Abstract
Autism spectrum disorder (ASD) is a heterogenous neurodevelopmental disorder which is characterized by impaired communication, and limited social interactions. The shortcomings of current clinical approaches which are based exclusively on behavioral observation of symptomology, and poor understanding of the neurological mechanisms underlying ASD necessitates the identification of new biomarkers that can aid in study of brain development, and functioning, and can lead to accurate and early detection of ASD. In this paper, we developed a deep-learning model called ASD-SAENet for classifying patients with ASD from typical control subjects using fMRI data. We designed and implemented a sparse autoencoder (SAE) which results in optimized extraction of features that can be used for classification. These features are then fed into a deep neural network (DNN) which results in superior classification of fMRI brain scans more prone to ASD. Our proposed model is trained to optimize the classifier while improving extracted features based on both reconstructed data error and the classifier error. We evaluated our proposed deep-learning model using publicly available Autism Brain Imaging Data Exchange (ABIDE) dataset collected from 17 different research centers, and include more than 1,035 subjects. Our extensive experimentation demonstrate that ASD-SAENet exhibits comparable accuracy (70.8%), and superior specificity (79.1%) for the whole dataset as compared to other methods. Further, our experiments demonstrate superior results as compared to other state-of-the-art methods on 12 out of the 17 imaging centers exhibiting superior generalizability across different data acquisition sites and protocols. The implemented code is available on GitHub portal of our lab at: https://github.com/pcdslab/ASD-SAENet.
Highlights
More than 1.5 Million children (Baio et al, 2018) in the US are affected by heterogenous Autism Spectrum Disorder (ASD) which has wide range of symptoms or characteristics such as limited communication, limited social interaction, and may exhibit repeated or limited interests or activities (American Psychiatric Association, 2013)
We focus on designing a deep learning algorithm that can extract, and distinguish between the functional features associated with Autism spectrum disorder (ASD) Functional magnetic resonance imaging (fMRI) brain scans as compared to healthy typical controls
Due to the limitation of the sample data, our model was evaluated using k-fold cross validation technique in which the dataset is randomly split into k equal sized samples, and one of these is used for getting the classification performance
Summary
More than 1.5 Million children (Baio et al, 2018) in the US are affected by heterogenous Autism Spectrum Disorder (ASD) which has wide range of symptoms or characteristics such as limited communication (including verbal and non-verbal), limited social interaction, and may exhibit repeated or limited interests or activities (American Psychiatric Association, 2013). Individuals with ASD have numerous challenges in daily life, and often develop comorbidities such as depression, anxiety disorder, or ADHD which may further complicate the diagnostic processes especially for ASD-SAENet: ASD Classification Using Sparse Deep-Learning Model young children (Mizuno et al, 2019). Language, and social interventions for children (under 24 months old) with ASD has shown to be especially effective (Bradshaw et al, 2015), and a delayed diagnosis can have more disastrous effects in the life of the child. ASD is associated with altered brain development in the early childhood but there are no reliable biomarkers that can be used for diagnosis (Lord et al, 2018). The shortcomings of current clinical approaches (National Collaborating Centre for Mental Health, 2009), and the poor understanding of the neurological mechanisms underlying ASD necessitates the identification of new biomarkers and computational techniques that can aid clinicians, and neuroscientists alike to understand the distinct way ASD brain works as compared to a typical brain
Sign-in/Register to view highlights & summary Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
