Classification of Dysphonic Voices in Parkinson’s Disease with Semi-Supervised Competitive Learning Algorithm
This article proposes a novel semi-supervised competitive learning (SSCL) algorithm for vocal pattern classifications in Parkinson’s disease (PD). The acoustic parameters of voice records were grouped into the families of jitter, shimmer, harmonic-to-noise, frequency, and nonlinear measures, respect...
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2022-07-01
|
| Series: | Biosensors |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2079-6374/12/7/502 |
| _version_ | 1797433788255436800 |
|---|---|
| author | Guidong Bao Mengchen Lin Xiaoqian Sang Yangcan Hou Yixuan Liu Yunfeng Wu |
| author_facet | Guidong Bao Mengchen Lin Xiaoqian Sang Yangcan Hou Yixuan Liu Yunfeng Wu |
| author_sort | Guidong Bao |
| collection | DOAJ |
| description | This article proposes a novel semi-supervised competitive learning (SSCL) algorithm for vocal pattern classifications in Parkinson’s disease (PD). The acoustic parameters of voice records were grouped into the families of jitter, shimmer, harmonic-to-noise, frequency, and nonlinear measures, respectively. The linear correlations were computed within each acoustic parameter family. According to the correlation matrix results, the jitter, shimmer, and harmonic-to-noise parameters presented as highly correlated in terms of Pearson’s correlation coefficients. Then, the principal component analysis (PCA) technique was implemented to eliminate the redundant dimensions of the acoustic parameters for each family. The Mann–Whitney–Wilcoxon hypothesis test was used to evaluate the significant difference of the PCA-projected features between the healthy subjects and PD patients. Eight dominant PCA-projected features were selected based on the eigenvalue threshold criterion and the statistical significance level (<i>p</i> < 0.05) of the hypothesis test. The SSCL algorithm proposed in this paper included the procedures of the competitive prototype seed selection, K-means optimization, and the nearest neighbor classifications. The pattern classification experimental results showed that the proposed SSCL method can provide the excellent diagnostic performances in terms of accuracy (0.838), recall (0.825), specificity (0.85), precision (0.846), F-score (0.835), Matthews correlation coefficient (0.675), area under the receiver operating characteristic curve (0.939), and Kappa coefficient (0.675), which were consistently better than those results of conventional KNN or SVM classifiers. |
| first_indexed | 2024-03-09T10:22:14Z |
| format | Article |
| id | doaj.art-688713e06684483e960888fc8a098b21 |
| institution | Directory Open Access Journal |
| issn | 2079-6374 |
| language | English |
| last_indexed | 2024-03-09T10:22:14Z |
| publishDate | 2022-07-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Biosensors |
| spelling | doaj.art-688713e06684483e960888fc8a098b212023-12-01T21:57:02ZengMDPI AGBiosensors2079-63742022-07-0112750210.3390/bios12070502Classification of Dysphonic Voices in Parkinson’s Disease with Semi-Supervised Competitive Learning AlgorithmGuidong Bao0Mengchen Lin1Xiaoqian Sang2Yangcan Hou3Yixuan Liu4Yunfeng Wu5School of Informatics, Xiamen University, 422 Si Ming South Road, Xiamen 361005, ChinaSchool of Informatics, Xiamen University, 422 Si Ming South Road, Xiamen 361005, ChinaSchool of Informatics, Xiamen University, 422 Si Ming South Road, Xiamen 361005, ChinaSchool of Informatics, Xiamen University, 422 Si Ming South Road, Xiamen 361005, ChinaSchool of Informatics, Xiamen University, 422 Si Ming South Road, Xiamen 361005, ChinaSchool of Informatics, Xiamen University, 422 Si Ming South Road, Xiamen 361005, ChinaThis article proposes a novel semi-supervised competitive learning (SSCL) algorithm for vocal pattern classifications in Parkinson’s disease (PD). The acoustic parameters of voice records were grouped into the families of jitter, shimmer, harmonic-to-noise, frequency, and nonlinear measures, respectively. The linear correlations were computed within each acoustic parameter family. According to the correlation matrix results, the jitter, shimmer, and harmonic-to-noise parameters presented as highly correlated in terms of Pearson’s correlation coefficients. Then, the principal component analysis (PCA) technique was implemented to eliminate the redundant dimensions of the acoustic parameters for each family. The Mann–Whitney–Wilcoxon hypothesis test was used to evaluate the significant difference of the PCA-projected features between the healthy subjects and PD patients. Eight dominant PCA-projected features were selected based on the eigenvalue threshold criterion and the statistical significance level (<i>p</i> < 0.05) of the hypothesis test. The SSCL algorithm proposed in this paper included the procedures of the competitive prototype seed selection, K-means optimization, and the nearest neighbor classifications. The pattern classification experimental results showed that the proposed SSCL method can provide the excellent diagnostic performances in terms of accuracy (0.838), recall (0.825), specificity (0.85), precision (0.846), F-score (0.835), Matthews correlation coefficient (0.675), area under the receiver operating characteristic curve (0.939), and Kappa coefficient (0.675), which were consistently better than those results of conventional KNN or SVM classifiers.https://www.mdpi.com/2079-6374/12/7/502Parkinson’s diseasesemi-supervised learningdysphoniaK-means clusteringcompetitive learningk-nearest neighbor |
| spellingShingle | Guidong Bao Mengchen Lin Xiaoqian Sang Yangcan Hou Yixuan Liu Yunfeng Wu Classification of Dysphonic Voices in Parkinson’s Disease with Semi-Supervised Competitive Learning Algorithm Biosensors Parkinson’s disease semi-supervised learning dysphonia K-means clustering competitive learning k-nearest neighbor |
| title | Classification of Dysphonic Voices in Parkinson’s Disease with Semi-Supervised Competitive Learning Algorithm |
| title_full | Classification of Dysphonic Voices in Parkinson’s Disease with Semi-Supervised Competitive Learning Algorithm |
| title_fullStr | Classification of Dysphonic Voices in Parkinson’s Disease with Semi-Supervised Competitive Learning Algorithm |
| title_full_unstemmed | Classification of Dysphonic Voices in Parkinson’s Disease with Semi-Supervised Competitive Learning Algorithm |
| title_short | Classification of Dysphonic Voices in Parkinson’s Disease with Semi-Supervised Competitive Learning Algorithm |
| title_sort | classification of dysphonic voices in parkinson s disease with semi supervised competitive learning algorithm |
| topic | Parkinson’s disease semi-supervised learning dysphonia K-means clustering competitive learning k-nearest neighbor |
| url | https://www.mdpi.com/2079-6374/12/7/502 |
| work_keys_str_mv | AT guidongbao classificationofdysphonicvoicesinparkinsonsdiseasewithsemisupervisedcompetitivelearningalgorithm AT mengchenlin classificationofdysphonicvoicesinparkinsonsdiseasewithsemisupervisedcompetitivelearningalgorithm AT xiaoqiansang classificationofdysphonicvoicesinparkinsonsdiseasewithsemisupervisedcompetitivelearningalgorithm AT yangcanhou classificationofdysphonicvoicesinparkinsonsdiseasewithsemisupervisedcompetitivelearningalgorithm AT yixuanliu classificationofdysphonicvoicesinparkinsonsdiseasewithsemisupervisedcompetitivelearningalgorithm AT yunfengwu classificationofdysphonicvoicesinparkinsonsdiseasewithsemisupervisedcompetitivelearningalgorithm |