Overcoming the Lack of Data to Improve Prediction and Treatment of Individuals with Autistic Spectrum Disorder and Attention Deficit Hyperactivity Disorder

The problem of lack of data remains a major drawback regardless of the continuous evolution of machine learning and deep learning models. This paper presents a modular and scalable architecture for the prediction and treatment of autism spectrum disorder (ASD) and attention deficit hyperactivity disorder (ADHD). With this architecture, therapists will be able to collect data from individuals from anywhere and at anytime, thanks to mobile devices, which will enable personalised monitoring. One of the main objectives of this ongoing project, which has a very widespread international projection within the framework of social inclusion, is the creation of a new collection of data due to the lack of data in this area. As a result, we will be able to place it in important repositories and specialised journals and thereby make it available to the scientific community. This architecture has been evaluated with several supervised and unsupervised machine learning algorithms in order to identify possible candidates for ASD/ADHD diagnosis. The initial results are very encouraging despite the small volume of data because they allow the possibility of developing personalised dashboards, which specialists can adapt to the personalised treatment of each individual according to the indicators obtained.

Automated summary

This paper presents a modular and scalable architecture for the prediction and treatment of autism spectrum disorder (ASD) and attention deficit hyperactivity disorder (ADHD). The architecture allows therapists to collect personalized data from individuals using mobile devices, enabling personalized monitoring. Initial results are very encouraging, despite the small volume of data, as it allows for the development of personalized dashboards for individualized treatment.