Development of a microcontroller EEG-based system for diagnosis of autism spectrum disorder in developing countries

Biomedical signals consists of records of electrical activity in the human body, and they represent the health status of individuals. The biomedical signal otherwise called neural signals includes Electro-encephalogram (EEG) Signals, Electro-cardiogram (ECG) Signals, Heart Rate Variability (HRV) Signals, Electro oculorgram (EOG) Signals, Electro-myogram (EMG) Signals, magnetoencephalography (MEG) signals, and electro gastrography (EGG). It is often difficult for health practitioners to visually examine these long records to diagnose a patient to arrive at conclusions. The process of classifying a biomedical signal is done by carefully attaching a signal to a disease state or healthy state and also the quality of features extracted from the signals, well pre – processed signal and the classification process determines the classification Accuracy (CA). The spectral content of the signals contains critical information on state of health of a person that can used for early detection of a particular disease. Developing an Automated system which could help in Automated classification of these signals can greatly assist the Doctors and non-technical individuals in the diagnostic process. In response to the above drawback, this paper intends to develop a prediction machine in a dual level approach using a Microcontroller EEG - based system and machine learning Algorithms for early diagnosis of Autism Spectrum Disorder using EEG signals. The performance of the proposed system will be evaluated using CA, amplitude, power frequency ratio, execution time, specificity, sensitivity, power spectral density ratio, memory usage and power consumption in a microcontroller platform time.