In 2 previous studies we have shown the ability of a special machine learning system applied to digital standard (19 channels) EEG data in distinguishing ASD from non ASD children with an overall accuracy rate of 100% and of 98.4% respectively. Since the equipment routinely available in neonatology units employ often few channels, we were curious to check if just two channels were enough to allow a good performance in the same cases of the above-mentioned studies. A continuous segment of artifact-free EEG data lasting 1 minute in ASCCI format from C3 and C4 EEG channels present in 2 previous studies, the first carried out in Italy on 25 subjects (15 ASD and 10 typicals) and the second carried out in US on 40 subjects (20 ASD and 20 with other NPI disord.) was used for features extraction and for subsequent analyses with advanced machine learning systems. A features extraction software package (Python tsfresh) applied on time-series raw data derived 1588 quantitative features. A special hybrid system called TWIST, coupling an evolutionary algorithm named Gen-D and a back propagation neural network was used to subdivide the dataset into training and testing sets as well as to select features yielding the maximum amount of information. After this intelligent preprocessing, 14 features were extracted from C3-C4 timeseries of study 1 and 31 C3-C4 timeseries of study 2 representing the EEG signature. Acting on these features the overall accuracy predictive capability of the best artificial neural network acting as classifier in deciphering autistic cases from typicals (study 1) and from other NP disorders (study 2) resulted 100 % for study 1 and 93.3 % for study 2. The results of this study suggest that also a minor part of EEG contain a precious information useful to detect autism if treated with advanced computational algorithms. This could allow in the future to use standard EEG from newborn to check if ASD signature is already present at birth.

* Autism Research Unit, Villa Santa Maria Foundation, Tavernerio, Italy
**Ephoran Multi Imaging Solutions, Milano, Italy