Combining speech sample and feature bilateral selection algorithm for classification of Parkinson's disease.
10.7507/1001-5515.201704061
- Author:
Xiaoheng ZHANG
1
;
Lirui WANG
2
;
Yao CAO
2
;
Pin WANG
2
;
Cheng ZHANG
2
;
Liuyang YANG
2
;
Yongming LI
3
;
Yanling ZHANG
4
;
Oumei CHENG
5
Author Information
1. Chongqing Radio & TV University, Chongqing 400052, P.R.China.
2. College of Communication Engineering, Chongqing University, Chongqing 400030,P.R.China.
3. College of Communication Engineering, Chongqing University, Chongqing 400030,P.R.China.yongmingli@cqu.edu.cn.
4. Department of Neurology, Southwest hospital, Third military medical university, Chongqing 400038, P.R.China.
5. Department of Neurology, The first affiliated hospital, Chongqing Medical University, Chongqing 400038, P.R.China.
- Publication Type:Journal Article
- Keywords:
Parkinson’s disease;
bilateral hybrid speech feature selection;
classification;
multiple kernel learning;
synergy effects
- From:
Journal of Biomedical Engineering
2018;34(6):942-948
- CountryChina
- Language:Chinese
-
Abstract:
Diagnosis of Parkinson's disease (PD) based on speech data has been proved to be an effective way in recent years. However, current researches just care about the feature extraction and classifier design, and do not consider the instance selection. Former research by authors showed that the instance selection can lead to improvement on classification accuracy. However, no attention is paid on the relationship between speech sample and feature until now. Therefore, a new diagnosis algorithm of PD is proposed in this paper by simultaneously selecting speech sample and feature based on relevant feature weighting algorithm and multiple kernel method, so as to find their synergy effects, thereby improving classification accuracy. Experimental results showed that this proposed algorithm obtained apparent improvement on classification accuracy. It can obtain mean classification accuracy of 82.5%, which was 30.5% higher than the relevant algorithm. Besides, the proposed algorithm detected the synergy effects of speech sample and feature, which is valuable for speech marker extraction.
