School of Information Science and Technology, University of Science and Technology of China, Hefei 230027, China.
Epilepsy Center, Department of Neurosurgery, The First Affiliated Hospital of USTC, Division of Life Sciences and Medicine, University of Science and Technology of China, Hefei 230001, China.
J Neurosci Methods. 2022 Apr 15;372:109557. doi: 10.1016/j.jneumeth.2022.109557. Epub 2022 Mar 8.
Early prediction of epilepsy seizures can warn the patients to take precautions and improve their lives significantly. In recent years, deep learning has become increasingly predominant in seizure prediction for its outstanding performance. With the aim of predicting unseen seizures, it is essential to guarantee the generalization ability of the model, especially considering the non-stationary nature of EEG and the scarcity of seizure events in EEG recordings. Stability training against extra perturbations is an intuitive and effective way to improve the model's ability to generalize. Though a great number of deep learning methods have been developed for seizure prediction, their strategies to increase generalization performance focus on improving the model's architecture itself, and few of them pay attention to the stability of the model against small perturbations.
In this study, we propose a novel consistency-based training strategy to address this issue. The proposed strategy underlines that a robust model should maintain consistent results for the same input under extra perturbations. Specifically, during training, we use stochastic augmentations to make the input vary from iteration to iteration and consider the output as a stochastic variable. Then a consistency constraint is constructed to penalize the difference between the current output and previous outputs. In this way, the generalization ability of the model will be fully enhanced.
To better verify the effectiveness of our proposed strategy, we implement it in two state-of-the-art models with public-available codes, including STFT CNN and Multi-view CNN. Notably, we compare with the first baseline on a scalp EEG dataset and the other on an intracranial EEG dataset. The results show that our strategy could improve the performance significantly for both of them.
Our strategy has increased the sensitivity by 7.1% and reduced the false prediction rate by 0.12/h on the first baseline while improving the AUC by 0.020 on the second baseline.
This study is easy to implement, providing a new solution to enhance the performance of seizure prediction.
早期预测癫痫发作可以警告患者采取预防措施，显著提高他们的生活质量。近年来，深度学习在癫痫预测中因其出色的表现而变得越来越重要。为了预测未见的癫痫发作，保证模型的泛化能力至关重要，特别是考虑到 EEG 的非平稳性和 EEG 记录中癫痫事件的稀缺性。针对额外扰动的稳定性训练是提高模型泛化能力的一种直观有效的方法。虽然已经开发了许多用于癫痫预测的深度学习方法，但它们提高泛化性能的策略主要集中在改进模型的架构本身，很少有方法关注模型对小扰动的稳定性。
在本研究中，我们提出了一种新的基于一致性的训练策略来解决这个问题。所提出的策略强调，一个稳健的模型应该在额外扰动下对相同的输入保持一致的结果。具体来说，在训练过程中，我们使用随机增强使输入在迭代中变化，并将输出视为随机变量。然后构建一致性约束来惩罚当前输出和以前输出之间的差异。通过这种方式，模型的泛化能力将得到充分增强。
为了更好地验证我们提出的策略的有效性，我们在两个具有公开可用代码的最先进模型中实现了它，包括 STFT CNN 和多视图 CNN。值得注意的是，我们在头皮 EEG 数据集上与第一个基线进行比较，在颅内 EEG 数据集上与另一个基线进行比较。结果表明，我们的策略可以显著提高两者的性能。
我们的策略在第一个基线中提高了 7.1%的敏感性，降低了 0.12/h 的假预测率，而在第二个基线中提高了 0.020 的 AUC。
本研究易于实施，为提高癫痫预测性能提供了新的解决方案。
Zotero 插件