Breakthrough in Autism Diagnosis Developed in Tenerife
A University of La Laguna (ULL) student in Tenerife has created an innovative artificial intelligence tool that helps diagnose autism spectrum disorder (ASD) by analyzing patients’ eye movement patterns. Roberto Chávez, an Electronics Engineering graduate with a PhD in Systems and Automation Engineering, developed this groundbreaking system as part of his doctoral thesis at ULL.
Understanding Autism Spectrum Disorder
ASD encompasses a range of neurodevelopmental conditions characterized by challenges with social interaction, communication difficulties, and atypical behavioral patterns. These may include trouble adapting to changes or unusual responses to sensory stimuli. While signs often appear in early childhood, many cases aren’t diagnosed until later developmental stages.
Revolutionizing Traditional Diagnosis Methods
Current ASD detection primarily relies on behavioral observation and structured interviews like the Autism Diagnostic Observation Schedule (ADOS), which depend heavily on clinicians’ subjective judgment. Chávez’s thesis addresses this limitation by creating a diagnostic support tool that uses eye-tracking data analysis and AI algorithms to identify visual patterns associated with autism.
How the AI-Powered System Works
The technology builds on research showing that visual patterns in people with ASD show distinctive characteristics from about six months of age. Chávez’s system captures these through advanced eye-tracking. His thesis advisor, Rosa Aguilar (ULL’s Department of Computer and Systems Engineering professor and former university rector), notes the project involved processing “extremely complex data” with samples collected every three milliseconds, each containing over 100 variables.
Cutting-Edge Data Processing Techniques
The system employs data science to extract patterns from massive information volumes, then applies machine learning algorithms – specifically the XGBoost model which combines simpler models for superior performance. Chávez also developed blink pattern analytics, pupil size measurement, and gaze sequence analysis techniques.
User-Friendly Diagnostic Platform
The final innovation is a plug-and-play web interface that simplifies diagnosis. Users simply upload eye-tracker data files, and the platform automatically processes them through all diagnostic steps without requiring technical expertise. This accessibility could revolutionize early autism detection, particularly in resource-limited settings.
Collaborative Research Achievement
Chávez worked with Professor José Luis González Mora, head of ULL’s Neurochemistry and Neuroimaging Research Laboratory, to develop this pioneering diagnostic tool that merges neuroscience with cutting-edge AI technology.
