Artificial Intelligence diagnosis

Post Reply New Topic

In a study, whose findings were published in the journal Complex & Intelligent Systems, the researchers examined and tested two sets of nonlinear, highly distinctive features of EEG signals extracted from children aged four to 13. One set belonged to 40 children diagnosed with autism spectrum disorder (ASD), while the other set was extracted from 37 children without any neurodevelopmental disorders. They were able to identify three distinct combinations of the EEG signal features that were "strongly predictive" of autism.

According to a media release, the researchers are now embarking on the next stage of their study where they will gather more data to test and develop a cloud-based system to automatically detect autism in a single brain scan.

The proposed cloud-based system will feed live data from a brain scan into a cloud service that runs on machine learning algorithms to analyse and rapidly classify the EEG signals into ASD and non-ASD categories.

In the US, there is an existing diagnostic support tool that helps paediatricians and primary care clinicians evaluate and diagnose autism in children. The Canvas Dx by pediatric behavioural health firm Cognoa uses AI algorithms to analyse videos showing a child's behaviour and responses to questions.

There is a real need for an automated diagnostic tool to aid healthcare professionals in diagnosing ASD accurately and early, particularly clinicians who may not have a lot of experience with children with autism," he added.