Computational Intelligence and Brain-Computer Interface Centre
Faculty of Engineering & Information Technology
University of Technology Sydney
Our Research Interest
Our main research goal focuses on translational neuroscience and machine intelligent systems, including algorithm development and brain-computer interface design. The research areas span from cognitive neuroscience research, to fundamental electronic circuits to signal and information processing, and to system realization and evaluation. The short-term research objective is to exploit computational intelligence methodologies for brain-machine Interface, and the long-term goal is to incorporate bio-inspired, brain-like computational capabilities into next-generation computers and robots.
Investigate natural cognition in operational environments
We integrate virtual-reality technology with brain imaging device to investigate human natural cognition. Neurophysiological evidence of mutual relationships between behavioral performance and neurocognition in
complex task paradigms and experimental environments elucidate our understanding of distributed brain dynamics, supporting natural human cognition and complex coordinated, multi-joint naturalistic behavior, and lead to improved understanding of brain-behavior relations in operating environments.
Decipher brain signatures of cognitive lapses, drowsiness, & motion sickness
By applying Intelligent Information Technology and Fuzzy Neural Network to the area of cognition neuroscience, we thoroughly investigate brain dynamics of a vehicle driver, especially when momentary cognitive lapses, drowsiness and motion-sickness are experienced. We also discover models and novel methods for the identification and interpretation of statistical relationships among high-dimensional data sets characterizing the dynamics of environment, behavior, and brain function during complex task performance.
Develop on-line closed-loop systems for human performance augmentation
We translate laboratory-oriented neurophysiological researches to design, develop, and test on-line closed-loop lapse detection and mitigation system featuring a mobile wireless dry-sensor EEG headgear, real-time EEG processing, artifact removal technology, cognitive state monitoring algorithm, and warning feedback system. When drivers experience lapses or fail to respond to emergency events during driving, auditory warning will be delivered to rectify the performance decrements.
Develop wearable and wireless neuroimaging device
We design and develop ultra-lightweight, wearable, wireless, low-cost, whole-head EEG systems featuring a miniature circuit for signal amplification and successively higher sensor densities that can allow assessment of brain activities of participants actively performing ordinary tasks in natural body positions and situations within a real operational environment. We have developed several types of novel dry-contact EEG sensors including foam-based, spring-loaded, and silicon-based sensors that can efficiently reduce the preparation time and provide excellent EEG signals without a conductive gel and skin preparations. Based on the invented dry sensors, a series of EEG helmets are developed, called MINDO4, MINDO16, MINDO32, MINDO64, and MINDO80 for measuring 4 to 80 channels of EEG signals.
Deploy our technology to real-world applications
In additional to EEG-based robotic arm system, we have successfully demonstrated the feasibility of the EEG system with advanced sensing technology and state-of-the-art algorithms to support people with healthcare needs in monitoring sleep, detecting headaches, treating depression, improving attention, and rehabilitation. We are also fusing multiple streams of psychophysiological information such as electrocardiography, electrooculography, actigraphy, galvanic skin response, to construct prototypes of neurocognitive brain-machine interface to improve overall human performance.
Human-machine autonomous (HMA) system
Our study further proposes a closed-loop HMA system to incorporate BCI technologies with machine intelligence. The HMA system dynamically aggregates decisions concerning uncertainties from both human and autonomy agents. A primary goal of the HMA is to balance the capabilities of human and autonomous machine in order to achieve and maintain optimal system performance more efficiently and robustly than either the human or
autonomous agents can achieve independently.
We are devoted into the study of dynamic cognition neuroengineering and has directed many cross-nation research projects in that field. Our research group attracts more than 10 collaborative projects every year. Professor CT Lin has been Chief Investigator (CI) of over 265 research projects funded by government and industry since 1992 and has a total research income (2004-2015) of $46,921K.
Of particular note is the Mega Research Project of Cognition and Neuroergonomics Collaborative Technology Alliance research project (sponsored by the US Army Research Lab) with funding of US$5 million for ten years, in which professor Lin led the participation between his research group and four universities from the US and Germany. The research areas we focus are ‘Real-World Neuroimaging’, ‘Brain-Computer Interface’, and ‘Advanced Computational Approach’.