Home > Psychiatry > ECNP 2020 > Autism and ADHD > Movement computing promising in analysing motor impairment in children with autism spectrum disorder

Movement computing promising in analysing motor impairment in children with autism spectrum disorder

Presented by
Mr Thomas Gargot, Hôpital La Pitié Salpêtrière, France
Conference
ECNP 2020
Doi
https://doi.org/10.55788/6b997e51


 

Using movement computing for patient assessment enables precise measurements of movements in the context of both experimental and at-home settings. These methods will likely allow researchers and clinicians to distinguish autism spectrum disorder from other motor disorders and facilitate improved monitoring of children’s progress in more familiar settings such as home or school [1].

Although autism spectrum disorder (ASD) is mainly described as a disorder of communication and socialisation, motor abnormalities are also common. These are significantly correlated with social, communicative, and behavioural impairments. Mr Thomas Gargot (Hôpital La Pitié Salpêtrière, France) and colleagues set up a study to identify computational methods to automatise the assessment of motor impairments in ASD [1]. This was done by a systematic review of the literature between 2000 and 2018, which included all articles discussing automatic assessment or new technologies regarding motor behaviour in children suffering from ASD.

Included in the review were 54 relevant articles, exploring static and kinetic equilibrium (i.e. posture and walking), fine motor skills, motor synchrony, and movements during interaction; all of which can be impaired in ASD patients.

Several devices were employed in these studies to capture relevant motor information, such as (3D) cameras, motion capture systems, and accelerometers. Since 2012, the number of studies has rapidly increased, in parallel with various technologies becoming less invasive, more precise, and more affordable. Examples are Wii boards and Kinect to record balance. Moreover, new algorithms such as deep learning enable researchers to measure posture with a simple camera. Depth sensors allow simple algorithms to record posture and less computational power, which allows for developing real-time measures in games developed for children with ASD.

Thus, open-source and in-house software have enabled the extraction of relevant data on this topic. In a few cases, these technologies have been used in games to measure motor status and progress of children with ASD. As a result, Mr Gargot and his team are currently evaluating the assessment of writing with an electronic tablet using a random forest algorithm.

 

  1. Gargot T, et al. Automatic assessment of motors impairments in autism spectrum disorders: a systematic review. P114. ECNP Congress 2020.




Posted on