SECOND PLACE
Omer Lerman
Autism detection using nonverbal communication features in ADOS tests
Omer Lerman1, Gianpaolo Alvari2, Marissa Hartston3, Batsheva Hadad3, Hagit Hel-Or1
1Department of Computer Science, University of Haifa, 2Department of Psychology and Cognitive Science, University of Trento, 3Department of Special Education, University of Haifa
Autism Spectrum Disorder (ASD) is a neurodevelopmental disorder that is marked by defects in social communication and interaction as well as repetitive behavioral patterns. ADOS is a standardized test for assessing and diagnosing ASD. The ADOS examiner assesses the subject’s behavior and responses and assigns a numeric ADOS score to different categories, ADOS scoring may be affected by the examiner’s subjective views and prior experience. In both clinical practice and research, there is a growing demand for standardized qualitative measures that can facilitate these diagnostic procedures. Using the recordings of ASD and control subjects, we measure
objective parameters such as eye contact duration, vocal pitch and intensity variation, changes in facial expression and quantification of body movement during the different ADOS tasks. These objective parameters will be compared with the subject’s ADOS score as assessed by the examiner in order to identify key parameters in ASD assessments.
Methods:
In this work, we use a novel 3-camera setup to collect audio-visual data
during ADOS testing which will be used to analyze subjects’ skills and
behaviors with emphasis on nonverbal communication signals, such as
speech prosody, gaze, facial expression and body movement.
Results:
Our analysis produced preliminary results in several modalities:
