Header image  
Tutorials: Nov 13, 2011, Shanghai, China
Main Conference: Nov 14-17, 2011, Shanghai, China
Workshops: Nov 18, 2011, Hangzhou, China
line decor
   Final program of ICONIP2011 and book of abstracts are available now.  
line decor

Workshops on Brain-Computer Interface and Applications

Keynote Speaker

Jinglong WU

Okayama University, Japan.

Cognitive Neuroscience for Brain-Computer Interface


        Recently, the exploration of human brain functioning has focused on basic research in cognitive neuroscience and frontiers technology of brain-computer interface. With the development of engineering technologies such as measurement technology, information technology, and artificial intelligence, we can record brain activation with millisecond in the time domain and with millimeter in the spatial domain, during basic cognitive function and higher-level brain function (e.g., memory, language, and attention). In present talk, I will introduce a new research approach that combines brain-computer interface and cognitive neuroscience with using four frontier research topics in my laboratory. They are 1) tactile perception and neurology, 2) brain function and attention, 3) brain function and basic visual cognition and 4) brain function and language.

Biographical Sketch

        Dr. Wu received a Ph.D. degree from Kyoto University, Japan in 1994. From 1999, he was an Associate Professor and was promoted to a permanent full Professor of Kagawa University, Japan in April, 2002. Since 2008 he has been a Professor and Lab. Head of Biomedical Engineering Laboratory of Okayama University, Japan. Dr. Wu has mainly been involved with the studies of human mechanism of visual, auditory and tactile information processing. He has made significant contributions to the study of the interactive human brain mechanism between lower function (sensation and perception) and higher function (attention and language) using electroencephalograph (EEG) and functional magnetic resonance imaging (fMRI) techniques. Dr. Wu became involved with several associations, and established an international society the Institute of Complex Medical Engineering. He also served as several associate editors of the international journals. He has served on the chair for many international conferences, and served on the leader for many projects of Japanese government, foundation, society and company. His hard work and dedication has earned him several Best Paper Awards, Tomoda's Prize, Gennai Grand Prize and Ozaki Foundation, respectively. Dr. Wu has contributed over 200 original papers which have been published in international journals and international congresses. In addition to his contributions, he is the author of six books and five patents.

Bo Hong

Associate Professor, Department of Biomedical Engineering, Tsinghua University, Beijing, China.
Associate Editor, IEEE Transactions on Neural Systems and Rehabilitation Engineering.

Sensory and cognitive neural markers for new brain computer interface


        Most of current event-related potential(ERP) based brain computer interfaces (BCI) employ attention explicitly or implicitly, to modulate the neural responses of target stimuli. However, user's attention level is hard to be maintained in a passive BCI paradigm, posing the challenge of advanced signal processing to extract unstable target ERP. Here, we proposed a series of new BCI paradigms, using subject's voluntary mental responses as neural markers along visual and auditory sensory streams, to produce more distinct target ERP, which makes it possible for reliable target detection using fewer trials of sensory stimulus. Google search and intention expression BCI systems for disable people are demostrated using these new paradigms. Possible extension of these BCIs to the cortial EEG level will also be discussed.

Biographical Sketch

        Bo Hong received his B.S. and Ph.D. degree of biomedical engineering from Tsinghua University, Beijing, China, in 1996 and 2001, respectively. During the year 1999 to 2000, he received research training of psychophysiology in the department of psychology, Chinese University of Hong Kong. He was assistant professor in Tsinghua University since 2001. On his sabbatical leave from 2004 to 2005, he was a visiting faculty in the department of biomedical engineering and the center for neural engineering at Johns Hopkins University, Baltimore, USA. Since 2005, he has been associate professor with department of Biomedical Engineering, School of Medicine, Tsinghua University, Beijing, China.

        He is currently associate editor of IEEE Trans. Neural Systems and Rehabilitation Engineering, and reviewer of Journal of Neural Engineering, IEEE Trans. Biomedical Engineering, Neuroscience Letters, Neurocomputing, etc.


National Key Laboratory of Science and Technology on Nano/Micro Fabrication Technology, Shanghai Jiao Tong University, China

Dry Electrode Array for Electroencephalography Recording


        A novel dry electroencephalography (EEG) electrode has been successfully fabricated for the vigilance analysis based on EEG signal acquisition, which is conventionally acquired by wet electrodes. Flexible substrate MEMS fabrication technique has been applied to fabricate the flexible dry micro-needle electrode array, which are chemical stable and bio-compatible. Featuring its cantilever structured micro-needles, the electrode was fabricated by wet etching of Cu sacrificial layer process. The electrodes were then released from substrate by PDMS lift-off process and packaged by polyimide, and finally fabricated a three-dimensional electrode array by multi-layer assembly. Ni was electroplated to fabricate the micro-needles, the wire and the pads of the dry electrode to obtain mechanical strength while Au was electroplated on the surface of micro-needles for bio-compatibility. The performance of the dry electrode has been tested by amplifier of EEG acquisition, Neuroscan, and compared to that of conventional wet electrode. The impedance of dry electrode was approximately 10 kΩ, and the similarity between the performances of dry and wet electrodes has been proved by comparison of their time domain signals and frequency domain signals. With the advantage of high yield, small size, simple assembly, fine shielding, good reliability and mechanical strength, the dry electrode is capable of convenient and painless EEG signal acquisition, being more widely used due to its possible portability.

Biographical Sketch

        Professor Di Chen received his diploma chemist and doctor of natural science from university of Hamburg, Germany in 1986 and 1990 respectively. From 1990 to 1992 he was post-doctoral researcher at Micro Parts Company, Germany. Since 1993 he was associate professor and professor at Shanghai Jiao Tong University. He is now vice director of Research Institute of Micro/nano Science and Technology, Shanghai Jiao Tong University. He has undertook many research projects in microfabrication technology and micro-electro-mechanical systems (MEMS) by Ministry of Science & Technology, Ministry of Education, National Natural Science Foundation of China, Education Commission of Shanghai Municipality, Science & Technology Commission of Shanghai Municipality, National State Key Laboratory, 3M and SIEMENS Company. He has published more than 130 papers and authorized 16 Chinese invention patents. The published papers were cited more than 270 times. He has written parts of 4 books in the field of MEMS and microsensors, and translated the book “Introduction to Microfabrication? He received the second prize of technology invention of China (2008), first prize of technology invention of Shanghai Municipality (2007) and second prize of natural science of Ministry of Education (2008). He is member of editorial board of several MEMS journals. His research fields are development and application of LIGA-like microfabrication technique, implantable microsystems, microfluidics and RF MEMS.

Weihua PEI

Institute of Semiconductors, CAS, China

Dry electrode and its applicaton in biopotential recording


        A micro-needle array based “dry?electrode was fabricated to detect biopotential and to develop a potable BCI system. Different methods were used to fabricate microneedles, and the characteristics of these methods were discussed. Testing results in vivo shows that the impedance value and recording ability of the dry electrode can be comparable with, if can not be better than, commercial electrode in a wide frequency range. An wearable Brain-computer interface (BCI) is designed and fabricated base on the dry electrodes. With dry electrodes, skin preparation and application of electrolytic gel are not required. The BCI system includes a wireless transmitter module and an receiver module, Electroencephalography (EEG) is acquired using dry electrodes, amplified and processed by an application specific integrate circuit (ASIC), and transmitted to the receiver by RF chip. The wearable transmitter module weighs only 39g. The transmitter consumes 60 mW of dc power and generates an output power of 0 dBm. With this BCI system, a trained subjects can play game or control a toy car with a light head band.

Biographical Sketch

        Dr. Weihua Pei received the Ph.D. degree in Microelectronics & solid electronics from state key laboratory on integrated optoelectronics, Institute of Semiconductors, Chinese Academy of Science, in 2005. He has worked as a Post-doctor at Max-Planck-Institute of Microstructure Physics and Department of Biomedical Engineering, Tsinghua university, from 2006-2008. He is now an associate professor at Institute of Semiconductors, CAS. His research focuses on the area of optoelectronic devices and integrated technology. The Focused Research Center is on developing micro-structure, biosensor or other devices for applications in neural engineering.

Hongtao LU

Department of Computer Science and Engineering, Shanghai Jiaotong University, China.

Vigilance Estimation and Real-time Monitoring Based on EEG and Infrared Facial Image Video


        We implemented a vigilance estimation and real-time monitoring system based on EEG and infrared facial image video. Several algorithms have been proposed to estimate vigilance. First, we use support vector machine (SVM) to identify light drowsiness state from other states to estimate vigilance level decline. Light drowsiness EEG is marked by alpha energy increasing to 50%. Alert EEG is marked by dominant beta activity and other EEG is labeled as sleep state. Samples of EEG data are trained by SVM using 4 features from each frequency band. Mutual information based feature selection method is used to reduce the dimension of features. Second, we use sparse representation of EEG to classify three vigilance states. Six features from each frequency band are got from samples of EEG data. Random feature is used to reduce the dimension of features. No training process is needed before the classification. Third, we use the continuous wavelet transform to extract a large set of features from EEG. We then use the random forest approach to rank the plenty of features and select the most important ones for classification.

        For the purpose of vigilance estimation from facial image video, we constructed an infrared illumination and video capture system, and then use the techniques of eye detection, eye tracking and head pose estimation to get facial parameters. Fuzzy integral method is used to obtain the vigilance result based on facial video.

Biographical Sketch

        Hongtao Lu got his Ph.D degree in 1997 from the Department of Radio Engineering of Southeast University, Nanjing, China. Then he become a postdoctoral fellow in the Department of Computer Science of Fudan University, where he had stayed for 2 years. From 1999, he join the Department of Computer Science and ngineering, Shanghai Jiaotong University. He is now a professor there. His research interests include machine learning, pattern recognition, computer vision.

Liqing ZHANG

MOE-Microsoft Key Laboratory for Intelligent Computing and Intelligent Systems, Department of Computer Science and Engineering, Shanghai Jiao Tong University, China.

EEG Tensor Representation and Spatial-Temporal Pattern Recognition for Real Time BCI Systems


        Brain Computer Interface (BCI) aims at transferring human's intents into commands to control computers or machines. Recognizing the human's intents from EEG data robustly in real time is critical for BCI systems. In this talk, we introduce tensor data structures to represent the EEG data and further develop three types of features extraction methods. Sparse constraint is imposed on the feature extraction for characterizing the sparseness of cognitive patterns in EEG data. And furthermore a regularized Tensor Discriminative Analysis (RTDA) is introduced to find the discriminant features of multi-task motor imaginary patterns. To verify the performance of the developed tensor features, computer simulations are given to compare the accuracy rate in classifying single trial EEG data, indicating that the tensor based features are able to capture discriminative features and thus achieve higher accuracy rate in cognitive pattern recognition.

Biographical Sketch

        His long term goal is to understand how intelligent information is processed in the brain and develop new type (brain-like) computational models and algorithms for visual and auditory information processing.

        Currently, his research interests cover brain-like computing model and its computing mechanism, visual information representation and global feature analysis, brain signal processing and brain-computer interface, perception and cognition computing model, statistical learning and inference. He has published more than 160 papers in international journals and conferences.

        He serves as the associate editor of "International Journal of Computational Intelligence and Neuroscience", the director of the committee of Biocybernetics and Biomedical engineering, Chinese Automation Association; member of Chinese Neural Network Society, member of neuroinformatics and neuroengineering committee, Chinese Neuroscience Association. He is also the reviewer of a number of international journals, such as IEEE Trans. Neural Networks,IEEE Trans. Signal Processing,IEEE Signal Processing Letters

Copyright© 2010-2011 International Conference on Neural Information Processing. All rights reserved.