نوع مقاله: مقاله کامل پژوهشی

نویسندگان

1 گروه مهندسی پزشکی، دانشکده مهندسی،دانشگاه فردوسی مشهد، مشهد - باشگاه پژوهشگران جوان و نخبگان، واحد مشهد، دانشگاه آزاد اسلامی، مشهد - قطب علمی رایانش نرم و پردازش هوشمند اطلاعات، دانشگاه فردوسی مشهد، مشهد

2 گروه مهندسی پزشکی، واحد مشهد ، دانشگاه آزاد اسلامی، مشهد

3 گروه مهندسی پزشکی، دانشکده مهندسی،دانشگاه فردوسی مشهد، مشهد - قطب علمی رایانش نرم و پردازش هوشمند اطلاعات، دانشگاه فردوسی مشهد، مشهد

4 گروه عصب شناسی، دانشکده علوم پزشکی، دانشگاه فردوسی مشهد، مشهد

5 گروه مهندسی پزشکی، واحد مشهد ، دانشگاه آزاد اسلامی، مشهد - باشگاه پژوهشگران جوان و نخبگان، واحد مشهد، دانشگاه آزاد اسلامی، مشهد

6 گروه مهندسی پزشکی، دانشکده مهندسی،دانشگاه فردوسی مشهد، مشهد - باشگاه پژوهشگران جوان و نخبگان، واحد مشهد، دانشگاه آزاد اسلامی، مشهد

10.22041/ijbme.2017.55213.1174

چکیده

در سال‌های اخیر، محققین تلاش‌های زیادی برای تشخیص بیماری پارکینسون از طریق یافتن ارتباط آن با سیگنال گفتار افراد انجام داده‌اند. همچنین پژوهش‌هایی در تعیین شدت بیماری و ارتباط آن با اختلالات صوتی انجام شده است. هدف این مقاله، ارزیابی و مقایسة توانایی دسته‌ ویژگی‌های مختلف استخراجی‌ از سیگنال گفتار، در تشخیص بیماری پارکینسون است. برای این منظور، 12 دسته ویژگی از سیگنال گفتار ارزیابی شده‌‌اند، تحلیل صدا روی قسمت آواسازی افراد انجام شده و واج /آ/ توسط افراد بیان شده است. با انتخاب بهترین ویژگی‌ها از هر دسته، که شامل 132 ویژگی است، به روش تسکین و اعمال آن به طبقه‌بندی کنندة ماشین بردار پشتیبان، مقایسه‌ای بین دسته ویژگی‌های مختلف انجام شد. همچنین با ترکیب ویژگی‌های منتخب از هر دسته، صحت تفکیک بسیار خوب 95.93 درصد، در جداسازی گروه سالم از بیمار به‌دست آمد. نتایج حاصل از این پژوهش، می‌تواند گامی بسیار مهم در تشخیص غیرتهاجمی بیماری پارکینسون باشد.

کلیدواژه‌ها

موضوعات

عنوان مقاله [English]

Optimal Feature Selection and Comparison for Automatic Detection of Parkinson's Disease Using Speech Signal

نویسندگان [English]

  • Hamid Azadi 1
  • Mohammad Ali Khalil Zade 2
  • Mohammad Reza Akbarzade Toutounchi 3
  • Hamid Reza Kobravi 2
  • Fariborz Rezaei Talab 4
  • Seyed Amir Ziafati Bagherzade 5
  • Alireza Noei Sarcheshme 6
  • Nina Shahsavan Pour 2

1 M.Eng,Bioelectric Department, Biomedical Engineering Faculty, Ferdowsi University of biomedical, Mashhad, Iran - Center of Excellence on Soft Computing and Intelligent Information Processing, Ferdowsi University, Mashhad, Iran

2 Ph.D, Bioelectric Department, Biomedical Engineering Faculty, Azad University, Mashhad, Iran

3 Professor, Bioelectric Department, Biomedical Engineering Faculty, Ferdowsi University, Mashhad, Iran - Center of Excellence on Soft Computing and Intelligent Information Processing, Ferdowsi University, Mashhad, Iran

5 M.Eng, Bioelectric Department, Biomedical Engineering Faculty, Azad University, Mashhad, Iran

6 M.Eng,Bioelectric Department, Biomedical Engineering Faculty, Ferdowsi University of biomedical, Mashhad, Iran

چکیده [English]

In recent years, researchers have tried hardly to diagnose Parkinson's disease through finding its relation with the patient's speech signal. Also, many studies have been performed on determining the intensity of the disease and its relation with vocal impairment measures. In this paper, we aim to assess and compare the ability of extracting different feature sets from speech signal in order to Parkinson's disease diagnosis. Therefore, 132 features were used to measure vocal impairments from the voice signal of individuals vocalizing phoneme /a/. Then, we used RELIEF feature selection method and applied it to Support Vector Machine (SVM) classifier to choose the best feature of each class. A comparison was made between different feature sets, and finally discrimination percent 95.93 was reached to separate patients from the healthy ones using the combination of selected features. Results obtained from this research can be a very important step toward diagnosing Parkinson's disease non-invasively.
 

کلیدواژه‌ها [English]

  • Parkinson's disease
  • Speech signal processing
  • RELIEFfeature selection
  • Support Vector Machine

[1]     P. M. Keeney, "Parkinson’s Disease Brain Mitochondrial Complex I Has Oxidatively Damaged Subunits and Is Functionally Impaired and Misassembled," The Journal of Neuroscience, vol. 26, no. 19, p. 5256 –5264, 2006.

[2]     C. G. Goetz, "The History of Parkinson's Disease: Early Clinical Descriptions and Neurological Therapies," Cold Spring Harb Perspect Med, 2011.

[3]     M. Novotný, "Automatic Evaluation of Articulatory Disorders in Parkinson’s Disease," IEEE/ACM TRANSACTIONS ON AUDIO, SPEECH, AND LANGUAGE PROCESSING, vol. 22, no. 9, 2014.

[4]     J. Jankovic, "Parkinson’s disease: Clinical features and diagnosis," Jornal of Neurology and Neurosurgery, vol. 79, p. 368–376, 2012.

[5]     M. C. Rodriguez-Oroz, M. Jahanshahi, P. Krack, I. Litvan, R. Macias, E. Bezard and J. A. Obeso, "Initial clinical manifestations of Parkinson’s disease: Features and pathophysiological mechanisms," Lancet Neurology, vol. 8, p. 1128–1139, 2009.

[6]     S. Lesage and A. Brice, "Parkinson’s disease: from monogenic forms to genetic susceptibility factors," Human Molecular Genetics, vol. 18, no. 1, pp. 48-59, 2009.

[7]     M. C. Henry and E. R. Stapleton, EMT Prehospital Care, Jones & Bartlett Publishers, 2011.

[8]     G. Becker, A. Müller, S. Braune, T. Büttner, R. Benecke, W. Greulich, W. Klein, G. Mark, J. Rieke and R. Thümler, "Early diagnosis of Parkinson’s disease," Jornal of Neurology, vol. 249, p. III/40–III/48, 2002.

[9]     S. Rewar, "A systematic review on Parkinson's disease," Indian Journal of Research in Pharmacy and Biotechnology3.2, pp. 176-185, 2015.

[10] R. J. A. K. Ho, C. Marigliani, J. L. Bradshaw and S. Gates, "Speech impairment in a large sample of people with Parkinson’s disease," Behavioral Neurology, vol. 11, p. 131–137, 1998.

[11] J. R. Orozco-Arroyave, F. Honig, J. D. A.-L. no, J. F. V. Bonilla and E. Noth, "Spectral and cepstral analyses for Parkinson’s disease detection in Spanish vowels and words," Jornal of Expert Systems, p. 1–10, 2015.

[12] Tsanas, M. Little, P. McSharry and L. Ramig, "Novel speech signal processing algorithms for high-accuracy classification of Parkinson’s disease," IEEE Trans. Biomed. Eng., vol. 59, no. 5, p. 1264–1271, 2012.

[13] M. A. Little, P. E. McSharry, E. J. Hunter, J. Spielman and L. O. Ramig, "Suitability of dysphonia measurements for telemonitoring of Parkinson’s disease," IEEE Transactions on Biomedical Engineering, vol. 56, no. 4, p. 1015–1022, 2009.

[14] H. Azadi, M.-R. Akbarzadeh-T, H.-R. Kobravi and A. Noei, "Presentation of a New Gender Dependent Feature Selection Approach for Diagnosis of Parkinson Disease using Speech Signal Processing," in Second International Congress on Technology, Communication and Knowledge (ICTCK 2015), Mashhad, Iran, 2015.

[15] M. Hariharan, K. Polat and R. Sindhu, "A new hybrid intelligent system for accurate detection of Parkinson’s disease," Comput. Methods Programs Biomed, vol. 113, no. 3, p. 904–913, 2014.

[16] W. Ji and Y. Li, "Energy-based feature ranking for assessing the dysphonia measurements in Parkinson detection," IET signal Process, vol. 6, no. 4, p. 300–305, 2012.

[17] Tsanas, M. A. Little, P. E. McSharry and L. O. Ramig, "Nonlinear speech analysis algorithms mapped to a standard metric achieve clinically useful quantification of average Parkinson’s disease symptom severity," J. Roy. Soc., vol. 8, p. 842–855, 2011.

[18] Kononenko, "Estimating attributes: Analysis and extensions of RELIEF," Estimating attributes in Machine Learning: ECML-94, p. 171–182, 1994.

[19] Kira and L. A. Rendell, "A practical approach to feature selection," Proceedings of the ninth international workshop on Machine learning, p. 249–256, 1992.

[20] V.Vapnik, The nature of statistical learning theory, Springer, 1999.

[21] Schölkopf, C. J. Burges and A. J. Smola, Advances in Kernel Methods: Support Vector Learning, Cambridge, MA: MIT Press, 1999.

[22] R. JD, P. A and L. JA., "Sensitivity analysis of kappa-fold cross validation in prediction error estimation," IEEE Trans. Pattern Anal. Mach. Intell., 2010.

[23] M. W. Powers, "Evaluation: From Precision, Recall and F-Measure to ROC, Informedness, Markedness & Correlation," Journal of Machine Learning Technologies, vol. 2, no. 1, pp. 37-63, 2011.