Brain signals are important quantitative data for understanding physiological
activities and diseases of human brain. Most existing studies pay attention to
supervised learning methods, which, however, require high-cost clinical labels.
In addition, the huge difference in the clinical patterns of brain signals
measured by invasive (e.g., SEEG) and non-invasive (e.g., EEG) methods leads to
the lack of a unified method. To handle the above issues, we propose to study
the self-supervised learning (SSL) framework for brain signals that can be
applied to pre-train either SEEG or EEG data. Intuitively, brain signals,
generated by the firing of neurons, are transmitted among different connecting
structures in human brain. Inspired by this, we propose MBrain to learn
implicit spatial and temporal correlations between different channels (i.e.,
contacts of the electrode, corresponding to different brain areas) as the
cornerstone for uniformly modeling different types of brain signals.
Specifically, we represent the spatial correlation by a graph structure, which
is built with proposed multi-channel CPC. We theoretically prove that
optimizing the goal of multi-channel CPC can lead to a better predictive
representation and apply the instantaneou-time-shift prediction task based on
it. Then we capture the temporal correlation by designing the
delayed-time-shift prediction task. Finally, replace-discriminative-learning
task is proposed to preserve the characteristics of each channel. Extensive
experiments of seizure detection on both EEG and SEEG large-scale real-world
datasets demonstrate that our model outperforms several state-of-the-art time
series SSL and unsupervised models, and has the ability to be deployed to
clinical practice.

本文提出了 MBrain，一种用于自监督学习的框架，可以用于预训练 SEEG 或 EEG 数据，并在大规模现实世界数据上进行了癫痫检测实验，表明我们的模型优于现有的一些时间序列自监督和无监督模型，并具有在临床实践中被部署的能力。