EmotionO+: Physiological signals knowledge representation and emotion reasoning model for mental health monitoring
EmotionO+: Physiological signals knowledge representation and emotion reasoning model for mental health monitoring
复制标题
DOI：
10.1109/bibm.2014.6999215
复制
发表时间：
2014-11
期刊：
影响因子：
--
通讯作者：
Yun Su;Bin Hu;Lixin Xu;Hanshu Cai;P. Moore;Xiaowei Zhang;Jing Chen
中科院分区：
文献类型：
--
作者：
Yun Su;Bin Hu;Lixin Xu;Hanshu Cai;P. Moore;Xiaowei Zhang;Jing Chen
情绪是抑郁状态的一个重要指标。基于脑电信号和功能近红外光谱等生理信号的情感识别在医疗领域的研究中具有重要意义。不同的医疗保健系统之间的情绪反应相关的生理信号数据的共享有可能有利于实验室为基础的医疗保健研究和“现实世界”的临床实践。然而，数据的管理和分发提出了重大挑战;应对这些挑战需要先进的数据表示、挖掘和集成工具。在本文中，我们提出了这样一个工具，它包含一个本体模型称为abstentionO+和规则集的基础上，这是由随机森林算法获得的情绪状态预测。它不仅提出了一种有效的方法，使语义表示的EEG和fNIRS数据，但也是一个情感知识挖掘工具。在eNTERFACE'06数据集中使用EEG数据的结果显示，我们提出的模型的准确性为99.11%，而使用C4.5算法的竞争方法为97.8%。实验结果表明，假设的方法是潜在的可用于早期预测和干预抑郁症。
Emotion is an important indicator of depressive conditions. Emotion recognition based on physiological signals such as electroencephalogram (EEG) and functional near-infrared spectroscopy (fNIRS) has gained significant attraction in healthcare domain research. Sharing of physiological signal data related to emotional response between different healthcare systems has the potential to benefit both laboratory-based healthcare research and `real-world' clinical practice. However, management and distribution of the data presents significant challenges; addressing these challenges requires advanced tools for data representation, mining and integration. In this paper we propose such a tool which contains an ontology model called EmotionO+ and rules set based on EEG, which is obtained by random forest algorithm to predict emotional state. It presents not only an effective method to enable semantic representation of the EEG and fNIRS data, but also an emotion knowledge mining tool. Results using EEG data in the eNTERFACE'06 dataset show an accuracy for our proposed model of 99.11% as compared to 97.8% for competing methods using the C4.5 algorithm. The experimental results demonstrate that the posited approach is potentially usable for early stage prediction and intervention for depressive disorders.