MetaBCI, my country's first comprehensive open source software platform for brain-computer interface, is officially released

Providing one-stop brain-computer interface information processing support and interconnection to support multi-center brain-computer interface research... Reporters learned from Tianjin University on November 20 that MetaBCI, my country's first comprehensive open source software platform for brain-computer interface, was officially released. This platform not only fills the software gap for independent research and development of open source ecology in the field of general brain-computer interface in my country, but also provides a basic open platform for cross-team collaborative innovation of brain-computer interface technology.

Brain-computer interface technology is known as the "information highway" for the human brain to communicate with the outside world. It is recognized as a key core technology for the new generation of human-computer interaction and human-computer hybrid intelligence. Building a complete brain-computer interface system requires both hardware and software support. Among them, the software content involves many key technical aspects such as visual and auditory stimulus presentation, data reading and preprocessing, data analysis and decoding, and online feedback.

Currently, most of the brain-computer interface software toolkits that have been published internationally are only oriented to a single link. Developers usually need to use multiple software in combination, which not only increases the R&D cost and cycle, but also improves cross-field cooperation and beginners' experience. The technical threshold is not conducive to the rapid iterative development of brain-computer interface technology.

The MetaBCI platform, jointly developed by Tianjin University, China Electronics Cloud Computer, Suishi Intelligence and other units, is written based on the internationally accepted open source language Python. It standardizes the brain-computer interface data structure and preprocessing process, develops a universal decoding algorithm framework, and uses dual Processes and dual threads improve the real-time efficiency of online systems and enable full-process processing of inducing, acquiring, analyzing, and converting user brain intentions.

The platform contains a total of 376 classes and functions, is compatible with 14 BCI public data sets, covers 16 data analysis methods and 53 brain-computer decoding models. All its codes have been publicly shared on GitHub, the world's largest open source programming and code hosting website, and The accompanying instruction manual provides platform-level technical support to global brain-computer interface developers and enthusiasts.

Introduction to MetaBCI platform architecture.

“The basic architecture of the MetaBCI platform consists of three major modules.” Professor Xu Minpeng, the technical leader of the project and deputy director of the Institute of Medical Engineering and Translational Medicine of Tianjin University, said that among them, the Brainda platform for offline analysis needs unifies existing public data sets The interface integrates a variety of major BCI data analysis methods and decoding algorithms to improve researchers' algorithm development efficiency; the Brainstim platform, which is oriented to stimulation presentation needs, can quickly create a brain-computer interface paradigm stimulation interface by providing a simple and efficient paradigm design module. ; The Brainflow platform for online development needs realizes real-time and high-speed data reading, data processing, result feedback and other functions, lowering the technical threshold of brain-computer interface online systems.

The basic MetaBCI platform is used to conduct steady-state visual evoked potential brain-computer interface experiments.

Cheng Longlong, a data scientist at China Electronics Information Industry Group and general manager of China Electronics Cloud Computer Technology Co., Ltd., said that the R&D team behind the MetaBCI platform is a joint industry-university-research youth army composed of young teachers, corporate engineers and master's and doctoral graduate students. In the future, the team will continue to work on the upgrade, evolution, and iteration of the MetaBCI platform, promote the formation of an open business format of joint research, sharing, and joint construction, open up new fields and new tracks, continuously enhance the global competitiveness of my country's brain-computer interface research, and shape new The new advantages of kinetic energy continue to promote the rapid development of the new generation of brain-computer intelligence.