Detection Of Autism Spectrum Disorder

Abstract

Autism spectrum disorder (ASD) presents a
neurological and developmental disorder that has an
impact on the social and cognitive skills of children
causing repetitive behaviors, restricted interests,
communication problems and difficulty in social
interaction. Early diagnosis of ASD can prevent from
its severity and prolonged effects. This project aims
to assist in the early detection of Autism Spectrum
Disorder (ASD) by integrating machine learning
models for behavioral screening and image
classification into a unified web-based application
using Streamlit. The application employs two
distinct pre-trained VotingClassifier models: one for
processing user responses to ASD screening
questionnaires and another for analyzing uploaded
images of autistic and non-autistic subjects.
Screening data is preprocessed and classified using
trained models, while images are resized,
preprocessed, and classified based on extracted
features.
Users interact with the app by answering survey
questions and optionally uploading images, with the
results of both screening and image-based
predictions displayed upon submission. The
application emphasizes accessibility and userfriendliness
while providing a disclaimer that it is not
a diagnostic tool but a supplementary aid for ASD
awareness and early intervention. This dual-model
approach enhances reliability and accuracy in
identifying potential ASD traits.