/ EN
2026-09-14 00:00 China Papers Foundations & Methods Translated from EN

Tsinghua Team Launches TrustBCI to Share BCI Data Without Moving It

Summary Training accurate BCI models depends on large, diverse EEG datasets, but that data is usually locked inside individual hospitals and labs, stored in inconsistent formats and restricted by privacy rules, which makes it hard to share. TrustBCI brings three functions into one platform: turning raw BCI datasets into assets that can be exchanged, using incentives to keep institutions contributing data, and letting users run computations without access to the raw records. The platform demonstrates two workflows, controllable data synthesis, which generates new datasets, and privacy-preserving query, which returns only query results. That suggests BCI models could be trained jointly across institutions without first pooling the data in one place.
Why it matters The BCI field's data shortage stems less from a lack of collection than from the fact that recorded data cannot leave the institution that holds it. This platform combines data assetization, contribution incentives and privacy-preserving computation in a single process and comes with a working web prototype, showing that cross-institution sharing is workable in engineering terms. It tackles data circulation, not decoding accuracy, and real-world uptake will depend on how many institutions are willing to connect their data.

BCIwiki (bciwiki.com) — Researchers at Tsinghua University have proposed TrustBCI, a platform for collaborative exchange of brain-computer interface data between institutions, reported in Proceedings of the VLDB Endowment on September 14, 2026. The team notes that high-quality BCI datasets are essential for developing accurate and generalizable models, yet in practice they remain isolated across institutions, heterogeneous in format and difficult to share under privacy constraints. Existing BCI data platforms address only part of the problem, the authors write, and offer no unified support for data contribution, privacy-preserving computation and downstream use.

According to the paper, TrustBCI combines three components: assetization tools that prepare raw BCI datasets, incentive mechanisms intended to sustain data contribution, and privacy-aware services that let users work with protected data without exposing raw records. The demonstration focuses on two representative workflows, controllable data synthesis and privacy-preserving data query, and shows dataset-level and query-level access to protected BCI data through a functional web prototype. The authors are Jianyu Yang, Jiannan Wang and Guoliang Li. More broadly, the team writes, TrustBCI illustrates how protected domain data can be turned into usable data services through a unified data platform.

The platform is presented as a demonstration prototype. The paper does not disclose how many institutions participate, the volume of data involved, any deployment, or performance overhead figures for privacy-preserving computation on BCI data.

Compiled by BCIwiki from public sources

Sources · 1
doi.org 2026-09-14

清华大学 timeline

2026-07 Beihang and Tsinghua Team Fuses EEG, Video and Motion to Flag VR Cybersickness All entries →
Read original ↗
Suggest a correction Revisions · none / EN
© 2026 BCIwiki.com Digest Topics Tips Subscribe Revisions About