Institute of Neural Engineering, University of Technology, Stremayrgasse 16/IV, Graz, 8010, Austria.
Med Biol Eng Comput. 2019 Nov;57(11):2347-2357. doi: 10.1007/s11517-019-02047-1. Epub 2019 Sep 14.
For Brain-Computer interfaces (BCIs), system calibration is a lengthy but necessary process for successful operation. Co-adaptive BCIs aim to shorten training and imply positive motivation to users by presenting feedback already at early stages: After just 5 min of gathering calibration data, the systems are able to provide feedback and engage users in a mutual learning process. In this work, we investigate whether the retraining stage of co-adaptive BCIs can be adapted to a semi-supervised concept, where only a small amount of labeled data is available and all additional data needs to be labeled by the BCI itself. The aim of the current work was to evaluate whether a semi-supervised co-adaptive BCI could successfully compete with a supervised co-adaptive BCI model. In a supporting two-class (190 trials per condition) BCI study based on motor imagery tasks, we evaluated both approaches in two separate groups of 10 participants online, while we simulated the other approach in each group offline. Our results indicate that despite the lack of true labeled data, the semi-supervised driven BCI did not perform significantly worse (p > 0.05) than the supervised counterpart. We believe that these findings contribute to developing BCIs for long-term use, where continuous adaptation becomes imperative for maintaining meaningful BCI performance. Graphical abstract In this work, we investigate whether the retraining stage of a co-adaptive BCI can be adapted to a semi-supervised concept, where only a small amount of labeled data is available and all additional data needs to be labeled by the BCI itself. In two groups of 10 persons, we evaluate a supervised as well as a semi-supervised approach. Our results indicate that despite the lack of true labeled data, the semi-supervised driven BCI did not perform significantly worse (p > 0.05) than the supervised counterpart.
对于脑机接口 (BCI) 来说，系统校准是成功运行的一个漫长但必要的过程。共适应 BCI 的目的是通过在早期阶段提供反馈来缩短训练时间并对用户产生积极的激励：在仅收集 5 分钟校准数据后，系统就能够提供反馈并让用户参与到一个共同学习的过程中。在这项工作中，我们研究了共适应 BCI 的再训练阶段是否可以适应半监督概念，即只有少量标记数据可用，并且所有其他数据都需要由 BCI 本身进行标记。目前工作的目的是评估半监督共适应 BCI 是否可以成功与监督共适应 BCI 模型竞争。在一项基于运动想象任务的支持两分类（每个条件 190 次试验）BCI 研究中，我们在两个独立的 10 名参与者在线组中分别评估了这两种方法，同时在每个组中模拟了另一种方法。我们的结果表明，尽管缺乏真实标记数据，半监督驱动的 BCI 并没有表现出明显的性能下降（p>0.05）与监督对应物相比。我们相信这些发现有助于开发用于长期使用的 BCI，其中持续的适应对于维持有意义的 BCI 性能至关重要。
图摘要 在这项工作中，我们研究了共适应 BCI 的再训练阶段是否可以适应半监督概念，即只有少量标记数据可用，并且所有其他数据都需要由 BCI 本身进行标记。我们在两组 10 人参与者中评估了监督和半监督方法。我们的结果表明，尽管缺乏真实标记数据，半监督驱动的 BCI 并没有表现出明显的性能下降（p>0.05）与监督对应物相比。
