Pregnancy risk factors in autism: a pilot study with artificial neural networks

Graph fusion prediction of autism based on attentional mechanisms

Autism spectrum disorder (ASD) is a pervasive developmental disorder, and the earlier detection and timely intervention for treatment positively affect the prognosis of patients. Deep learning algorithms based on graph structure have achieved good results in autism prediction in recent years. However, there are problems with standardized operations in extracting features and combining neighborhood node features with the structure of the graph dependent, which limits the generalization ability of the trained model to other graph structures. In this paper, we propose a graph fusion autism prediction model based on attentional mechanisms(AGF) to address the above problems. The AGF model represents the overall population (patients or healthy controls) as a sparse graph, where nodes are subjects, and non-imaging features are integrated as edge weights. Different weights can be defined for different nodes in the neighborhood through the attention mechanism without relying on prior knowledge of the graph structure. The model is also able to dynamically fuse multiple sparse graphs obtained from different non-imaging features by way of training weight assignment. Its performance is also compared with several other models (e.g., S-AGF, GCN, etc.), and the results show that it has superior prediction accuracy compared to the baseline model. The results show that this improvement of graph fusion works better on the ABIDE databases, and the classification accuracy can reach 73.9%. The datasets and source code are freely available at https://github.com/chengyu-github1012/Graph-Fusion.git. Strengths and limitations of this study: graph fusion; disease prediction; noise.

Not breastfeeding and risk of autism spectrum disorders among children: a meta-analysis

BACKGROUND

To our knowledge, this is the first meta-analysis of the association of not breastfeeding and the risk of autism spectrum disorder (ASD) based on observational studies.

PURPOSE

This meta-analysis aimed to evaluate of the association of not breastfeeding and the risk of ASD.

METHODS

Three databases (PubMed, Web of Science, and Scopus) were systematically searched until December 2021. Heterogeneity was determined using the chi-square test and its quantity was measured using the I2 statistic. The Begg line regression test was used to assess publication bias. A random-effects model was used to analyze the data. Seven studies were included in this meta-analysis.

RESULTS

The total study population included 3,270 individuals. According to the random-effects model, the estimated odds ratio of the risk of ASD associated with not breastfeeding was 1.81 (95% confidence interval, 1.35-2.27; I2=0%).

CONCLUSION

The results of the included studies were homogeneous. Our findings showed that not breastfeeding is a risk factor for ASD. These results suggest the importance of breastfeeding in decreasing the risk of ASD in children.

A Meta-Analysis on the Impact of Prenatal and Early Childhood Antimicrobial Use on Autism Spectrum Disorders

OBJECTIVE

To investigate the impact of prenatal and early childhood antimicrobial use on autism spectrum disorders (ASD).

DATA SOURCES

We searched PubMed and Embase databases for relevant studies from inception to August 2022.

STUDY SELECTION AND DATA EXTRACTION

Peer-reviewed, observational studies were all acceptable. Raw data were extracted into a predefined worksheet and quality analysis was performed using the Newcastle-Ottawa Scale.

DATA SYNTHESIS

Nineteen studies were identified in the meta-analysis. Prenatal antimicrobial exposure was not associated with ASD (P = 0.06 > 0.05), whereas early childhood antimicrobial exposure was associated with an increased odds ratio of ASD (OR = 1.17, 95% CI = [1.08-1.27], P value < 0.001). The sibling-matched analysis, with a very limited sample size, suggested that neither prenatal (P = 0.47 > 0.05) nor early childhood (P = 0.13 > 0.05) antimicrobial exposure was associated with ASD. Medical professionals may need to take the possible association into consideration when prescribing an antimicrobial in children.

CONCLUSIONS

Early childhood antimicrobial exposure could increase the incidence of ASD. In future studies, it would be necessary to control for confounding factors, such as genetic factors, parenteral age at birth, or low birthweight, to further validate the association.

Early Life Antibiotic Exposure and the Subsequent Risk of Autism Spectrum Disorder and Attention Deficit Hyperactivity Disorder: A Systematic Review and Meta-Analysis

This study was conducted to assess this association between early life antibiotic exposure and the risk of autism spectrum disorder (ASD) and attention deficit hyperactivity disorder (ADHD) in later life. The results showed that early life antibiotic exposure was associated with an increased risk of ASD (OR = 1.13, 95% confidence interval (CI): 1.07-1.21) or ADHD (OR = 1.18, 95% CI: 1.1-1.27). However, this association for ASD (OR = 1.04, 95% CI: 0.97-1.11) or ADHD (OR = 0.98, 95% CI: 0.94-1.02) disappeared when data from sibling-matched studies were pooled. The statistically significant association between early life antibiotic exposure and ASD or ADHD in later life can be partially explained by unmeasured genetic and familial confounding factors.

Prenatal maternal infection and risk for autism in offspring: A meta-analysis

While prenatal maternal infection has received attention as a preventable and treatable risk factor for autism, findings have been inconsistent. This paper presents the results of a meta-analysis to determine whether the weight of the evidence supports such an association. Studies with a categorical diagnosis of autism as the outcome and an assessment of its association with prenatal maternal infection or fever (or the data necessary to compute this association) were included. A total of 36 studies met these criteria. Two independent reviewers extracted data on study design, methods of assessment, type of infectious agent, site of infection, trimester of exposure, definition of autism, and effect size. Analyses demonstrated a statistically significant association of maternal infection/fever with autism in offspring (OR = 1.32; 95% CI = 1.20-1.46). Adjustment for evident publication bias slightly weakened this association. There was little variation in effect sizes across agent or site of infection. Small differences across trimester of exposure were not statistically significant. There was some evidence that recall bias associated with status on the outcome variable leads to differential misclassification of exposure status. Nonetheless, the overall association is only modestly reduced when studies potentially contaminated by such bias are removed. Although causality has not been firmly established, these findings suggest maternal infection during pregnancy confers an increase in risk for autism in offspring. Given the prevalence of this risk factor, it is possible that the incidence of autism would be reduced by 12%-17% if maternal infections could be prevented or safely treated in a timely manner. LAY SUMMARY: This study is a meta-analysis of the association of maternal infection during pregnancy and subsequent autism in offspring. In combining the results from 36 studies of this association we find that a significant relationship is present. The association does not vary much across the types of infections or when they occur during pregnancy. We conclude that the incidence of autism could be substantially reduced if maternal infections could be prevented or safely treated in a timely manner.

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.

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.

Self-Reported Autistic Traits Using the AQ: A Comparison between Individuals with ASD, Psychosis, and Non-Clinical Controls

The term "autism" was originally coined by Eugen Bleuler to describe one of the core symptoms of schizophrenia. Even if autism spectrum disorder (ASD) and schizophrenia spectrum disorders (SSD) are now considered two distinct conditions, they share some clinical features. The present study aimed to investigate self-reported autistic traits in individuals with ASD, SSD, and non-clinical controls (NCC), using the Autism-Spectrum Quotient (AQ), a 50-item questionnaire. The study was conducted in the Psychiatry Unit of Policlinico "G. Rodolico", Catania, Italy. The AQ was administered to 35 adults with ASD, 64 with SSD, and 198 NCC. Overall, our data showed that the ASD sample scored significantly higher than NCC. However, no significant differences were detected between individuals with ASD and SSD. Notably, the three groups scored similarly in the subscale "attention to detail". AQ showed good accuracy in differentiating ASD from NCC (AUC = 0.84), while discriminant ability was poor in the clinical sample (AUC = 0.63). Finally, AQ did not correlate with clinician-rated ADOS-2 scores in the ASD sample. Our study confirms that symptoms are partially overlapping in adults with ASD and psychosis. Moreover, they raise concerns regarding the usefulness of AQ as a screening tool in clinical populations.

Prenatal maternal stress and risk of neurodevelopmental disorders in the offspring: a systematic review and meta-analysis

PURPOSE

Exposure to prenatal stress has been reported to affect the risk of adverse neurodevelopmental outcomes in the offspring; however, there is currently no clear consensus. The aim of this systematic review and meta-analysis was to examine the existing literature on the association between prenatal stress and autism spectrum disorder (ASD) and attention-deficit hyperactivity disorder (ADHD) in the offspring.

METHODS

Based on a registered protocol, we searched several electronic databases for articles in accordance with a detailed search strategy. We performed this study following the Preferred Reporting Items for Systematic reviews and Meta-Analyses (PRISMA).

RESULTS

Prenatal stress was significantly associated with an increased risk of both ASD (pooled OR 1.64 [95% CI 1.15-2.34]; I² = 90%; 15 articles) and ADHD (pooled OR 1.72 [95% CI 1.27-2.34]; I² = 85%; 12 articles).

CONCLUSIONS

This study suggests that prenatal stress may be associated with ASD and ADHD; however, several limitations in the reviewed literature should be noted including significant heterogeneity and there is a need for carefully controlled future studies in this area.

The Association between Autism Spectrum Disorder and Pre- and Postnatal Antibiotic Exposure in Childhood-A Systematic Review with Meta-Analysis

Autism spectrum disorder (ASD) is a developmental disorder that begins in early childhood and has been associated with several environmental and genetic factors. We aimed to conduct two-side meta-analyses to determine the association between ASD and pre- and postnatal antibiotic exposure in childhood. We searched PubMed, Embase, Web of Science, and Cochrane Library for articles published up to February 2019. We evaluated observational studies that assessed the association between ASD and antibiotic exposure. Of 1459 articles, nine studies were used in the meta-analysis. We found that early antibiotic exposure, including pre- and postnatal, significantly increased the ASD risk in children. Furthermore, early antibiotic exposure, including pre- and postnatal, was significantly increased in children with ASD. Specifically, prenatal antibiotic exposure was significantly increased in children with ASD; however, postnatal antibiotic exposure was not. Our results indicate an association between ASD and early antibiotic exposure; specifically, that prenatal antibiotic exposure is an important risk factor of ASD in children.

Application of Single-Nucleotide Polymorphisms in the Diagnosis of Autism Spectrum Disorders: A Preliminary Study with Artificial Neural Networks

Autism spectrum disorder (ASD) includes different neurodevelopmental disorders characterized by deficits in social communication, and restricted, repetitive patterns of behavior, interests or activities. Based on the importance of early diagnosis for effective therapeutic intervention, several strategies have been employed for detection of the disorder. The artificial neural network (ANN) as a type of machine learning method is a common strategy. In the current study, we extracted genomic data for 487 ASD patients and 455 healthy individuals. All individuals were genotyped in certain single-nucleotide polymorphisms within retinoic acid-related orphan receptor alpha (RORA), gamma-aminobutyric acid type A receptor beta3 subunit (GABRB3), synaptosomal-associated protein 25 (SNAP25) and metabotropic glutamate receptor 7 (GRM7) genes. Subsequently, we used the "Keras" package to create and train the ANN model. For cross-validation, samples were divided into ten folds. In the training process, initially, the first fold was preserved for validation and the other folds were used to train the model. The validation fold was then used to evaluate model performance. The k-fold cross-validation method was used to ensure model generalizability and to prevent overfitting. Local interpretable model-agnostic explanations (LIME) were applied to explain model predictions at the data sample level. The output of loss function was evaluated in the training process for each fold in the k-fold cross-validation model. Finally, the number of losses was reduced to less than 0.6 after 200 epochs (except in two cases). The accuracy, sensitivity and specificity of our model were 73.67%, 82.75% and 63.95%, respectively. The area under the curve (AUC) was 80.59. Consequently, in the current study, we propose an ANN-based method for differentiating ASD status from healthy status with adequate power.

Evaluation of machine learning algorithms for improved risk assessment for Down's syndrome

Prenatal screening generates a great amount of data that is used for predicting risk of various disorders. Prenatal risk assessment is based on multiple clinical variables and overall performance is defined by how well the risk algorithm is optimized for the population in question. This article evaluates machine learning algorithms to improve performance of first trimester screening of Down syndrome. Machine learning algorithms pose an adaptive alternative to develop better risk assessment models using the existing clinical variables. Two real-world data sets were used to experiment with multiple classification algorithms. Implemented models were tested with a third, real-world, data set and performance was compared to a predicate method, a commercial risk assessment software. Best performing deep neural network model gave an area under the curve of 0.96 and detection rate of 78% with 1% false positive rate with the test data. Support vector machine model gave area under the curve of 0.95 and detection rate of 61% with 1% false positive rate with the same test data. When compared with the predicate method, the best support vector machine model was slightly inferior, but an optimized deep neural network model was able to give higher detection rates with same false positive rate or similar detection rate but with markedly lower false positive rate. This finding could further improve the first trimester screening for Down syndrome, by using existing clinical variables and a large training data derived from a specific population.

Artificial neural network detects human uncertainty

Artificial neural networks (ANNs) are known to be a powerful tool for data analysis. They are used in social science, robotics, and neurophysiology for solving tasks of classification, forecasting, pattern recognition, etc. In neuroscience, ANNs allow the recognition of specific forms of brain activity from multichannel EEG or MEG data. This makes the ANN an efficient computational core for brain-machine systems. However, despite significant achievements of artificial intelligence in recognition and classification of well-reproducible patterns of neural activity, the use of ANNs for recognition and classification of patterns in neural networks still requires additional attention, especially in ambiguous situations. According to this, in this research, we demonstrate the efficiency of application of the ANN for classification of human MEG trials corresponding to the perception of bistable visual stimuli with different degrees of ambiguity. We show that along with classification of brain states associated with multistable image interpretations, in the case of significant ambiguity, the ANN can detect an uncertain state when the observer doubts about the image interpretation. With the obtained results, we describe the possible application of ANNs for detection of bistable brain activity associated with difficulties in the decision-making process.

Association between hypertensive disorders of pregnancy and risk of autism in offspring: a systematic review and meta-analysis of observational studies

Background

Autism spectrum disorder (ASD) is a common severe pervasive neurodevelopmental disorder of undetermined etiology. Environmental exposures, especially pregnancy complications, have been increasingly recognized as a potential risk factor for ASD. Our aim was to (1) systematically evaluate the association between hypertensive disorders of pregnancy (HDP) and the risk of ASD in offspring, (2) specifically draw a subgroup analysis of disease severity in patients with HDP to achieve more sufficient evidence on this issue.

Results

A total of 21 studies were identified with more than 6.5 million participants, including 31,027 ASD probands. A comparative meta-analysis established that offspring born premature to HDP were significantly associated with ASD than matched controls (OR = 1.42, 95% CI: 1.34–1.50). Subgroup analysis of clinical classification include: (1) gestational hypertension, (2) pre-eclampsia, (3) chronic hypertension complicating pregnancy (CHP). The offspring of mothers with pre-eclampsia and CHP have slightly higher risk (OR = 1.43; OR = 1.48, respectively) of ASD than those of mothers with gestational hypertension (OR = 1.37). In consistence with most previous researches, higher ASD prevalence was observed in male than female (OR = 1.38), indicating a potential role for gender in the pathophysiology of ASD.

Materials and Methods

We conducted a systematic literature search on PubMed, EMBASE, Web of Science, PsycINFO database and China National Knowledge Infrastructure up to Jun. 2017. Statistical analysis was performed using Stata 10.0.

Conclusions

This meta-analysis implies a possible link between HDP and the risk of ASD in offspring. However, further investigation should be conducted to confirm this conclusion, and intensive prenatal surveillance and early prediction for ASD is needed.

Interplay between maternal Slc6a4 mutation and prenatal stress: a possible mechanism for autistic behavior development

The low activity allele of the maternal polymorphism, 5HTTLPR, in the serotonin transporter, SLC6A4, coupled with prenatal stress is reported to increase the risk for children to develop autism spectrum disorder (ASD). Similarly, maternal Slc6a4 knock-out and prenatal stress in rodents results in offspring demonstrating ASD-like characteristics. The present study uses an integrative genomics approach to explore mechanistic changes in early brain development in mouse embryos exposed to this maternal gene-environment phenomenon. Restraint stress was applied to pregnant Slc6a4 ^+/+ and Slc6a4 ^+/- mice and post-stress embryonic brains were assessed for whole genome level profiling of methylome, transcriptome and miRNA using Next Generation Sequencing. Embryos of stressed Slc6a4 ^+/+ dams exhibited significantly altered methylation profiles and differential expression of 157 miRNAs and 1009 genes affecting neuron development and cellular adhesion pathways, which may function as a coping mechanism to prenatal stress. In striking contrast, the response of embryos of stressed Slc6a4 ^+/- dams was found to be attenuated, shown by significantly reduced numbers of differentially expressed genes (458) and miRNA (0) and genome hypermethylation. This attenuated response may pose increased risks on typical brain development resulting in development of ASD-like characteristics in offspring of mothers with deficits in serotonin related pathways during stressful pregnancies.

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.

Does prenatal stress alter the developing connectome?

Human neurodevelopment requires the organization of neural elements into complex structural and functional networks called the connectome. Emerging data suggest that prenatal exposure to maternal stress plays a role in the wiring, or miswiring, of the developing connectome. Stress-related symptoms are common in women during pregnancy and are risk factors for neurobehavioral disorders ranging from autism spectrum disorder, attention deficit hyperactivity disorder, and addiction, to major depression and schizophrenia. This review focuses on structural and functional connectivity imaging to assess the impact of changes in women's stress-based physiology on the dynamic development of the human connectome in the fetal brain.