Saba-Sadiya Sari, Chantland Eric, Alhanai Tuka, Liu Taosheng, Ghassemi Mohammad M
Human Augmentation and Artificial Intelligence Lab, Department of Computer Science, Michigan State University, East Lansing, MI, United States.
Neuroimaging of Perception and Attention Lab, Department of Psychology, Michigan State University, East Lansing, MI, United States.
Front Digit Health. 2021 Jan 22;2:608920. doi: 10.3389/fdgth.2020.608920. eCollection 2020.
Electroencephalography (EEG) is used in the diagnosis, monitoring, and prognostication of many neurological ailments including seizure, coma, sleep disorders, brain injury, and behavioral abnormalities. One of the primary challenges of EEG data is its sensitivity to a breadth of non-stationary noises caused by physiological-, movement-, and equipment-related artifacts. Existing solutions to artifact are deficient because they require experts to manually explore and annotate data for artifact segments. Existing solutions to artifact or removal are deficient because they assume that the incidence and specific characteristics of artifacts are similar across both subjects and tasks (i.e., "one-size-fits-all"). In this paper, we describe a novel EEG noise-reduction method that uses representation learning to perform patient- and task-specific artifact detection and correction. More specifically, our method extracts 58 clinically relevant features and applies an ensemble of unsupervised outlier detection algorithms to identify EEG artifacts that are unique to a given task and subject. The artifact segments are then passed to a deep encoder-decoder network for unsupervised . We compared the performance of classification models trained with and without our method and observed a 10% relative improvement in performance when using our approach. Our method provides a flexible end-to-end unsupervised framework that can be applied to novel EEG data without the need for expert supervision and can be used for a variety of clinical decision tasks, including coma prognostication and degenerative illness detection. By making our method, code, and data publicly available, our work provides a tool that is of both immediate practical utility and may also serve as an important foundation for future efforts in this domain.
脑电图（EEG）用于多种神经系统疾病的诊断、监测和预后评估，包括癫痫、昏迷、睡眠障碍、脑损伤和行为异常。EEG数据的主要挑战之一是其对由生理、运动和设备相关伪迹引起的广泛非平稳噪声敏感。现有的伪迹解决方案存在缺陷，因为它们需要专家手动探索和注释数据中的伪迹段。现有的伪迹检测或去除解决方案存在缺陷，因为它们假设伪迹的发生率和特定特征在不同受试者和任务中相似（即“一刀切”）。在本文中，我们描述了一种新颖的EEG降噪方法，该方法使用表示学习来执行针对患者和任务的伪迹检测和校正。更具体地说，我们的方法提取58个临床相关特征，并应用一组无监督异常检测算法来识别特定任务和受试者独有的EEG伪迹。然后将伪迹段传递给深度编码器 - 解码器网络进行无监督处理。我们比较了使用和不使用我们的方法训练的分类模型的性能，发现使用我们的方法时性能相对提高了10%。我们的方法提供了一个灵活的端到端无监督框架，可应用于新的EEG数据，无需专家监督，可用于各种临床决策任务，包括昏迷预后评估和退行性疾病检测。通过公开我们的方法、代码和数据，我们的工作提供了一个既具有直接实用价值，又可能为该领域未来的努力奠定重要基础的工具。
