Multi-Channel Inception-DenseNet for Autism Classification: A Robust Deep Learning Model

Autism is a neurodevelopmental condition that affects communication, social interactions, and behavioral patterns. Although the precise cause is still unknown, studies suggest that both genetic and environmental factors influence brain development. People with autism may encounter difficulties in understanding social cues, exhibit repetitive behaviors, and develop intense interests in specific topics. Early diagnosis is essential as it enables timely interventions that can significantly improve cognitive and behavioral development. However, diagnosing autism remains a challenging task due to variations in symptoms and reliance on subjective assessments by medical professionals. In this research, we explored deep learning techniques to detect autism using EEG medical imaging. Our proposed model, Multi-Channel Inception DenseNet, was designed to enhance feature extraction and classification, achieving an impressive accuracy of 98.64 percentage on the dataset. This approach leverages artificial intelligence to support medical experts in diagnosing autism more quickly and accurately, reducing inconsistencies in traditional diagnostic methods. By integrating deep learning into medical imaging analysis, this research highlights the capability of AI-powered solutions in improving early detection and intervention strategies. Upcoming research will concentrate on enhancing the model using larger datasets and real-world clinical applications to further enhance autism diagnosis and treatment.