A machine learning approach to seizure detection in a rat model of post-traumatic epilepsy.

A machine learning approach to seizure detection in a rat model of post-traumatic epilepsy.
复制标题

在创伤后癫痫的大鼠模型中,一种机器学习方法。

DOI:
10.1038/s41598-023-40628-1
复制
发表时间:
2023-09-22
期刊:
影响因子:
4.6
通讯作者:
--
中科院分区:
综合性期刊3区
文献类型:
--
作者:

文献摘要

参考文献

相似文献

癫痫是一种常见的神经系统疾病,经常使用啮齿动物模型进行研究,通过脑电图 (EEG) 识别癫痫发作。鉴于技术进步,脑电图的大型数据集已广泛使用,并且适合机器学习方法来识别癫痫发作。虽然已经针对人类脑电图探索了这种方法,但识别啮齿动物脑电图癫痫发作的机器学习方法是有限的。我们利用预先设计的深度卷积神经网络 (DCNN) GoogLeNet 对图像进行分类以进行癫痫发作识别。通过复用两秒脑电图片段的频谱内容(尺度图)、峰度和熵来生成训练图像。对超过 2200 小时的脑电图数据进行了癫痫发作的评分,与视觉评分相比,DCNN 识别出 95.6% 的癫痫发作,假阳性率为 34.2%(1.52/小时)。多重图像优于单独的尺度图(尺度图-峰度-熵 0.956 ± 0.010,尺度图 0.890 ± 0.028,t(7) = 3.54,p < 0.01),并且专门针对个体动物训练的 DCNN 优于在动物之间使用 DCNN (动物内 0.960 ± 0.0094,动物间 0.811 ± 0.015,t(30) = 5.54,p < 0.01)。对于该数据集,DCNN 方法优于先前描述的利用较长局部线长度(根据 EEG 小波分解计算)来识别癫痫发作的算法。我们展示了预先设计的 DCNN 的新颖用途,该 DCNN 被构建用于对图像进行分类,利用脑电图频谱内容、峰度和熵的多重图像,以高灵敏度快速、客观地识别大鼠脑电图大型数据集中的癫痫发作。
Epilepsy is a common neurologic condition frequently investigated using rodent models, with seizures identified by electroencephalography (EEG). Given technological advances, large datasets of EEG are widespread and amenable to machine learning approaches for identification of seizures. While such approaches have been explored for human EEGs, machine learning approaches to identifying seizures in rodent EEG are limited. We utilized a predesigned deep convolutional neural network (DCNN), GoogLeNet, to classify images for seizure identification. Training images were generated through multiplexing spectral content (scalograms), kurtosis, and entropy for two-second EEG segments. Over 2200 h of EEG data were scored for the presence of seizures, with 95.6% of seizures identified by the DCNN and a false positive rate of 34.2% (1.52/h), as compared to visual scoring. Multiplexed images were superior to scalograms alone (scalogram-kurtosis-entropy 0.956 ± 0.010, scalogram 0.890 ± 0.028, t(7) = 3.54, p < 0.01) and a DCNN trained specifically for the individual animal was superior to using DCNNs across animals (intra-animal 0.960 ± 0.0094, inter-animal 0.811 ± 0.015, t(30) = 5.54, p < 0.01). For this dataset the DCNN approach is superior to a previously described algorithm utilizing longer local line lengths, calculated from wavelet-decomposition of EEG, to identify seizures. We demonstrate the novel use of a predesigned DCNN constructed to classify images, utilizing multiplexed images of EEG spectral content, kurtosis, and entropy, to rapidly and objectively identifies seizures in a large dataset of rat EEG with high sensitivity.
DOI: 10.1097/00004691-200302000-00005
发表时间: 2003-02-01
影响因子: 2.4
作者:
Benbadis, SR;Tatum, WO
通讯作者: Tatum, WO
DOI: 10.1212/wnl.0b013e3181e8e6d0
发表时间: 2010-07-20
期刊: NEUROLOGY
影响因子: 9.9
作者:
Raymont, V.;Salazar, A. M.;Grafman, J.
通讯作者: Grafman, J.
DOI: 10.3389/fninf.2018.00095
发表时间: 2018-12-10
影响因子: 3.5
作者:
Zhou, Mengni;Tian, Cheng;Xiang, Jie
通讯作者: Xiang, Jie
DOI: 10.1186/s40708-020-00105-1
发表时间: 2020-05-25
期刊: Brain informatics
影响因子: --
作者:
Siddiqui, Mohammad Khubeb;Morales-Menendez, Ruben;Hussain, Nasir
通讯作者: Hussain, Nasir
DOI: 10.1016/j.nbd.2015.10.008
发表时间: 2016-01
影响因子: 6.1
作者:
Langberg, Tomer;Dashek, Ryan;Mulvey, Bernard;Miller, Kimberly A.;Osting, Susan;Stafstrom, Carl E.;Sutula, Thomas P.
通讯作者: Sutula, Thomas P.