Using machine-learning techniques, Professor Dennis Wall and colleagues were able to achieve highly accurate classification of children with Autism that only required a small selection of items from the Autism Diagnostic Observation Schedule-Generic (ADOS). The ADOS, a semi-structured and standardized assessment used with children suspected of having Autism, has four 30- to 60-minute parts where children are observed for social interaction, communication, play, and imaginative use of materials; it is very widely used in diagnosis of Autism, Pervasive Developmental Disorder Not Otherwise Specified (PDDNOS), and non-spectrum disorders. As they reported in Translational Psychiatry, Professor Wall’s team was able to determine that 8 of the 29 items in Module 1 of the ADOS were sufficient to classify autism with 100% accuracy.
Dubbed the “ADTree,” the Wall team’s classifier overlaps with the revised algorithms for scoring the ADOS, which were developed in part to address the length of the original ADOS. Also, because the lower number of items in the ADTree does not require all of the activities in the original ADOS, the duration of Module 1 can be shortened as well.
Professor Wall and his colleagues concluded their report with this summary:
Currently, autism spectrum disorder is diagnosed through behavioral exams and questionnaires that require significant time investment for both parents and clinicians. In our study, we performed a data-driven approach to select a reduced set of questions from one of the most widely used instruments for behavioral diagnosis, the ADOS. Using machine-learning algorithms, we found the ADTree to perform with almost perfect sensitivity, specificity and accuracy in distinguishing individuals with autism from individuals without autism. The ADTree classifier consisted of eight questions, 72.4% less than the complete ADOS Module 1, and performed with >99% accuracy when applied to independent populations of individuals with autism, misclassifying only 2 out of 446 cases. Given this reduction in the number of items without appreciable loss in accuracy, our findings may help to guide future efforts, chiefly including mobile health approaches, to shorten the evaluation and diagnosis process overall such that families can receive care earlier than under current diagnostic modalities.
Wall, D. P., Kosmicki, J., DeLuca, T. F., Harstad, E., & Fusaro, V. A. (2012). Use of machine learning to shorten observation-based screening and diagnosis of autism. Translational Psychiatry, 2, e100; doi:10.1038/tp.2012.10