article · IEEE Access
Autism Spectrum Disorder is a neurodevelopmental condition affecting social communication, making early diagnosis crucial for mitigating its effects. This research presents a diagnostic framework using Magnetic Resonance Imaging brain scans combined with a hybrid deep learning model. The process begins by removing non-brain tissues, followed by image segmentation using a combination of Fuzzy C Means and Gaussian Mixture Models to isolate cortical and sub-cortical regions. A deep convolutional neural network, specifically VGG-16, is employed to extract intricate region-of-interest functional connectivity features. These features are then classified using a Residual Network whose hyperparameters are tuned with a Dwarf Mongoose optimisation algorithm. The combined approach reduces computational complexity and achieves an autism detection accuracy of 99.83 percent on brain scan data.
Early detection of Autism Spectrum Disorder can assist in reducing the long-term impact of the condition on individuals. Demonstrating that automated artificial intelligence systems can analyse brain scans with high precision provides a potential foundation for objective, image-based diagnostic aids, helping clinical specialists identify neurodevelopmental conditions sooner and with greater consistency.
This technology could potentially support software developers building diagnostic decision-support tools for radiologists and clinical specialists assessing autism. Given that the abstract reports only an experimental classification model evaluated on brain image data without clinical trials or system integration, the research represents an early-stage algorithmic proof of concept that requires further testing and regulatory validation before any clinical or commercial deployment.
AI-generated from the published abstract. Always read the original work before citing.
The neurodevelopmental Autism Spectrum Disorder (ASD) causes problems in social communication. Earlier diagnosis of ASD from brain image is necessary for reducing the effect of disorder. In this paper, deep Convolutional Neural Network (CNN) with Dwarf Mongoose optimized Residual Network (DM-ResNet) is proposed for the classification of autism disorder from Magnetic Resonance Imaging (MRI) brain images. Initially, the input brain images are preprocessed to remove the non-brain tissues. The preprocessed images are segmented with hybrid Fuzzy C Means (FCM) and Gaussian Mixture Model (GMM) which partition the image into sub groups to make it easier for classification by reducing the complexity. FCM-GMM segments the volume into predefined cortical and sub cortical regions. After segmentation, the features are extracted with Visual Geometry Group (VGG)-16 networks which comprised of several tiny kernels with filters for enhancing the depth of network and permit to extract complicated and discriminative features. Region of Interest (ROI) based functional connectivity feature is extracted with VGG-16 and these features are classified with DM optimized ResNet. The hyper parameters are optimized with DM optimization algorithm which improves the accuracy of classifier. By using the proposed approach, the accuracy of autism detection is improved to 99.83%.
This page summarises published work. The authoritative version sits with the publisher.
DOI: 10.1109/access.2023.3325701
Is something wrong with this record? Report it or request removal.
Discussion
Have you built on this work, tried to replicate it, or seen it applied in practice? Share what you know. Verified researchers and MARATTO™ domain experts can open a discussion, and any member can reply. Contributions are reviewed before they appear.
No discussion yet. Open the first thread.
New to MARATTO™? Create a free account.
