Multi-View Intact Space Learning for Tinnitus Classification in Resting State EEG
- 115 Downloads
Tinnitus is a common but obscure auditory disease to be studied, and there are still in the lack of effective methods developed to treat tinnitus universally. Although electroencephalogram (EEG) is widely applied to the diagnosis of tinnitus, there are few machine learning methods developed to classify tinnitus patients from healthy people based on the EEG data. Moreover, there is still room for improving the classification performance due to the insufficient existing studies. Therefore, in order to improve the performance of classification based on the EEG data, we introduce a multi-view intact space learning method to characterize the EEG signals by feature extraction in a latent intact space. Considering the fact that there are only a small number of subjects available for study, we conduct the classification for valid segments of EEG data of each subject. In this way, the dataset can be enlarged and the classification performance can be improved. By combining different views of EEG data, a considerable result is achieved on classification by using Support Vector Machine classifier, with accuracy, recall, precision, F1 to be 99.23, 99.72, 98.97, 99.34% respectively. This proposed method is an effective and objective method to classify the tinnitus patients from healthy people, further researches are needed to explore the machine learning method in classification and prediction of the effectiveness of tinnitus interventions based on the EEG response of tinnitus individuals.
KeywordsTinnitus Classification Multi-view learning
We would like to acknowledge Dr. Fei Zhao for his proof reading.
This work was funded by National Natural Science Foundation of China (Grant No. 81600808), Natural Science Foundation of Guangdong Province (Grant No. 2016A030313318 and 2015A030310134), Guangdong Natural Science Funds for Distinguished Young Scholar (2016A030306014) and Tip-top Scientific and Technical Innovative Youth Talents of Guangdong special support program (2016TQ03X542).
Compliance with Ethical Standards
Conflict of interest
The authors declared that they have no conflict of interest.
- 5.Emami Y, Bayrak C (2017) EEG analysis of evoked potentials of the brain to develop a mathematical model for classifying tinnitus datasets. In: 2017 IEEE international symposium on medical measurements and applications (MeMeA), IEEE, pp 379–384Google Scholar
- 6.Han J, Chen H, Liu N, Yan C, Li X (2017) CNNs-based rgb-d saliency detection via cross-view transfer and multiview fusion. IEEE Trans Cybern 99:1–13Google Scholar
- 12.Huang L, Chao HY, Wang CD (2017) Multi-view intact space clustering. In: Proceedings of the 4th Asian conference on pattern recognition, pp 500–505Google Scholar
- 15.Li PZ, Cai YX, Wang CD, Liang MJ, Zheng YQ (2018) Higher-order brain network analysis for auditory disease. Neural Process Lett. https://doi.org/10.1007/s11063-018-9815-7
- 16.Li PZ, Li JH, Wang CD (2016) A SVM-based EEG signal analysis: an auxiliary therapy for tinnitus. In: Proceedings of advances in brain inspired cognitive systems: 8th international conference, BICS 2016, Beijing, China, November 28–30, 2016, vol 8. Springer, pp 207–219Google Scholar
- 18.Lin KY, Huang L, Wang CD, Chao HY (2018) Multi-view proximity learning for clustering. In: Proceedings of the 23rd international conference on database systems for advanced applicationsGoogle Scholar
- 19.Lin KY, Wang CD, Meng YQ, Zhao ZL (2017) Multi-view unit intact space learning. In: 10th international conference on knowledge science, engineering and management, pp 211–223Google Scholar
- 20.Manzano M, Guillén A, Rojas I, Herrera LJ (2017) Deep learning using EEG data in time and frequency domains for sleep stage classification. In: International work-conference on artificial neural networks. Springer, pp 132–141Google Scholar
- 21.Meyer M, Luethi MS, Neff P, Langer N, Büchi S (2014) Disentangling tinnitus distress and tinnitus presence by means of EEG power analysis. Neural Plast 2014:1–13Google Scholar
- 29.Wang SJ, Cai YX, Sun ZR, Wang CD, Zheng YQ (2017) Tinnitus EEG classification based on multi-frequency bands. In: International conference on neural information processing. Springer, pp 788–797Google Scholar
- 32.Xu H, Plataniotis KN (2016) Affective states classification using EEG and semi-supervised deep learning approaches. In: 2016 IEEE 18th international workshop on multimedia signal processing (MMSP). IEEE, pp 1–6Google Scholar
- 33.Xu L, Wang CD, Liang MJ, Cai YX, Zheng YQ (2018) Brain network regional synchrony analysis in deafness. Biomed Res Int. https://doi.org/10.1155/2018/6547848
- 36.Yildirim A, Halici U (2013) Analysis of dimension reduction by PCA and adaboost on spelling paradigm EEG data. In: 2013 6th international conference on biomedical engineering and informatics (BMEI). IEEE, pp 192–196Google Scholar
- 39.Zhang GY, Wang CD, Huang D, Zheng WS, Zhou YR (2018) TW-Co-k-means: two-level weighted collaborative k-means for multi-view clustering. Knowl Based Syst. https://doi.org/10.1016/j.knosys.2018.03.009