASD, and the prioritized cell type-specific ASD-associated genes and gene sets may be used as potential biomarkers of ASD.

Electroencephalography in Assessment of Autism Spectrum Disorders: A Review

Electroencephalography (EEG) can further out our understanding of autistic spectrum disorders (ASD) neurophysiology. Epilepsy and ASD comorbidity range between 5 and 46%, but its temporal relationship, causal mechanisms and interplay with intellectual disability are still unknown. Epileptiform discharges with or without seizures go as high as 60%, and associate with epileptic encephalopathies, conceptual term suggesting that epileptic activity can lead to cognitive and behavioral impairment beyond the underlying pathology. Seizures and ASD may be the result of similar mechanisms, such as abnormalities in GABAergic fibers or GABA receptor function. Epilepsy and ASD are caused by a number of genetic disorders and variations that induce such dysregulation. Similarly, initial epilepsy may influence synaptic plasticity and cortical connection, predisposing a growing brain to cognitive delays and behavioral abnormalities. The quantitative EEG techniques could be a useful tool in detecting and possibly measuring dysfunctions in specific brain regions and neuronal regulation in ASD. Power spectra analysis reveals a U-shaped pattern of power abnormalities, with excess power in the low and high frequency bands. These might be the consequence of a complicated network of neurochemical changes affecting the inhibitory GABAergic interneurons and their regulation of excitatory activity in pyramidal cells. EEG coherence studies of functional connectivity found general local over-connectivity and long-range under-connectivity between different brain areas. GABAergic interneuron growth and connections are presumably impaired in the prefrontal and temporal cortices in ASD, which is important for excitatory/inhibitory balance. Recent advances in quantitative EEG data analysis and well-known epilepsy ASD co-morbidity consistently indicate a role of aberrant GABAergic transmission that has consequences on neuronal organization and connectivity especially in the frontal cortex.

Early screening of autism spectrum disorder using cry features

The increase in the number of children with autism and the importance of early autism intervention has prompted researchers to perform automatic and early autism screening. Consequently, in the present paper, a cry-based screening approach for children with Autism Spectrum Disorder (ASD) is introduced which would provide both early and automatic screening. During the study, we realized that ASD specific features are not necessarily observable in all children with ASD and in all instances collected from each child. Therefore, we proposed a new classification approach to be able to determine such features and their corresponding instances. To test the proposed approach a set of data relating to children between 18 to 53 months which had been recorded using high-quality voice recording devices and typical smartphones at various locations such as homes and daycares was studied. Then, after preprocessing, the approach was used to train a classifier, using data for 10 boys with ASD and 10 Typically Developed (TD) boys. The trained classifier was tested on the data of 14 boys and 7 girls with ASD and 14 TD boys and 7 TD girls. The sensitivity, specificity, and precision of the proposed approach for boys were 85.71%, 100%, and 92.85%, respectively. These measures were 71.42%, 100%, and 85.71% for girls, respectively. It was shown that the proposed approach outperforms the common classification methods. Furthermore, it demonstrated better results than the studies which used voice features for screening ASD. To pilot the practicality of the proposed approach for early autism screening, the trained classifier was tested on 57 participants between 10 to 18 months. These 57 participants consisted of 28 boys and 29 girls and the results were very encouraging for the use of the approach in early ASD screening.

Brief Report: Preliminary Feasibility of the TEDI: A Novel Parent-Administered Telehealth Assessment for Autism Spectrum Disorder Symptoms in the First Year of Life

Families with early concerns about infant symptoms of ASD have limited access to experienced professionals for screening and guidance. Telehealth has been used to reduce access disparities in other pediatric populations and has shown promise in parent-implemented interventions for ASD. We investigated the feasibility of a novel level-2 telehealth assessment of infants' early social communication and ASD symptoms, the Telehealth Evaluation of Development for Infants (TEDI). Parents of eleven infants aged 6-12 months were coached to administer specific semi-structured behavioral probes. Initial feasibility, reliability, and acceptability benchmarks were met. These findings suggest the feasibility of screening infants via telehealth, and are supportive of further large-scale efforts to validate this method for longitudinal monitoring of symptomatic infants in community settings.

An fMRI Feature Selection Method Based on a Minimum Spanning Tree for Identifying Patients with Autism

Autism spectrum disorder (ASD) is a neurodevelopmental disorder originating in infancy and childhood that may cause language barriers and social difficulties. However, in the diagnosis of ASD, the current machine learning methods still face many challenges in determining the location of biomarkers. Here, we proposed a novel feature selection method based on the minimum spanning tree (MST) to seek neuromarkers for ASD. First, we constructed an undirected graph with nodes of candidate features. At the same time, a weight calculation method considering both feature redundancy and discriminant ability was introduced. Second, we utilized the Prim algorithm to construct the MST from the initial graph structure. Third, the sum of the edge weights of all connected nodes was sorted for each node in the MST. Then, N features corresponding to the nodes with the first N smallest sum were selected as classification features. Finally, the support vector machine (SVM) algorithm was used to evaluate the discriminant performance of the aforementioned feature selection method. Comparative experiments results show that our proposed method has improved the ASD classification performance, i.e., the accuracy, sensitivity, and specificity were 86.7%, 87.5%, and 85.7%, respectively.

Principles for Guiding the Selection of Early Childhood Neurodevelopmental Risk and Resilience Measures: HEALthy Brain and Child Development Study as an Exemplar

The vast individual differences in the developmental origins of risk and resilience pathways combined with sophisticated capabilities of big data science increasingly point to the imperative of large, neurodevelopmental consortia to capture population heterogeneity and key variations in developmental trajectories. At the same time, such large-scale population-based designs involving multiple independent sites also must weigh competing demands. For example, the need for efficient, scalable assessment strategies must be balanced with the need for nuanced, developmentally sensitive phenotyping optimized for linkage to neural mechanisms and specification of common and distinct exposure pathways. Standardized epidemiologic batteries designed for this purpose such as PhenX (consensus measures for Phenotypes and eXposures) and the National Institutes of Health (NIH) Toolbox provide excellent "off the shelf" assessment tools that are well-validated and enable cross-study comparability. However, these standardized toolkits can also constrain ability to leverage advances in neurodevelopmental measurement over time, at times disproportionately advantaging established measures. In addition, individual consortia often expend exhaustive effort "reinventing the wheel," which is inefficient and fails to fully maximize potential synergies with other like initiatives. To address these issues, this paper lays forth an early childhood neurodevelopmental assessment strategy, guided by a set of principles synthesizing developmental and pragmatic considerations generated by the Neurodevelopmental Workgroup of the HEALthy Brain and Child Development (HBCD) Planning Consortium. These principles emphasize characterization of both risk- and resilience-promoting processes. Specific measurement recommendations to HBCD are provided to illustrate application. However, principles are intended as a guiding framework to transcend any particular initiative as a broad neurodevelopmentally informed, early childhood assessment strategy for large-scale consortia science.

Factors influencing classification of frequency following responses to speech and music stimuli

Successful mapping of meaningful labels to sound input requires accurate representation of that sound's acoustic variances in time and spectrum. For some individuals, such as children or those with hearing loss, having an objective measure of the integrity of this representation could be useful. Classification is a promising machine learning approach which can be used to objectively predict a stimulus label from the brain response. This approach has been previously used with auditory evoked potentials (AEP) such as the frequency following response (FFR), but a number of key issues remain unresolved before classification can be translated into clinical practice. Specifically, past efforts at FFR classification have used data from a given subject for both training and testing the classifier. It is also unclear which components of the FFR elicit optimal classification accuracy. To address these issues, we recorded FFRs from 13 adults with normal hearing in response to speech and music stimuli. We compared labeling accuracy of two cross-validation classification approaches using FFR data: (1) a more traditional method combining subject data in both the training and testing set, and (2) a "leave-one-out" approach, in which subject data is classified based on a model built exclusively from the data of other individuals. We also examined classification accuracy on decomposed and time-segmented FFRs. Our results indicate that the accuracy of leave-one-subject-out cross validation approaches that obtained in the more conventional cross-validation classifications while allowing a subject's results to be analysed with respect to normative data pooled from a separate population. In addition, we demonstrate that classification accuracy is highest when the entire FFR is used to train the classifier. Taken together, these efforts contribute key steps toward translation of classification-based machine learning approaches into clinical practice.

Predictive modeling of neurobehavioral state and trait variation across development

A key goal of human neurodevelopmental research is to map neural and behavioral trajectories across both health and disease. A growing number of developmental consortia have begun to address this gap by providing open access to cross-sectional and longitudinal 'big data' repositories. However, it remains challenging to develop models that enable prediction of both within-subject and between-subject neurodevelopmental variation. Here, we present a conceptual and analytical perspective of two essential ingredients for mapping neurodevelopmental trajectories: state and trait components of variance. We focus on mapping variation across a range of neural and behavioral measurements and consider concurrent alterations of state and trait variation across development. We present a quantitative framework for combining both state- and trait-specific sources of neurobehavioral variation across development. Specifically, we argue that non-linear mixed growth models that leverage state and trait components of variance and consider environmental factors are necessary to comprehensively map brain-behavior relationships. We discuss this framework in the context of mapping language neurodevelopmental changes in early childhood, with an emphasis on measures of functional connectivity and their reliability for establishing robust neurobehavioral relationships. The ultimate goal is to statistically unravel developmental trajectories of neurobehavioral relationships that involve a combination of individual differences and age-related changes.

Automatic Identification of High-Risk Autism Spectrum Disorder: A Feasibility Study Using Video and Audio Data Under the Still-Face Paradigm

It is reported that the symptoms of autism spectrum disorder (ASD) could be improved by effective early interventions, which arouses an urgent need for large-scale early identification of ASD. Until now, the screening of ASD has relied on the child psychiatrist to collect medical history and conduct behavioral observations with the help of psychological assessment tools. Such screening measures inevitably have some disadvantages, including strong subjectivity, relying on experts and low-efficiency. With the development of computer science, it is possible to realize a computer-aided screening for ASD and alleviate the disadvantages of manual evaluation. In this study, we propose a behavior-based automated screening method to identify high-risk ASD (HR-ASD) for babies aged 8-24 months. The still-face paradigm (SFP) was used to elicit baby's spontaneous social behavior through a face-to-face interaction, in which a mother was required to maintain a normal interaction to amuse her baby for 2 minutes (a baseline episode) and then suddenly change to the no-reaction and no-expression status with 1 minute (a still-face episode). Here, multiple cues derived from baby's social stress response behavior during the latter episode, including head-movements, facial expressions and vocal characteristics, were statistically analyzed between HR-ASD and typical developmental (TD) groups. An automated identification model of HR-ASD was constructed based on these multi-cue features and the support vector machine (SVM) classifier; moreover, its screening performance was satisfied, for all the accuracy, specificity and sensitivity exceeded 90% on the cases included in this study. The experimental results suggest its feasibility in the early screening of HR-ASD.

Bi-Dimensional Approach Based on Transfer Learning for Alcoholism Pre-disposition Classification via EEG Signals

Recent statistics have shown that the main difficulty in detecting alcoholism is the unreliability of the information presented by patients with alcoholism; this factor confusing the early diagnosis and it can reduce the effectiveness of treatment. However, electroencephalogram (EEG) exams can provide more reliable data for analysis of this behavior. This paper proposes a new approach for the automatic diagnosis of patients with alcoholism and introduces an analysis of the EEG signals from a two-dimensional perspective according to changes in the neural activity, highlighting the influence of high and low-frequency signals. This approach uses a two-dimensional feature extraction method, as well as the application of recent Computer Vision (CV) techniques, such as Transfer Learning with Convolutional Neural Networks (CNN). The methodology to evaluate our proposal used 21 combinations of the traditional classification methods and 84 combinations of recent CNN architectures used as feature extractors combined with the following classical classifiers: Gaussian Naive Bayes, K-Nearest Neighbor (k-NN), Multilayer Perceptron (MLP), Random Forest (RF) and Support Vector Machine (SVM). CNN MobileNet combined with SVM achieved the best results in Accuracy (95.33%), Precision (95.68%), F1-Score (95.24%), and Recall (95.00%). This combination outperformed the traditional methods by up to 8%. Thus, this approach is applicable as a classification stage for computer-aided diagnoses, useful for the triage of patients, and clinical support for the early diagnosis of this disease.

Pain phenotypes classified by machine learning using electroencephalography features

Pain is a multidimensional experience mediated by distributed neural networks in the brain. To study this phenomenon, EEGs were collected from 20 subjects with chronic lumbar radiculopathy, 20 age and gender matched healthy subjects, and 17 subjects with chronic lumbar pain scheduled to receive an implanted spinal cord stimulator. Analysis of power spectral density, coherence, and phase-amplitude coupling using conventional statistics showed that there were no significant differences between the radiculopathy and control groups after correcting for multiple comparisons. However, analysis of transient spectral events showed that there were differences between these two groups in terms of the number, power, and frequency-span of events in a low gamma band. Finally, we trained a binary support vector machine to classify radiculopathy versus healthy subjects, as well as a 3-way classifier for subjects in the 3 groups. Both classifiers performed significantly better than chance, indicating that EEG features contain relevant information pertaining to sensory states, and may be used to help distinguish between pain states when other clinical signs are inconclusive.

Future Directions for Infant Identification and Intervention for Autism Spectrum Disorder from a Transdiagnostic Perspective

By the time they are typically detected, neurodevelopmental disorders like autism spectrum disorder (ASD) are already challenging to treat. Preventive and early intervention strategies in infancy are critical for improving outcomes over the lifespan with significant cost savings. However, the impact of prevention and early intervention efforts is dependent upon our ability to identify infants most appropriate for such interventions. Because there may be significant overlap between prodromal symptoms across neurodevelopmental disorders and child psychopathology more broadly which may wax and wane across development, we contend that the impact of prevention and early intervention efforts will be heightened by identifying early indicators that may overlap across ASD and other commonly co-occurring disorders. This paper summarizes the existing literature on infant symptoms and identification of ASD to demonstrate the ways in which a transdiagnostic perspective could expand the impact of early identification and intervention research and clinical efforts, and to outline suggestions for future empirical research programs addressing current gaps in the identification-to-treatment pipeline. We propose four recommendations for future research that are both grounded in developmental and clinical science and that are scalable for early intervention systems: (1) development of fine-grained, norm-referenced measures of ASD-relevant transdiagnostic behavioral domains; (2) identification of shared and distinct mechanisms influencing the transition from risk to disorder; (3) determination of key cross-cutting treatment strategies (both novel and extracted from existing approaches) effective in targeting specific domains across disorders; and (4) integration of identified measures and treatments into existing service systems.

Investigating electroencephalography signals of autism spectrum disorder (ASD) using Higuchi Fractal Dimension

Autism spectrum disorder (ASD) is a neurodevelopmental disorder with a deficit of social relationships, interaction, sense of imagination, and constrained interests. Early diagnosis of ASD will aid in devising appropriate training procedures and placing those children in the normal stream. The objective of this research is to analyze the brain response for auditory/visual stimuli in Typically Developing (TD) and children with autism through electroencephalography (EEG). Brain dynamics in the EEG signal can be analyzed well with the help of nonlinear feature primitives. Recent research reveals that, application of fractal-based techniques proves to be effective to estimate of degree of nonlinearity in a signal. This research attempts to analyze the effect of brain dynamics with Higuchi Fractal Dimension (HFD). Also, the performance of the fractal based techniques depends on the selection of proper hyper-parameters involved in it. One of the key parameters involved in computation of HFD is the time interval parameter 'k'. Most of the researches arbitrarily fixes the value of 'k' in the range of all channels. This research proposes an algorithm to estimate the optimal value of the time parameter for each channel. Sub-band analysis was also carried out for the responding channels. Statistical analysis on the experimental reveals that a difference of 30% was observed between autistic and Typically Developing children.

Epigenetic tuning of brain signal entropy in emergent human social behavior

BACKGROUND

How the brain develops accurate models of the external world and generates appropriate behavioral responses is a vital question of widespread multidisciplinary interest. It is increasingly understood that brain signal variability-posited to enhance perception, facilitate flexible cognitive representations, and improve behavioral outcomes-plays an important role in neural and cognitive development. The ability to perceive, interpret, and respond to complex and dynamic social information is particularly critical for the development of adaptive learning and behavior. Social perception relies on oxytocin-regulated neural networks that emerge early in development.

METHODS

We tested the hypothesis that individual differences in the endogenous oxytocinergic system early in life may influence social behavioral outcomes by regulating variability in brain signaling during social perception. In study 1, 55 infants provided a saliva sample at 5 months of age for analysis of individual differences in the oxytocinergic system and underwent electroencephalography (EEG) while listening to human vocalizations at 8 months of age for the assessment of brain signal variability. Infant behavior was assessed via parental report. In study 2, 60 infants provided a saliva sample and underwent EEG while viewing faces and objects and listening to human speech and water sounds at 4 months of age. Infant behavior was assessed via parental report and eye tracking.

RESULTS

We show in two independent infant samples that increased brain signal entropy during social perception is in part explained by an epigenetic modification to the oxytocin receptor gene (OXTR) and accounts for significant individual differences in social behavior in the first year of life. These results are measure-, context-, and modality-specific: entropy, not standard deviation, links OXTR methylation and infant behavior; entropy evoked during social perception specifically explains social behavior only; and only entropy evoked during social auditory perception predicts infant vocalization behavior.

CONCLUSIONS

Demonstrating these associations in infancy is critical for elucidating the neurobiological mechanisms accounting for individual differences in cognition and behavior relevant to neurodevelopmental disorders. Our results suggest that an epigenetic modification to the oxytocin receptor gene and brain signal entropy are useful indicators of social development and may hold potential diagnostic, therapeutic, and prognostic value.

Central coherence in adults with a high-functioning autism spectrum disorder. In a search for a non-self-reporting screening tool

BACKGROUND

Autism spectrum disorder in adults, especially high-functioning ones, is often difficult to differentiate from other mental disorders. Therefore, many adults with ASD are misdiagnosed, and their social difficulties are not adequately addressed. Moreover, frequent comorbid issues make diagnosis a challenging prospect. Most of the available screening and diagnostic tools rely on self-reporting, which can be a biased method. Weak Central Coherence is one of the main cognitive theories of ASD. According to research, individuals with ASD are slower in comparison to typically developed control on the uptake of context. The study goal was to see if the central coherence tasks could be used as a reliable screening marker that differentiates between high-functioning ASD and typically developed controls.

METHOD

Thirty males with ASD (as in DSM-5) and 30 demographically matched controls were investigated with Central Coherence Inferences Tests. Tests' scores and reaction times needed to complete the tasks in both groups were compared.

RESULTS

High-functioning participants with ASD achieved a similar score in central coherence tests as the typically developed control group, but they needed significantly more time to solve them. The ROC analysis for both central coherence tests revealed AUC values of 0.73 in differentiating ASD from typically developed controls.

CONCLUSIONS

The results are discussed in reference to the clinical application of central coherence as a possible screening marker. Further research directions are proposed in terms of differential diagnosis of adults with ASD.