A Reconfigurable General Behavior Data Acquisition System For Motor Brain Machine Interface

Lichao Wang,Guanghao Sun,Zhaozhong Gui,Yiwen Wang,Shaomin Zhang,Xiaoxiang Zheng
DOI: https://doi.org/10.1109/NER.2015.7146578
2015-01-01
Abstract:In motor brain machine interface (BMI) research, it is necessary to collect behavior data and neural signals of subjects for neural decoding. However, most behavior data acquisition systems are customized for specific experiments because of variety of experiment goals. It is difficult for researchers to reconfigure their existing system for a new experiment environment, whenever they need to design a new behavioral experiment. In this paper, a reconfigurable general behavior data acquisition system was developed. It can provide integral structure and rich interfaces. The system consists of a microcontroller, three types of behavior data input interfaces, three types of system output interfaces and a communication module. It can acquire, digitize, process behavior data and send them to personal computer or smartphone. To evaluate the facility of this system, two typical behavioral experiments used in BMI studies were carried out. In these experiments, monkeys were trained to perform track, center-out, reach and grasp tasks. By reconfiguring the system with some external devices, not only all kinds of cues and triggers could be generated successfully, but also event information and behavior data of the upper limb movements were recorded in real time. The results show that researchers can easily connect this system with various types of input and output devices and reconfigure system settings according to their specific research purposes. The system can also provide a useful tool for those end users without strong engineering background to quickly set up their experiment environment.
What problem does this paper attempt to address?