Deep Learning With Convolutional Neural Networks for Motor Brain-Computer Interfaces Based on Stereo-Electroencephalography (SEEG)
Deep Learning With Convolutional Neural Networks for Motor Brain-Computer Interfaces Based on Stereo-Electroencephalography (SEEG)
复制标题
DOI：
10.1109/jbhi.2023.3242262
复制
发表时间：
2023-02
影响因子：
7.7
通讯作者：
Xiaolong Wu;Shize Jiang;Guangye Li;Shengjie Liu;B. Metcalfe;Liang Chen;Dingguo Zhang
中科院分区：
文献类型：
--
作者：
Xiaolong Wu;Shize Jiang;Guangye Li;Shengjie Liu;B. Metcalfe;Liang Chen;Dingguo Zhang
目的：基于卷积神经网络（CNN）的深度学习在脑机接口（bci）的头皮脑电图（EEG）中取得成功。然而，所谓的“黑匣子”方法的解释及其在基于立体脑电图（SEEG）的脑机接口中的应用在很大程度上仍然未知。因此，本文对深度学习方法对SEEG信号的解码性能进行了评估。方法：招募30例癫痫患者，设计5种手部和前臂运动模式。使用6种方法，包括滤波器组公共空间模式（FBCSP）和5种深度学习方法（EEGNet、浅层和深层CNN、ResNet和一种名为STSCNN的深层CNN变体）对SEEG数据进行分类。通过各种实验研究了窗口、模型结构以及ResNet和STSCNN解码过程的影响。结果：EEGNet、FBCSP、shallow CNN、deep CNN、STSCNN和ResNet的平均分类准确率分别为35 $\pm$ 6.1%、38 $\pm$ 4.9%、60 $\pm$ 3.9%、60 $\pm$ 3.3%、61 $\pm$ 3.2%和63 $\pm$ 3.1%。进一步分析表明，该方法在谱域中具有明显的可分性。结论：ResNet和STSCNN分别取得了第一和第二高的解码精度。STSCNN表明，额外的空间卷积层是有益的，并且可以从空间和光谱角度部分解释解码过程。意义：本研究首次探讨了深度学习在SEEG信号上的性能。此外，本文还证明了所谓的“黑盒”方法可以部分解释。
Objective: Deep learning based on convolutional neural networks (CNN) has achieved success in brain-computer interfaces (BCIs) using scalp electroencephalography (EEG). However, the interpretation of the so-called ‘black box’ method and its application in stereo-electroencephalography (SEEG)-based BCIs remain largely unknown. Therefore, in this paper, an evaluation is performed on the decoding performance of deep learning methods on SEEG signals. Methods: Thirty epilepsy patients were recruited, and a paradigm including five hand and forearm motion types was designed. Six methods, including filter bank common spatial pattern (FBCSP) and five deep learning methods (EEGNet, shallow and deep CNN, ResNet, and a deep CNN variant named STSCNN), were used to classify the SEEG data. Various experiments were conducted to investigate the effect of windowing, model structure, and the decoding process of ResNet and STSCNN. Results: The average classification accuracy for EEGNet, FBCSP, shallow CNN, deep CNN, STSCNN, and ResNet were 35 $\pm$ 6.1%, 38 $\pm$ 4.9%, 60 $\pm$ 3.9%, 60 $\pm$ 3.3%, 61 $\pm$ 3.2%, and 63 $\pm$ 3.1% respectively. Further analysis of the proposed method demonstrated clear separability between different classes in the spectral domain. Conclusion: ResNet and STSCNN achieved the first- and second-highest decoding accuracy, respectively. The STSCNN demonstrated that an extra spatial convolution layer was beneficial, and the decoding process can be partially interpreted from spatial and spectral perspectives. Significance: This study is the first to investigate the performance of deep learning on SEEG signals. In addition, this paper demonstrated that the so-called ‘black-box’ method can be partially interpreted.