I worked on using machine learning to classify alzheimer's vs healthy controls using EEG signals. The features used were magnitudes of FFT and CWT over various frequency bands and have obtained some very good results as well.
1. Astolfi, Laura; Toppi, Jlenia; Borghini, Gianluca; Vecchiato, Giovanni; Isabella, R.; Vico Fallani, Fabrizio de et al.: Study of the functional hyperconnectivity between couples of pilots during flight simulation. An EEG Hyperscanning Study. In: IEEE Eng Med Biol Soc., Bd. 2011, S. 2338–2341. Online verfügbar unter http://ieeexplore.ieee.org/servlet/opac?punumber=6067544.
2. Astolfi, Laura; Toppi, Jlenia; Vico Fallani, Fabrizio de; Cincotti, Febo; Wilke, Christopher T.; Yuan, Han et al. (2011): Methods for the EEG Hyperscanning. Simultaneous Recordings from Multiple Subjects during Social Interaction. In: 2011 8th International Symposium on Noninvasive Functional Source Imaging of the Brain and Heart and the 2011 8th International Conference on Bioelectromagnetism. Online verfügbar unter http://ieeexplore.ieee.org/servlet/opac?punumber=5910695.
3. Fukuda, Haruaki; Shiomi, Masahiro; Nakagawa, Kayako; Ueda, Kazuhiro (2012): 'Midas touch' in human-robot interaction. New York, NY: ACM. Online verfügbar unter http://dl.acm.org/citation.cfm?id=2157689.
4. Nakamura, Tsukasa; Tomita, Yohei; Ito, Shin-ichi; Mitsukura, Yasue (2010): A method of obtaining sense of touch by using EEG. 19th IEEE International Symposium on Robot and Human Interactive Communication ; 13 - 15 Sept. 2010, Viareggio, Italy. Piscataway, NJ: IEEE. Online verfügbar unter http://ieeexplore.ieee.org/servlet/opac?punumber=5593942.
I need to analyze EEG data for before arithmetic task and during arithmetic task. Please suggest me about perfect statistical features. Thank in advanced.