A novel hybrid deep learning model using MEResNext for autism spectrum disorder detection.

Researchers developed a computer program that uses artificial intelligence to help detect autism. The program showed promising results in testing, correctly identifying autism 95.3% of the time. However, the study doesn't provide important details like how many people were tested or how the results compare to current diagnostic methods used by doctors.

This study presents a novel artificial intelligence approach for detecting autism spectrum disorder using a hybrid deep learning model called MEResNeXt. The model combines three stages: data preprocessing using Yeo-Johnson transformation, feature selection using a Double exponential Smoothing-Elk Herd Optimizer algorithm, and ASD detection using the MEResNeXt network. The researchers report high performance metrics with 95.3% accuracy, 96.5% sensitivity, and 94.8% specificity. However, critical details about the dataset, sample size, validation methods, and clinical context are not provided in the abstract, limiting assessment of the model's real-world applicability.

While the reported performance metrics appear promising, the lack of validation details, sample characteristics, and comparison with current diagnostic practices limits immediate clinical applicability. Further research with proper validation is needed before clinical implementation.

No sample size reported, validation methodology unclear, no comparison with existing diagnostic tools, dataset characteristics not described, and real-world clinical applicability uncertain due to lack of implementation details.

A neurological disease, named autism spectrum disorder (ASD), is portrayed through divergence with social interaction, communication, and repetitive activities. As heredity is an important source, initial identification and treatment can diminish the need for wide examinations and expensive medical procedures for those with ASD. Prolonged effects and severity can be avoided at a timely determination of ASD. This study develops a hybrid deep-learning method for ASD detection.

Pre-processing, feature selection and ASD detection are three steps involved. In the pre-processing phase, input data is passed in which noise and artefacts present in the data are removed using Yeo-Jhonson transformation. Next, by integrating Double exponential Smoothing (DES) in Elk Herd Optimizer (EHO), Double exponential Smoothing-Elk Herd Optimizer (DeSEHO), is proposed to select features. At last, the features that are selected are subjected to the ASD detection phase.

In this stage, the Moments Embedding ResNeXt (MEResNeXt), which is the combination of Moments Embedding Network (MoNet) and ResNeXt, is used for ASD detection. The results of the research show that MEResNeXt performed better in terms of traditional models which show accuracy of 95.3 %, sensitivity of 96.5 %, and specificity of 94.8 %.