Jacobs Daniel, Liu Yuhan H, Hilton Trevor, Del Campo Martin, Carlen Peter L, Bardakjian Berj L
1Institute of Biomaterials and Biomedical Engineering, University of TorontoTorontoONM5S 3G9Canada.
2Department of Electrical and Computer EngineeringUniversity of TorontoTorontoONM5S 3G9Canada.
IEEE J Transl Eng Health Med. 2019 Aug 16;7:2000203. doi: 10.1109/JTEHM.2019.2926257. eCollection 2019.
To investigate the feasibility of improving the performance of an EEG-based multistate classifier (MSC) previously proposed by our group.
Using the random forest (RF) classifiers on the previously reported dataset of patients, but with three improvements to classification logic, the specificity of our alarm algorithm improves from 82.4% to 92.0%, and sensitivity from 87.9% to 95.2%.
The MSC could be a useful approach for seizure-monitoring both in the clinic and at home.
Three improvements to the MSC are described. Firstly, an additional check using RF outputs is made prior to alarm to confirm increasing probability of a seizure onset state. Secondly, a post-alarm detection horizon that accounts for the seizure state duration is implemented. Thirdly, the alarm decision window is kept constant.
研究改进我们小组之前提出的基于脑电图的多状态分类器（MSC）性能的可行性。
在先前报告的患者数据集上使用随机森林（RF）分类器，但对分类逻辑进行了三项改进，我们的警报算法的特异性从82.4%提高到92.0%，敏感性从87.9%提高到95.2%。
MSC可能是一种在临床和家庭中进行癫痫发作监测的有用方法。
描述了对MSC的三项改进。首先，在发出警报之前使用RF输出进行额外检查，以确认癫痫发作起始状态的概率增加。其次，实施了一个考虑癫痫发作状态持续时间的警报后检测范围。第三，警报决策窗口保持不变。
