SOUL: An Energy-Efficient Unsupervised Online Learning Seizure Detection Classifier
SOUL: An Energy-Efficient Unsupervised Online Learning Seizure Detection Classifier
复制标题
DOI：
10.1109/jssc.2022.3172231
复制
发表时间：
2021-10
影响因子：
5.4
通讯作者：
A. Chua;M. I. Jordan;R. Muller
中科院分区：
文献类型：
--
作者：
A. Chua;M. I. Jordan;R. Muller
记录神经活动和检测癫痫发作的植入式设备已被用于发出警告或触发神经刺激以抑制癫痫发作。典型的癫痫检测系统依赖于高精度离线训练的机器学习分类器，当癫痫发作模式长时间变化时，需要人工重新训练。对于植入式癫痫检测系统，可以采用低功耗、边缘在线学习算法来动态适应神经信号漂移，从而在没有外界干预的情况下保持较高的精度。这项工作提出了SOUL：基于随机梯度下降的在线无监督逻辑回归分类器。在初始的离线训练阶段之后，连续的在线无监督分类器更新应用于原位，这提高了对漂移癫痫发作特征患者的敏感性。SOUL在两个人类脑电图（EEG）数据集上进行了测试：波士顿儿童医院和麻省理工学院（CHB-MIT）头皮脑电图数据集和长($\！>$ 100 h)颅内脑电图数据集。对于两个数据集，它能够分别达到97.5%和97.9%的平均灵敏度，特异性为bb0 95%。与典型的癫痫检测分类器相比，灵敏度在长期数据上最多提高8.2%。SOUL采用台湾积电（TSMC） 28纳米工艺制造，占地0.1 mm2，能效达到1.5 nJ/分类，比目前最先进的能效至少提高24倍。
Implantable devices that record neural activity and detect seizures have been adopted to issue warnings or trigger neurostimulation to suppress epileptic seizures. Typical seizure detection systems rely on high-accuracy offline-trained machine learning classifiers that require manual retraining when seizure patterns change over long periods of time. For an implantable seizure detection system, a low-power, at-the-edge, online learning algorithm can be employed to dynamically adapt to the neural signal drifts, thereby maintaining high accuracy without external intervention. This work proposes SOUL: Stochastic-gradient-descent-based Online Unsupervised Logistic regression classifier. After an initial offline training phase, continuous online unsupervised classifier updates are applied in situ, which improves sensitivity in patients with drifting seizure features. SOUL was tested on two human electroencephalography (EEG) datasets: the Children’s Hospital Boston and the Massachusetts Institute of Technology (CHB-MIT) scalp EEG dataset and a long ( $\!>$ 100 h) intracranial EEG dataset. It was able to achieve an average sensitivity of 97.5% and 97.9% for the two datasets, respectively, at >95% specificity. Sensitivity improved by at most 8.2% on long-term data when compared to a typical seizure detection classifier. SOUL was fabricated in Taiwan Semiconductor Manufacturing Company (TSMC’s) 28 nm process occupying 0.1 mm2 and achieves 1.5 nJ/classification energy efficiency, which is at least $24\times $ more efficient than state-of-the-art.