Feb. 23, 2012
In the laboratory of Qiushi Academy for Advanced Studies on Zijingang Campus, Zhejiang University, a monkey named "Jianhui” was practicing four different moves of grabbing, hooking, holding and pinching to deal with four objects of different shapes in front of him. The right move and thus the successful holding of the objects would win him a bonus of drinking. What’s amazing was that in a place not too far away, a mechanical hand was performing exactly the same hand moves as Jianhui did, just like being synchronized by “telepathy” with Jianhui.
What’s described above is the latest research progress made by Brain-Computer Interface Research Team at Qiushi Academy for Advanced Studies. The team successfully captured and deciphered the neural signals of the four gestures aforementioned in the monkey’s brain, thus making it possible for the monkey’s mind power to control the external mechanical hand. This achievement represents the highest level brain-machine interface research in China, and is complementary with the relevant international cutting-edge research. The research in this field brings hope to disabled people; more importantly, it establishes a research model for the comprehensive deciphering of fine neural signals in brain and transmitting such signals to machines.
Brain is like a container filled with liquid and neurons. Hundreds of billions of neurons interact by transmitting tiny electrical impulses and give orders to body parts so that the body can perform actions such as walking, eye-brow lifting, etc. The neural signals have long been double Dutch to human. But from the 1990s, the rapid development of computer science gives growing hope of understanding this double Dutch. The technology called Brain-Machine Interface (BMI) is committed to the establishment of transmission of signals from brain to external devices such as artificial limbs. Such signal transmissions allow the brain signals to reach the external devices to perform independent actions for mobility-impaired people, even in case of spinal cord injury and neural pathway damage.
As Professor Zheng Xiaoxiang of the Brain-Computer Interface Research Team introduces, the research in this field needs the collaboration of multi-disciplines, including neuroscience, information engineering, medicine and so on. The key is to develop real-time multi-channel neuron collecting, processing and deciphering technology with high accuracy and mutual adaptability.
First of all, brain surgeons from the Second Affiliated Hospital of Zhejiang University performed a surgery on Jianhui’s brain, implanting two chips connected to more than 200 neurons in his motor cortex. The other end of the chip is connected to a computer which records the real-time neural signals issued by Jianhui with every move. On a computer screen, these electroencephalograms (EEG) are converted to “sounds” of different frequency and length. Researchers in biomedical engineering, computer information and medicine develop a real-time analysis system based on the neural signals acquired which deciphers the 200 neural discharge signals recorded and eventually differentiates Jianhui’s “secret language” of the four moves grabbing, hooking, holding and pinching.
When Jianhui is happily playing with his toys, his brain signals are intercepted and deciphered and then transmitted to the mechanical hand. Hence the telepathy. “Hand moves are associated with at least several hundreds of thousands of neurons. We now decipher the moves based on the signals of about 200 neurons. Of course, the orders we produced are still distant from the truly flexible finger moves in complexity and fineness.” Prof. Zheng commented.
If “telepathy” can apply to various parts of human body, it can greatly improve the quality of life of physically disabled people. Scientists around the world are working to make this real. In 2008, scientists at University of Pittsburgh announced the successful control of mechanical arms based on monkey’s mind power. In October 2011, scientists at Medical Center of Duke University published a paper on Nature, declaring that they not only made it possible for the monkey’s mind to control the movement of a virtual palm, but also felt the tactile signals of the virtual palm touching objects. The uniqueness of Zhejiang University research lies in the fact that the neural signals captured are finger signals with higher complexity and finer than those in arm movements. The mechanical hand controlled by Jianhui’s mind is one of the most sophisticated mechanical hands in the world, developed by the laboratory headed by Prof. Carrozza at Sant’ Anna School of Advanced Studies in Italy. Now Jianhui is also working well with another domestically developed mechanical hand.
Qiushi Academy for Advanced Studies was founded in 2006 and has been endeavoring to conduct interdisciplinary brain-computer interface research. The academy has established collaborative academic relations with top international research institutions in this field. In 2006, the research team built an animal navigation system implanted into the brain of a rat to make it walk in accordance with human orders. In 2007, the team developed an EEG-based Chinese typewriter. In 2008, the team started the deciphering of the neural signals on rats’ motor cortex. In 2010, the deciphering of the arm movements of monkeys was successfully achieved. In addition, the team has developed a number of neurological rehabilitation equipment now in clinical trials.