Autism Spectrum Disorder (ASD) is a common neurodevelopmental disorder with core symptoms that include language communication difficulties, social interaction challenges, and repetitive, restricted behaviors. It has long-term impacts on individuals and families. However, due to the lack of biological markers and unified diagnostic methods, clinical diagnosis of ASD still relies on questionnaire screening and physician judgment, which often leads to burdens on medical resources and delays in diagnosis.
To improve screening efficiency, a research team led by Shur-Fen Gau, Vice Superintendent of National Taiwan University Hospital, has developed AQ-13, a new generation of artificial intelligence-based autism screening and diagnostic aid tool, using machine learning techniques.
Efficient and Accurate: AQ-13's Clinical and Market Potential
AQ-13 analyzes vast amounts of questionnaire data from children with autism and normal children using AI models. It selects 13 high-discriminative questions, significantly simplifying the original 50-question AQ scale, thus reducing the time required for completion while maintaining high diagnostic accuracy. The core advantages of this technology include:
AQ-13 has broad application potential, targeting both professional medical personnel and the general public. It is designed to assist pediatricians, child psychiatrists, family physicians, psychotherapists, and other relevant professionals in performing initial ASD screenings. It is also suitable for use by school teachers and parents, who can self-administer the questionnaire to assess a child's risk for ASD. Compared to traditional questionnaire-based screening tools, which are time-consuming and involve complicated diagnostic processes, AQ-13 fills a gap in the market by offering a simple, quick, and efficient solution, lowering the diagnostic threshold and improving screening accessibility.
According to the National Development Council, the population of children and adolescents aged 4 to 18 in Taiwan is approximately 2.97 million. With over 1.4 billion Mandarin-speaking people worldwide, and an estimated ASD prevalence rate of 1%, at least 14 million people globally could benefit from this technology. Additionally, AQ-13 features an objective, consistent AI-based automated scoring mechanism, which eliminates human error, enhances diagnostic standardization, and can be promoted to the Mandarin-speaking market through digital platforms, becoming a key tool for clinical and community screenings.
Expert Interview: Precision Diagnosis Brings New Opportunities for Early Treatment
Regarding the development prospects of AQ-13, Vice Superintendent Gau mentioned that early diagnosis and treatment of autism are crucial for the future development of patients. AQ-13's high accuracy enables rapid screening, allowing parents and healthcare professionals to detect high-risk cases earlier and refer them for timely professional treatment. This will have a significant impact on the quality of life for children with ASD. In the future, the research team plans to further optimize AQ-13 and integrate digital technologies to expand its application scenarios. They have already developed an initial website and user interface and plan to collaborate with professional publishers and digital development teams to accelerate AQ-13's commercialization. They also plan to introduce value-added services such as case management systems and report generation to provide more comprehensive ASD screening and diagnostic solutions. With continuous improvements and market promotion, AQ-13 is expected to become the standard tool for early ASD screening, offering a more convenient and effective diagnostic method for the global Mandarin-speaking autism community.
Resource: AI精挑13題高鑑別度問題 快速篩檢自閉症利早期治療
