Deep learning techniques for automated detection of autism spectrum disorder based on thermal imaging

2021 ◽  
pp. 095441192110247
Author(s):  
Kavya Ganesh ◽  
Snekhalatha Umapathy ◽  
Palani Thanaraj Krishnan
Keyword(s):  
Neural Network ◽  
Autism Spectrum Disorder ◽  
Skin Temperature ◽  
Thermal Imaging ◽  
Imaging Modality ◽  
Autism Spectrum ◽  
Spectrum Disorder ◽  
Facial Skin ◽  
Autistic Children ◽  
Thermal Images

Children with autism spectrum disorder have impairments in emotional processing which leads to the inability in recognizing facial expressions. Since emotion is a vital criterion for having fine socialisation, it is incredibly important for the autistic children to recognise emotions. In our study, we have chosen the facial skin temperature as a biomarker to measure emotions. To assess the facial skin temperature, the thermal imaging modality has been used in this study, since it has been recognised as a promising technique to evaluate emotional responses. The aim of this study was the following: (1) to compare the facial skin temperature of autistic and non-autistic children by using thermal imaging across various emotions; (2) to classify the thermal images obtained from the study using the customised convolutional neural network compared with the ResNet 50 network. Fifty autistic and fifty non-autistic participants were included for the study. Thermal imaging was used to obtain the temperature of specific facial regions such as the eyes, cheek, forehead and nose while we evoked emotions (Happiness, anger and sadness) in children using an audio-visual stimulus. Among the emotions considered, the emotion anger had the highest temperature difference between the autistic and non-autistic participants in the region’s eyes (1.9%), cheek (2.38%) and nose (12.6%). The accuracy obtained by classifying the thermal images of the autistic and non-autistic children using Customised Neural Network and ResNet 50 Network was 96% and 90% respectively. This computer aided diagnostic tool can be a predictable and a steadfast method in the diagnosis of the autistic individuals.

Development of Neural Structure and Function in Autism Spectrum Disorder: Potential Implications for Learning Language

2020 ◽  
Vol 29 (4) ◽  
pp. 1783-1797
Author(s):  
Kelly L. Coburn ◽  
Diane L. Williams
Keyword(s):  
Autism Spectrum Disorder ◽  
Language Development ◽  
Diffusion Tensor ◽  
Autism Spectrum ◽  
Language Intervention ◽  
Spectrum Disorder ◽  
Autistic Children ◽  
Neural Structure ◽  
Complex Information ◽  
And Function

Purpose Neurodevelopmental processes that begin during gestation and continue throughout childhood typically support language development. Understanding these processes can help us to understand the disruptions to language that occur in neurodevelopmental conditions, such as autism spectrum disorder (ASD). Method For this tutorial, we conducted a focused literature review on typical postnatal brain development and structural and functional magnetic resonance imaging, diffusion tensor imaging, magnetoencephalography, and electroencephalography studies of the neurodevelopmental differences that occur in ASD. We then integrated this knowledge with the literature on evidence-based speech-language intervention practices for autistic children. Results In ASD, structural differences include altered patterns of cortical growth and myelination. Functional differences occur at all brain levels, from lateralization of cortical functions to the rhythmic activations of single neurons. Neuronal oscillations, in particular, could help explain disrupted language development by elucidating the timing differences that contribute to altered functional connectivity, complex information processing, and speech parsing. Findings related to implicit statistical learning, explicit task learning, multisensory integration, and reinforcement in ASD are also discussed. Conclusions Consideration of the neural differences in autistic children provides additional scientific support for current recommended language intervention practices. Recommendations consistent with these neurological findings include the use of short, simple utterances; repetition of syntactic structures using varied vocabulary; pause time; visual supports; and individualized sensory modifications.

scholarly journals Multisite Autism Spectrum Disorder Classification Using Convolutional Neural Network Classifier and Individual Morphological Brain Networks

2021 ◽  
Vol 14 ◽  
Author(s):  
Jingjing Gao ◽  
Mingren Chen ◽  
Yuanyuan Li ◽  
Yachun Gao ◽  
Yanling Li ◽  
...  
Keyword(s):  
Neural Network ◽  
Autism Spectrum Disorder ◽  
Convolutional Neural Network ◽  
Structural Mri ◽  
Autism Spectrum ◽  
Brain Network ◽  
Spectrum Disorder ◽  
Structural Covariance ◽  
Almost All ◽  
New Framework

Autism spectrum disorder (ASD) is a range of neurodevelopmental disorders with behavioral and cognitive impairment and brings huge burdens to the patients’ families and the society. To accurately identify patients with ASD from typical controls is important for early detection and early intervention. However, almost all the current existing classification methods for ASD based on structural MRI (sMRI) mainly utilize the independent local morphological features and do not consider the covariance patterns of these features between regions. In this study, by combining the convolutional neural network (CNN) and individual structural covariance network, we proposed a new framework to classify ASD patients with sMRI data from the ABIDE consortium. Moreover, gradient-weighted class activation mapping (Grad-CAM) was applied to characterize the weight of features contributing to the classification. The experimental results showed that our proposed method outperforms the currently used methods for classifying ASD patients with the ABIDE data and achieves a high classification accuracy of 71.8% across different sites. Furthermore, the discriminative features were found to be mainly located in the prefrontal cortex and cerebellum, which may be the early biomarkers for the diagnosis of ASD. Our study demonstrated that CNN is an effective tool to build the framework for the diagnosis of ASD with individual structural covariance brain network.

scholarly journals Alterations of the endocannabinoid system and its therapeutic potential in autism spectrum disorder

Open Biology ◽  
2021 ◽  
Vol 11 (2) ◽  
Author(s):  
Mingyang Zou ◽  
Yu Liu ◽  
Shu Xie ◽  
Luxi Wang ◽  
Dexin Li ◽  
...  
Keyword(s):  
Autism Spectrum Disorder ◽  
Animal Model ◽  
Social Preference ◽  
Therapeutic Potential ◽  
Hydrolytic Enzyme ◽  
Endocannabinoid System ◽  
Autism Spectrum ◽  
Spectrum Disorder ◽  
Autistic Children ◽  
Novel Approach

Autism spectrum disorder (ASD) is a group of developmental disabilities, the aetiology of which remains elusive. The endocannabinoid (eCB) system modulates neurotransmission and neuronal plasticity. Evidence points to the involvement of this neuromodulatory system in the pathophysiology of ASD. We investigated whether there is a disruption to the eCB system in ASD and whether pharmacological modulation of the eCB system might offer therapeutic potential. We examined three major components of the eCB system—endogenous cannabinoids, their receptors and associated enzymes—in ASD children as well as in the valproic acid (VPA) induced animal model in autism. Furthermore, we specifically increased 2-arachidonoylglycerol (2-AG) levels by administering JZL184, a selective inhibitor of monoacylglycerol lipase which is the hydrolytic enzyme for 2-AG, to examine ASD-like behaviours in VPA-induced rats. Results showed that autistic children and VPA-induced rats exhibited reduced eCB content, increased degradation of enzymes and upregulation of CBRs. We found that repetitive and stereotypical behaviours, hyperactivity, sociability, social preference and cognitive functioning improved after acute and chronic JZL184 treatment. The major efficacy of JZL184 was observed after administration of a dosage regimen of 3 mg kg −1 , which affected both the eCB system and ASD-like behaviours. In conclusion, a reduced eCB signalling was observed in autistic children and in the ASD animal model, and boosting 2-AG could ameliorate ASD-like phenotypes in animals. Collectively, the results suggested a novel approach to ASD treatment.

scholarly journals Consecrate Recurrent Neural Network Classifier for Autism Classification

2019 ◽  
Vol 9 (1) ◽  
pp. 2033-2041
Keyword(s):  
Neural Network ◽  
Autism Spectrum Disorder ◽  
Recurrent Neural Network ◽  
Short Term Memory ◽  
Sequence Data ◽  
Large Data ◽  
Autism Spectrum ◽  
Spectrum Disorder ◽  
Neural Network Classifier ◽  
Gene Sequences

Most recent discoveries in Autism Spectrum Disorder (ASD) detection and classification studies reveal that there is a substantial relationship between Autism disorders and gene sequences. This work is indented to classify the autism spectrum disorder groups and sub-groups based on the gene sequences. The gene sequences are large data and perplexed for handling with conventional data mining or classification procedures. The Consecrate Recurrent Neural Network Classifier for Autism Classification (CRNNC-AC) work is introduced in this work to classify autism disorders using gene sequence data. A dedicated Elman [1] type Recurrent Neural Network (RNN) is introduced along with a legacy Long Short-Term Memory (LSTM) [2] in this classifier. The LSTM model is contrived to achieve memory optimization to eliminate memory overflows without affecting the classification accuracy. The classification quality metrics [3] such as Accuracy, Sensitivity, Specificity and F1-Score are concerned for optimization. The processing time of the proposed method is also measured to evaluate the pertinency.

scholarly journals Psychological Stress Among Mothers of Children with Autism Spectrum Disorder and Down Syndrome in the City of Kufra

2021 ◽  
Vol 2 (1) ◽  
pp. 29-41
Author(s):  
Mansour Mohammed Ali Bopaeda
Keyword(s):  
Autism Spectrum Disorder ◽  
Down Syndrome ◽  
Psychological Stress ◽  
Children With Autism ◽  
Autism Spectrum ◽  
Spectrum Disorder ◽  
Autistic Children ◽  
Children With Down Syndrome ◽  
Level Of Significance ◽  
The City

Having a baby is a huge responsibility. It often happens that mothers, especially when they are having their first baby, feel stressed and incompetent in their role as mothers even as their children normally grows. The purpose of this article is to identify the general features of psychological stress among mothers of autism spectrum disorder and Down syndrome and to identify the differences and relationship in psychological stress between mothers of autism and down syndrome. In this study participated a group of mothers with children who have autism (n = 44) and mothers of children with Down syndrome (n = 44). The findings made proved that the general features of psychological stress are high Among mothers of autistic children, while it was low among mothers of Down syndrome, there were also statistically significant differences in the level of psychological stress between mothers of autism and Down syndrome, and a correlation was found between mothers of autism and Down syndrome at the level of significance (0.01).

Effect of physical therapy on 7- to 10-year-old children with autism spectrum disorder: A retrospective study in a university day hospital

2021 ◽  
Vol 85 (4) ◽  
pp. 385-404
Author(s):  
Jokthan Guivarch ◽  
Elisabeth Jouve ◽  
Elodie Avenel ◽  
François Poinso ◽  
Laura Conforti-Roussel
Keyword(s):  
Autism Spectrum Disorder ◽  
Retrospective Study ◽  
Motor Impairment ◽  
Autism Spectrum ◽  
The Body ◽  
Spectrum Disorder ◽  
Control Group ◽  
Autistic Children ◽  
Dance Movement ◽  
Cascading Effect

More than half of children who have autism spectrum disorder (ASD) suffer from motor impairment. In a retrospective study, the authors investigated the effect of a body-mediated workshop with dance movement therapy (DMT) on the motor skills and social skills of children with ASD by comparing 10 autistic children aged 7 to 10 years who benefited from DMT with 10 autistic children in a control group. Scores on the Movement Assessment Battery for Children and the Vineland Adaptive Behavior Scale were compared. The body-mediated workshop had significant benefits for motricity, especially manual dexterity, and for relational skills. A body-mediated workshop may have a multimodal effect and requires transmodal training. Regarding the mechanisms that explain the benefits and the cascading effect, the roles of imitation and multimodal connections are important.

scholarly journals Interpretable Learning Approaches in Resting-State Functional Connectivity Analysis: The Case of Autism Spectrum Disorder

2020 ◽  
Vol 2020 ◽  
pp. 1-12
Author(s):  
Jinlong Hu ◽  
Lijie Cao ◽  
Tenghui Li ◽  
Bin Liao ◽  
Shoubin Dong ◽  
...  
Keyword(s):  
Neural Network ◽  
Autism Spectrum Disorder ◽  
Functional Connectivity ◽  
Resting State ◽  
Piecewise Linear ◽  
Autism Spectrum ◽  
Spectrum Disorder ◽  
Fmri Data ◽  
Support Vector ◽  
Resting State Functional Connectivity

Deep neural networks have recently been applied to the study of brain disorders such as autism spectrum disorder (ASD) with great success. However, the internal logics of these networks are difficult to interpret, especially with regard to how specific network architecture decisions are made. In this paper, we study an interpretable neural network model as a method to identify ASD participants from functional magnetic resonance imaging (fMRI) data and interpret results of the model in a precise and consistent manner. First, we propose an interpretable fully connected neural network (FCNN) to classify two groups, ASD versus healthy controls (HC), based on input data from resting-state functional connectivity (rsFC) between regions of interests (ROIs). The proposed FCNN model is a piecewise linear neural network (PLNN) which uses piecewise linear function LeakyReLU as its activation function. We experimentally compared the FCNN model against widely used classification models including support vector machine (SVM), random forest, and two new classes of deep neural network models in a large dataset containing 871 subjects from ABIDE I database. The results show the proposed FCNN model achieves the highest classification accuracy. Second, we further propose an interpreting method which could explain the trained model precisely with a precise linear formula for each input sample and decision features which contributed most to the classification of ASD versus HC participants in the model. We also discuss the implications of our proposed approach for fMRI data classification and interpretation.

Sign in / Sign up

Export Citation Format

Share Document