Automatic recognition of therapy progress among children with autism

Abstract

The article presents a research study on recognizing therapy progress among children with autism spectrum disorder. The progress is recognized on the basis of behavioural data gathered via five specially designed tablet games. Over 180 distinct parameters are calculated on the basis of raw data delivered via the game flow and tablet sensors -i.e. touch screen, accelerometer and gyroscope. The results obtained confirm the possibility of recognizing progress in particular areas of development. The recognition accuracy exceeds 80%. Moreover, the study identifies a subset of parameters which appear to be better predictors of therapy progress than others. The proposed method -consisting of data recording, parameter calculation formulas and prediction models -might be implemented in a tool to support both therapists and parents of autistic children. Such a tool might be used to monitor the course of the therapy, modify it and report its results.Autism is a complex developmental disorder that influences the ability to communicate and learn. Autism is nowadays a growing challenge, as the number of children diagnosed with autism is rising worldwide 1 . The disorder exhibits a spectrum of symptoms that might range from mild to severe in a particular case, varying from skill to skill and from child to child. This makes diagnosis and therapy progress evaluation difficult and at the same time crucial for the effectiveness of the therapy. Early identification and proper therapy translates into a greater chance for preventing a person with autism from social exclusion. Therefore, any idea that might improve therapeutic practice is worth investigating. This paper presents one such idea, which incorporates the analysis of behavioural characteristics observed in autistic children while they play specially designed tablet games. This interdisciplinary study combines computer science and special pedagogy, by applying computer technologies and machine learning methods in the process of therapy for autistic children.Computer technologies may support both the diagnosis of autism and related therapy, although most solutions so far refer to the process of therapy. There are numerous applications designed for individuals with autism. These applications focus on particular issues by teaching specific skills-e.g. expressing needs, learning certain behaviours 2 , improving verbal communication, answering questions, interacting with other people in typical situations 3 , recognizing and expressing emotions 4-6 . These solutions take advantage of the fact, that children with autism are usually enthusiastic about tasks supported by computer technology, which offers a predictable framework without causing stress 2 .Another group of tools is designed for therapists. One of the most commonly used solutions are computer versions of standardized questionnaires that evaluate an individual's state 7 . Another popular way of supporting the therapy is to provide the experts with video recordings of the children's behaviour. However, there are also some...

Precision Telemedicine through Crowdsourced Machine Learning: Testing Variability of Crowd Workers for Video-Based Autism Feature Recognition

“…Since autism consists of a largely behavioral phenotype, video data are a particularly powerful and rich means of capturing the range of social symptoms a child may exhibit in a fast and virtually cost-free manner. Accurate diagnoses and behavioral classifications have been inferred from categorical ordinal labels extracted by untrained humans from the short video clips [ 5 , 6 , 7 , 8 , 9 , 10 ], which are recorded by digital mobile and wearable interventions during use by the child or administering parent [ 11 , 12 , 13 , 14 , 15 , 16 , 17 , 18 , 19 , 20 , 21 , 22 ]. Such a process can be scaled through crowdsourcing platforms, which allow distributed workers from around the globe to perform short on-demand tasks.…”

Implementation of Wavelet Analysis on Thermal Images for Affective States Recognition of Children With Autism Spectrum Disorder

“…In many research [49]- [51] and from the consultation of expert therapists, guided questionnaire responded by the caregivers, done in a very controlled and structured environment was the best option to gage the affective state of the autistic children. In this research the questionnaire was used to verify the classifier and help guided the development of classifier in the prior stage based on the response obtained from the control group (the TD children who answered the questionnaire themselves).…”

Sensor-Based Technology for Social Information Processing in Autism: A Review

“…For monitoring social interactions, Winoto et al [79] created a machine-learning-based social interaction coding of 3D data around a target user. Kolakowska et al [80] approached automatic progress recognition with different tablet games. Over a 6-month time window, they were able to identify movement patterns in their study group of children with ASD ( N = 40), that not only related to development in fine motor skills but also other fields like communication and socio-emotional skills.…”