Enhancing an open-source brain-computer interface software for greater adoption and physiologic data sharing

点击翻译按钮获取中文摘要

SUMMARY There are high technological and software demands associated with conducting brain-computer interface (BCI) research. BCIs are computer-facilitated systems that rely on direct, real-time measures of brain activity for environmental interaction. In order to accelerate the development and accessibility of BCIs, the OHSU BCI research team, during the parent award (R01DC009834), created an open-source software platform, BciPy, available on GitHub. BciPy is written in Python in order to reduce many barriers in the field related to data sharing, storage, and testing new algorithms with existing datasets. BciPy already engages over 30 community members with 17,500 downloads. This supplement allows us to increase the user community through better engagement tooling and integration with cloud services to ensure that experimental data are more accessible and faster to obtain by the broader community than is currently available. This administrative supplement proposes to (1) enhance the use of established open-source BciPy software and (2) accelerate the management, storage, and sharing, via cloud services, of electroencephalography (EEG) and other physiological data. A unique scientific contribution for data science is our dataset, physiologic data acquired from people with severe speech and physical impairments (SSPI) for use in BCI research. Two specific aims are proposed that achieve the quality recognized by the FAIR guiding principles for scientific data management and stewardship. Specific Aim 1 will improve data sharing and enhance software dissemination and collaboration. We will extend BciPy to support a flexible and extensible architecture for specifying and encoding task-, session-, user-, and experiment-level metadata, so that investigators can easily describe their own tasks and protocols for effective data dissemination. We will refine BciPy to accomplish experiment-level management and tagging in line with proposed data standardization methods, such as the Brain Imaging Data Structure (BIDS). We will launch a website at bcipy.github.io to communicate about our listservs, Slack channels, and future workshops. Published guidelines will initiate data sharing protocols by setting up storage resources for automatic upload using Amazon Web Services. Specific Aim 2 will facilitate cloud integration of BciPy data. Offline evaluation of user-generated research algorithms will be encouraged with technical assistance available. Proposed sharing and functionality efforts will significantly increase adoption and implementation of BciPy software and our exceptional acquired physiological data within the international BCI research and development community. Data storage mechanisms are proposed that remain affordable and sustainable in order to support long-term maintenance. The OHSU BCI research team is uniquely positioned for this open data science supplement. With capacity and proven expertise, these aims will significantly move the field forward to improve health outcomes for patients, create stronger widespread end- user engagement and facilitate innovative collaboration.