Reconfigurable Architecture for Autism Detection

Autism syndrome disorder (ASD) is a neuropsychiatric disorder generally recognised by audio-visual discrepancy in the children. This discrepancy is normally identified by the social reciprocity and interactive skills compared to normal children between three to five years of age. The basic objective of this work is to identify the dominant features related to autism, perform optimum feature selection for increased accuracy and train a model for detection of autism using hybrid technique of constructing feature vector using quantitative approach and EEG. Further a robust customized reconfigurable architecture is proposed for the detection of ASD. The proposed custom hardware can be the part of concerned medical applications oriented architecture. The developed system will be helpful for medical practitioners to monitor the development of the children at certain intervals of times after the birth. - - SEEGSignal Questionrnaire $tatistics Pre- Pre processing processing Sectior A;PC based 3 Processing Feature Parametric 1 extraction Extraction 7 Seletiori P aaeers F w Li re Sp i ic aamKm a t Recorifigurable FPGA based Archieture Secton 8: 10 Reonfigurable k Architecture 12 Graphics LCD Display Figure 1