After receiving permission to access the large Autism repository, we have been assessing the best approach for analysis. The datasets contain a range of information, including EEG, Eye movement, ECG and genomic information.

We plan using deep learning to obtain representations of these datasets to see if predictions can be made to better cluster types of autism. By having richer clusters of autism, we will gain higher significance levels when predicting treatments and symptoms, which will lead to better learning models.

In this way, we hope to split up the colours on the Autism spectrum to find features of each individual colour