Parkinson’s Disease Detection Based on Spectrogram-Deep Convolutional Generative Adversarial Network Sample Augmentation
As an essential biological feature of human beings, voiceprint is increasingly used in medical research and diagnosis, especially in identifying Parkinson's Disease (PD). This paper proposes a Spectrogram Deep Convolutional Generative Adversarial Network (S-DCGAN) for sample augmentation to ove...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2020-01-01
|
| Series: | IEEE Access |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/9257451/ |
| _version_ | 1819173335913725952 |
|---|---|
| author | Zhi-Jing Xu Rong-Fei Wang Juan Wang Da-Hai Yu |
| author_facet | Zhi-Jing Xu Rong-Fei Wang Juan Wang Da-Hai Yu |
| author_sort | Zhi-Jing Xu |
| collection | DOAJ |
| description | As an essential biological feature of human beings, voiceprint is increasingly used in medical research and diagnosis, especially in identifying Parkinson's Disease (PD). This paper proposes a Spectrogram Deep Convolutional Generative Adversarial Network (S-DCGAN) for sample augmentation to overcome the limited amount of existing patient voiceprint datasets and samples. S-DCGAN generates a high-resolution spectrogram by increasing network layers, adding the Spectral Normalization (SN) method, and combining feature matching strategy. The high-similarity and low-distortion spectrogram are selected in light of Structural Similarity Index (SSIM) values and Peak Signal to Noise Ratio (PSNR) to augment the samples. Fréchet Inception Distance (FID) and GAN-train result show the generalization ability of the generated data. We construct the ResNet50 model with a Global Average Pooling(GAP) layer to extract the voiceprint features and classify them effectively to improve recognition accuracy. The GAP suppresses the over-fitting problem and optimizes quickly. Finally, on the Sakar dataset, comparative experiments were conducted on different models and classification methods. Results show that the S-DCGAN-ResNet50 hybrid model can achieve the highest voiceprint recognition accuracy of 91.25% and specificity of 92.5%, which can distinguish between PD patients and healthy people more precisely compared with DCGAN-ResNet50. It augments the application environment of voiceprint recognition in the medical field and makes it universal in different datasets. |
| first_indexed | 2024-12-22T20:21:27Z |
| format | Article |
| id | doaj.art-67bc142d939e46adbb6a884153b8a9b9 |
| institution | Directory Open Access Journal |
| issn | 2169-3536 |
| language | English |
| last_indexed | 2024-12-22T20:21:27Z |
| publishDate | 2020-01-01 |
| publisher | IEEE |
| record_format | Article |
| series | IEEE Access |
| spelling | doaj.art-67bc142d939e46adbb6a884153b8a9b92022-12-21T18:13:50ZengIEEEIEEE Access2169-35362020-01-01820688820690010.1109/ACCESS.2020.30377759257451Parkinson’s Disease Detection Based on Spectrogram-Deep Convolutional Generative Adversarial Network Sample AugmentationZhi-Jing Xu0https://orcid.org/0000-0002-1182-4537Rong-Fei Wang1https://orcid.org/0000-0002-1845-0283Juan Wang2https://orcid.org/0000-0002-7944-3685Da-Hai Yu3https://orcid.org/0000-0002-4663-6165College of Information Engineering, Shanghai Maritime University, Shanghai, ChinaCollege of Information Engineering, Shanghai Maritime University, Shanghai, ChinaCollege of Information Engineering, Shanghai Maritime University, Shanghai, ChinaShanghai Experimental School, Shanghai, ChinaAs an essential biological feature of human beings, voiceprint is increasingly used in medical research and diagnosis, especially in identifying Parkinson's Disease (PD). This paper proposes a Spectrogram Deep Convolutional Generative Adversarial Network (S-DCGAN) for sample augmentation to overcome the limited amount of existing patient voiceprint datasets and samples. S-DCGAN generates a high-resolution spectrogram by increasing network layers, adding the Spectral Normalization (SN) method, and combining feature matching strategy. The high-similarity and low-distortion spectrogram are selected in light of Structural Similarity Index (SSIM) values and Peak Signal to Noise Ratio (PSNR) to augment the samples. Fréchet Inception Distance (FID) and GAN-train result show the generalization ability of the generated data. We construct the ResNet50 model with a Global Average Pooling(GAP) layer to extract the voiceprint features and classify them effectively to improve recognition accuracy. The GAP suppresses the over-fitting problem and optimizes quickly. Finally, on the Sakar dataset, comparative experiments were conducted on different models and classification methods. Results show that the S-DCGAN-ResNet50 hybrid model can achieve the highest voiceprint recognition accuracy of 91.25% and specificity of 92.5%, which can distinguish between PD patients and healthy people more precisely compared with DCGAN-ResNet50. It augments the application environment of voiceprint recognition in the medical field and makes it universal in different datasets.https://ieeexplore.ieee.org/document/9257451/Parkinson’s diseaseResNet50S-DCGANsample augumentationspectrogram |
| spellingShingle | Zhi-Jing Xu Rong-Fei Wang Juan Wang Da-Hai Yu Parkinson’s Disease Detection Based on Spectrogram-Deep Convolutional Generative Adversarial Network Sample Augmentation IEEE Access Parkinson’s disease ResNet50 S-DCGAN sample augumentation spectrogram |
| title | Parkinson’s Disease Detection Based on Spectrogram-Deep Convolutional Generative Adversarial Network Sample Augmentation |
| title_full | Parkinson’s Disease Detection Based on Spectrogram-Deep Convolutional Generative Adversarial Network Sample Augmentation |
| title_fullStr | Parkinson’s Disease Detection Based on Spectrogram-Deep Convolutional Generative Adversarial Network Sample Augmentation |
| title_full_unstemmed | Parkinson’s Disease Detection Based on Spectrogram-Deep Convolutional Generative Adversarial Network Sample Augmentation |
| title_short | Parkinson’s Disease Detection Based on Spectrogram-Deep Convolutional Generative Adversarial Network Sample Augmentation |
| title_sort | parkinson x2019 s disease detection based on spectrogram deep convolutional generative adversarial network sample augmentation |
| topic | Parkinson’s disease ResNet50 S-DCGAN sample augumentation spectrogram |
| url | https://ieeexplore.ieee.org/document/9257451/ |
| work_keys_str_mv | AT zhijingxu parkinsonx2019sdiseasedetectionbasedonspectrogramdeepconvolutionalgenerativeadversarialnetworksampleaugmentation AT rongfeiwang parkinsonx2019sdiseasedetectionbasedonspectrogramdeepconvolutionalgenerativeadversarialnetworksampleaugmentation AT juanwang parkinsonx2019sdiseasedetectionbasedonspectrogramdeepconvolutionalgenerativeadversarialnetworksampleaugmentation AT dahaiyu parkinsonx2019sdiseasedetectionbasedonspectrogramdeepconvolutionalgenerativeadversarialnetworksampleaugmentation |