Tuyển tập báo cáo các nghiên cứu khoa học quốc tế ngành hóa học dành cho các bạn yêu hóa học tham khảo đề tài: Signal Processing Institute, Swiss Federal Institute of | EURASIP Journal on Applied Signal Processing 2005 19 3087-3088 2005 Hindawi Publishing Corporation Editorial Jean-Marc Vesin Signal Processing Institute Swiss Federal Institute of Technology 1015 Lausanne Switzerland Email Touradj Ebrahimi Signal Processing Institute Swiss Federal Institute of Technology 1015 Lausanne Switzerland Email Brain-computer interfaces BCI an emerging domain in the field of man-machine interaction have attracted increasing attention in the last few years. Among the reasons for such an interest one may cite the expansion of neurosciences the development of powerful information processing and machine learning techniques as well as the mere fascination for control of the physical world with human thoughts. BCI pose significant challenges at both the biomedical and the data processing levels. Brain processes are not fully understood yet. Also the information on the dynamics of these processes up to now gathered mainly with electroencephalographic EEG or functional magnetic resonance imaging fMRI systems is incomplete and more than often noisy. As such it is important for BCI applications to determine how physically the maximum amount of information can be extracted and to design efficient tools both to process the data and to classify the results. This special issue presents nine papers exhibiting a rather balanced state of research and development in BCI. Three papers deal with information extraction three with signal processing aspects and three present applications. Moreover while most current efforts concentrate on continuous EEGbased techniques fMRI implanted microwire electrode and evoked potential-based techniques are also presented. In the first batch of three papers on information extraction A. Meyer-Base et al. study independent component analysis ICA and unsupervised clustering techniques and combine them to produce task-related activation maps for fMRI datasets. M. Schroder et al. .
