In this STSM some identification and de-identification methods with two completely different physiological treats were studied. EEG and Hand, offer a set of entirely opposite strategies to analyze in recognition operation. With the expertise on EEG from the host institution some focus headed EEG methods that can also work with hand geometry. With a public EEG database from UCI some classifications methods were tested. Volatile and non-volatile finance models were also deployed using normalization techniques to improve results.
The signals have been analyzed with the usual sub-bands showing the best recognition rates with the Beta and Gamma sub-bands. These sub-bands showed better results than the raw EEG signal with a thorough classification analysis (Decision Trees, Discriminant Analysis, SVM, KNN and Ensemble Classifiers).