A Real-Time Depth of Anesthesia Monitoring System Based on Deep Neural Network With Large EDO Tolerant EEG Analog Front-End
A Real-Time Depth of Anesthesia Monitoring System Based on Deep Neural Network With Large EDO Tolerant EEG Analog Front-End
复制标题
基于深度神经网络的大EDO容限脑电模拟前端麻醉深度实时监测系统
DOI：
10.1109/tbcas.2020.2998172
复制
发表时间：
2020-08-01
影响因子：
5.1
通讯作者：
Kim, Seong-Jin
中科院分区：
文献类型：
--
作者：
Park, Yongjae;Han, Su-Hyun;Kim, Seong-Jin
在本文中，我们结合深度学习框架 AnesNET 提出了一种基于实时脑电图 (EEG) 的麻醉深度 (DoA) 监测系统。在所提出的系统中实现了 EEG 模拟前端 (AFE)，它可以使用粗略数字直流伺服环路补偿 +/- 380 mV 电极直流偏移。基于脑电图的 MAC（EEGMAC）是一种准确预测 DoA 的新指标，旨在应用于使用挥发性麻醉剂和静脉麻醉剂麻醉的患者。所提出的深度学习协议由四层卷积神经网络和两个密集层组成。此外，我们优化了深度神经网络（DNN）的复杂性，以在Raspberry Pi 3等微型计算机上运行，实现了经济高效的小型DoA监控系统。原型 AFE 采用 110 nm CMOS 制造，每通道功耗为 4.33 μW，0.5 至 100 Hz 范围内的输入参考噪声为 0.29 μ Vrms，噪声效率因数为 2.2。所提出的 DNN 使用来自 374 名在手术中接受吸入麻醉剂的受试者的预先记录的脑电图数据进行了评估，平均平方误差和绝对误差分别为 0.048 和 0.05。通过静脉注射麻醉剂进行麻醉的受试者的 EEGMAC 也显示出与脑电双频指数值良好的一致性，证实了所提出的 DoA 指数适用于两种麻醉剂。使用 Raspberry Pi 3 实现的监测系统可在 20 毫秒内估计 EEGMAC，这比文献中的 BIS 估计快了约千倍。
In this article, we present a real-time electroencephalogram (EEG) based depth of anesthesia (DoA) monitoring system in conjunction with a deep learning framework, AnesNET. An EEG analog front-end (AFE) that can compensate +/- 380-mV electrode DC offset using a coarse digital DC servo loop is implemented in the proposed system. The EEG-based MAC, EEGMAC, is introduced as a novel index to accurately predict the DoA, which is designed for applying to patients anesthetized by both volatile and intravenous agents. The proposed deep learning protocol consists of four layers of convolutional neural network and two dense layers. In addition, we optimize the complexity of the deep neural network (DNN) to operate on a microcomputer such as the Raspberry Pi 3, realizing a cost-effective small-size DoA monitoring system. Fabricated in 110-nm CMOS, the prototype AFE consumes 4.33 mu W per channel and has the input-referred noise of 0.29 mu Vrms from 0.5 to 100 Hz with the noise efficiency factor of 2.2. The proposed DNN was evaluated with pre-recorded EEG data from 374 subjects administrated by inhalational anesthetics under surgery, achieving an average squared and absolute errors of 0.048 and 0.05, respectively. The EEGMAC with subjects anesthetized by an intravenous agent also showed a good agreement with the bispectral index value, confirming the proposed DoA index is applicable to both anesthetics. The implemented monitoring system with the Raspberry Pi 3 estimates the EEGMAC within 20 ms, which is about thousand-fold faster than the BIS estimation in literature.