Simulating Autism Spectrum Disorder Diagnosis Using Tsallis Entropy

This study explores the application of Tsallis entropy, a non-extensive entropy measure, to analyze diagnostic patterns of Autism Spectrum Disorder (ASD) through simulation. Synthetic data, reflecting real-world ASD behavioral metrics such as social interaction deficits and repetitive behaviors, are generated to compute Tsallis entropy, quantifying diagnostic complexity and uncertainty. Using Python, the simulation analyzes entropy variations across mild, moderate, and severe ASD severity levels. Results indicate that Tsallis entropy can distinguish diagnostic profiles, offering insights into ASD heterogeneity. Limitations and future research directions are discussed.