Automated sleep state classification of wide-field calcium imaging data via multiplex visibility graphs and deep learning.
Automated sleep state classification of wide-field calcium imaging data via multiplex visibility graphs and deep learning.
复制标题
DOI：
10.1016/j.jneumeth.2021.109421
复制
发表时间：
2022-01-15
影响因子：
3
通讯作者：
Anastasio, Mark A.
中科院分区：
文献类型：
--
作者：
Zhang, Xiaohui;Landsness, Eric C.;Chen, Wei;Miao, Hanyang;Tang, Michelle;Brier, Lindsey M.;Culver, Joseph P.;Lee, Jin-Moo;Anastasio, Mark A.
关键词：
广域钙成像(WFCI)可以监测小鼠大脑皮质的神经动力学。当应用于睡眠研究时，WFCI数据通过使用辅助脑电和肌电记录被手动评分为清醒、非快速眼动(NREM)和快速眼动睡眠状态。然而，这一过程非常耗时，而且往往存在评分者之间和评分者内部的可靠性和侵入性较低的问题。因此，需要一种仅对WFCI数据进行操作的自动睡眠状态分类方法。提出了一种混合两步法。在第一步中，将时空WFCI数据映射到多路可见图(MVG)。随后，将二维卷积神经网络(2DCNN)应用于MVG，将其分为觉醒、NREM和REM。睡眠状态分类的准确率为84%，科恩的κ为0.67。该方法还有效地应用于觉醒/睡眠二分类(精度=0.82，κ=0.6 2)和觉醒/睡眠/麻醉/运动四级分类(精度=0.74，κ=0.6 6)。梯度加权类激活图显示，CNN以睡眠状态特定的方式关注MVG的短期和长期时间连接。当使用单独的大脑区域时，大脑皮层后部区域的睡眠状态分类性能最高，当考虑到整个大脑皮层的活动时。在3小时的WFCI记录中，MVG-CNN的κ为0.65，与基于人类脑电/肌电的评分的κ为0.60相当。MVG-CNN混合方法从WFCI数据中准确地对睡眠状态进行分类，并将使未来使用WFCI进行睡眠重点研究成为可能。
Wide-field calcium imaging (WFCI) allows for monitoring of cortex-wide neural dynamics in mice. When applied to the study of sleep, WFCI data are manually scored into the sleep states of wakefulness, non-REM (NREM) and REM by use of adjunct EEG and EMG recordings. However, this process is time-consuming and often suffers from low inter- and intra-rater reliability and invasiveness. Therefore, an automated sleep state classification method that operates on WFCI data alone is needed. A hybrid, two-step method is proposed. In the first step, spatial-temporal WFCI data is mapped to multiplex visibility graphs (MVGs). Subsequently, a two-dimensional convolutional neural network (2D CNN) is employed on the MVGs to be classified as wakefulness, NREM and REM. Sleep states were classified with an accuracy of 84% and Cohen’s κ of 0.67. The method was also effectively applied on a binary classification of wakefulness/sleep (accuracy=0.82, κ = 0.62) and a four-class wakefulness/sleep/anesthesia/movement classification (accuracy=0.74, κ = 0.66). Gradient-weighted class activation maps revealed that the CNN focused on short- and long-term temporal connections of MVGs in a sleep state-specific manner. Sleep state classification performance when using individual brain regions was highest for the posterior area of the cortex and when cortex-wide activity was considered. On a 3-hour WFCI recording, the MVG-CNN achieved a κ of 0.65, comparable to a κ of 0.60 corresponding to the human EEG/EMG-based scoring. The hybrid MVG-CNN method accurately classifies sleep states from WFCI data and will enable future sleep-focused studies with WFCI.
登录
查看更多内容
影响因子：
2.9
作者：
Brankack, Jurij;Kukushka, Valeriy I.;Draguhn, Andreas
通讯作者：
Draguhn, Andreas
影响因子：
4.8
作者：
Collop, Nancy A.
通讯作者：
Collop, Nancy A.
影响因子：
5.6
作者：
Dodet, Pauline;Chavez, Mario;Arnulf, Isabelle
通讯作者：
Arnulf, Isabelle
影响因子：
3.2
作者：
Gaiduk, Maksym;Penzel, Thomas;Seepold, Ralf
通讯作者：
Seepold, Ralf
影响因子：
37.8
作者：
Goldberger, AL;Amaral, LAN;Stanley, HE
通讯作者：
Stanley, HE