Mobile eeg based hunger and satiety classification

Surface electroencephalography (EEG) measurements that can be performed in hospitals and laboratories have reached wearable and portable level with the development of today's technologies. Artificial intelligence assisted, brain computer interface (BCI) systems play an important role in the processing of EEG signals of individuals with disabilities and their interaction with the outside world. Especially, with the increasing population, researches to support the basic needs of individuals in need of home care are becoming widespread. In this study, it is aimed to design the BCI system that will detect the hunger and satiety status of the people in computer environment through EEG measurements. In this context, the database was created by recording EEG signals in the eyes open, eyes closed and Event Related Potential (ERP) scenarios of 20 healthy participants in the first stage of the study. In preprocessing, EEG signals are cleaned from noise using low pass, high pass and notch filters. The properties are extracted by obtaining the maximum, minimum, average, median, variance, kurtosis, mode, difference between maximum and minimum values, standard deviation and skewness values of the ERP signals. Wavelet Packet Transformation (WPT) method was used to analyze EEG measurements with eyes open and closed. In the feature selection, the properties with the most successful accuracy rate, which are the inputs of the Linear Discriminant Analysis (LDA) classifier, are used as the inputs of artificial intelligence algorithms. In the study, the classification performances of the EEG signal in the open and satiety states of Coiflet 1 and Daubechies 4 wavelets were compared by using Multilayer Artificial Neural Network, Support Vector Machine, k Nearest Neighbor and Decision Tree algorithms. As a result of the study, the accuracy rate was determined as 97.62%, 95% and 85% with three different measurement methods and analysis.