Zhang Zheng, Li Xin, Geng Fengji, Huang Kejie
Annu Int Conf IEEE Eng Med Biol Soc. 2021 Nov;2021:600-603. doi: 10.1109/EMBC46164.2021.9630363.
In the past decade, the rapid development of machine learning has dramatically improved the performance of epileptic detection with Electroencephalography (EEG). However, only a small amount of labeled epileptic data is available for training because labeling requires numerous neurologists. This paper proposes a one-step semi-supervised epilepsy detection system to reduce the labeling cost by fully utilizing the unlabeled data. The proposed neural network training strategy enables a more robust and accurate decision boundary by forcing the consistency of the double predictions on the same unlabeled data. The results show that the Area Under Receiver Operating Characteristic (AUROC) curves of our proposed model are 10.3% and 4.9% higher than the supervised methods on CHB-MIT and Kaggle datasets, respectively.
在过去十年中，机器学习的快速发展极大地提高了利用脑电图（EEG）进行癫痫检测的性能。然而，由于标记需要众多神经科医生，只有少量带标记的癫痫数据可用于训练。本文提出了一种一步半监督癫痫检测系统，以通过充分利用未标记数据来降低标记成本。所提出的神经网络训练策略通过强制对同一未标记数据的双重预测保持一致，从而实现更稳健、准确的决策边界。结果表明，我们所提出模型的受试者工作特征曲线下面积（AUROC）在CHB - MIT和Kaggle数据集上分别比监督方法高10.3%和4.9%。
