An Efficient Autism Spectrum Disorder Classification using Modified Artificial Immune System

Authors

  • Abinaya S
  • Dr. W. Rose Varuna

DOI:

https://doi.org/10.70135/seejph.vi.3244

Abstract

Autism Spectrum Disorder (ASD) is a childhood disability that interferes with social interaction and communication, as well as patterns of behavior. This paper introduces a method for the classification of Autism Spectrum Disorder (ASD) employing a Modified Artificial Immune System (M-AIS). The proposed system improves the AIS by adding dynamic feature extraction and optimization that improves the classification of sensory, motor and genetic condition related data. The model fits to existing ASD diagnostic models that have shortcomings like static features classification and rigidity in data features. The modified AIS uses clonal selection, mutation, and affinity maturation to refine decision boundaries to increase diagnostic accuracy. The proposed system was evaluated on features from ASD data; the system was accurate and fast in its classification. The modified AIS offered improved real-time adaptability and highest 95.12% accuracy of the predictions thatis more robust than existing machine learning models. This method is a good solution for early diagnosis of ASDs because it offers clinicians a better, flexible instrument for testing existence of ASD features in people.

Downloads

Published

2025-01-03

How to Cite

S, A., & Varuna, D. W. R. (2025). An Efficient Autism Spectrum Disorder Classification using Modified Artificial Immune System. South Eastern European Journal of Public Health, 139–150. https://doi.org/10.70135/seejph.vi.3244

Issue

Section

Articles