Exploiting Large Language Models for Diagnosing Autism Associated Language Disorders and Identifying Distinct...

Area of Science:

• Computational linguistics
• Developmental psychology
• Artificial intelligence in healthcare

Background:

• Diagnosing language disorders in autism spectrum disorder (ASD) is challenging due to subjective traditional methods.
• Existing assessment tools often lack the speed and precision required for timely diagnosis.

Purpose of the Study:

• To explore the use of Large Language Models (LLMs) for enhancing the sensitivity and precision of autism diagnosis.
• To identify and profile key linguistic features associated with ASD-related language disorders.

Main Methods:

• Utilized LLMs' natural language understanding capabilities to analyze linguistic patterns in individuals with ASD.
• Employed a zero-shot learning configuration to evaluate model performance against baseline methods.

Main Results:

• The LLM-based approach demonstrated over a 10% increase in sensitivity and positive predictive value compared to baseline models.
• Identified ten key linguistic features, including echolalia and pronoun reversal, crucial for ASD diagnosis.

Conclusions:

• LLMs offer a promising, supplementary tool for improving the accuracy and efficiency of diagnosing ASD-related language patterns.
• The identified linguistic features can inform the development of personalized treatment strategies for individuals with ASD.