Abstract
Parkinson’s disease (PD) is a neuron-related disorder due to the decrease in dopaminergic neurons present in the midbrain. For the last few decades, speech is an emerging interest in the analysis and detection of PD. In this study, a predictive machine learning framework based on extreme gradient boosting (XGBoost) features selection and a stacked ensemble approach is presented to investigate the voice tremor of people suffering from PD. The proposed framework consists of two stages: In the first stage the optimized features are obtained using XGBoost features selection, and in the second stage a PD detection system is developed using stacked ensemble classifiers. Leave one subject out (LOSO) cross-validation shows that the proposed framework gives average accuracy of up to 95.07% compared to results obtained with individual classifiers. Additionally, it was also concluded that reduced features had given the highest classification accuracy compared to the raw features set which saves training time and enhances the prediction accuracy.
Similar content being viewed by others
References
K Dashtipour A Tafreshi J Lee B Crawley 2018 Speech disorders in Parkinson's disease: pathophysiology, medical management and surgical approaches Neurodegener. Dis. Manag. 8 5 337 348
A Roberts D Post 2018 Information content and efficiency in the spoken discourse of individuals with Parkinson's disease J. Speech Lang. Hear. Res. 61 9 2259 2274
J Mühlhaus H Frieg K Bilda U Ritterfeld 2017 Game-based speech rehabilitation for people with Parkinson’s disease Lect. Notes Comput. Sci. 10279 76 85
MA Little PE McSharry EJ Hunter J Spielman LO Ramig 2009 Suitability of dysphonia measurements for telemonitoring of Parkinson's disease IEEE Trans. Biomed. Eng. 56 4 1015 1022
J Rusz R Čmejla H Ruzickova E Růžička 2011 Quantitative acoustic measurements for characterization of speech and voice disorders in early untreated Parkinson’s disease J. Acoust. Soc. Am. 129 1 350 367
A Tsanas MA Little PE McSharry J Spielman LO Ramig 2012 Novel speech signal processing algorithms for high accuracy classification of Parkinsons disease IEEE Trans. Biomed. Eng. 59 1264 1271
S Lahmiri A Shmuel 2019 Detection of Parkinson’s disease based on voice patterns ranking and optimized support vector machine Biomed. Signal Process. Control 49 427 433
M Novotný J Rusz R Čmejla E Růžička 2014 Automatic evaluation of articulatory disorders in Parkinson’s disease IEEE/ACM Trans. Audio Speech Lang. Process. 22 9 1366 1378
JR Orozco-Arroyave JD Arias-Londoño JF Vargas-Bonilla E Nöth 2013 Analysis of speech from people with Parkinson’s disease through nonlinear dynamics Lect. Notes Comput. Sci. 7911 112 119
BE Sakar ME Isenkul CO Sakar A Sertbas F Gurgen S Delil H Apaydin O Kursun 2013 Collection and analysis of a Parkinson speech dataset with multiple types of sound recordings IEEE J. Biomed. Health Inform. 17 828 834
JR Orozco-Arroyave F Hönig JD Arias-Londoño JF Vargas-Bonilla E Nöth 2015 Spectral and cepstral analyses for Parkinson’s disease detection in Spanish vowels and words Expert. Syst. 32 688 697
T Khan J Westin M Dougherty 2014 Cepstral separation difference: A novel approach for speech impairment quantification in Parkinson’s disease Biocybern. Biomed. Eng. 34 1 25 34
Tsanas, Athanasios, et al. Accurate telemonitoring of Parkinson’s disease progression by non-invasive speech tests. Nat. Preced. 1–1 (2009)
P Gómez-Vilda J Mekyska JM Ferrández D Palacios-Alonso A Gómez-Rodellar V Rodellar-Biarge Z Galaz Z Smekal I Eliasova M Kostalova I Rektorova 2017 Parkinson’s disease detection from speech articulation neuromechanics Front. Neuroinform. 11 1 17
C. Okan Sakar 2019 A comparative analysis of speech signal processing algorithms for Parkinson’s disease classification and the use of the tunable Q-factor wavelet transform Appl. Soft Comput. 74 255 263
B Karan SS Sahu K Mahto 2020 Parkinson disease prediction using intrinsic mode function based features from speech signal Biocybern. Biomed. Eng. 40 1 249 264
B Karan SS Sahu 2021 An improved framework for Parkinson’s disease prediction using variational mode decomposition-hilbert spectrum of speech signal Biocybern. Biomed. Eng. 41 717 732
A Tsanas 2012 Novel speech signal processing algorithms for high-accuracy classification of Parkinson's disease IEEE Trans. Biomed. Eng. 59 5 1264 1271
G Solana-Lavalle J-C Galán-Hernández R Rosas-Romero 2020 Automatic Parkinson disease detection at early stages as a pre-diagnosis tool by using classifiers and a small set of vocal features Biocybern. Biomed. Eng. 40 1 505 516
G Solana-Lavalle R Rosas-Romero 2021 Analysis of voice as an assisting tool for detection of Parkinson's disease and its subsequent clinical interpretation Biomed. Signal Process. Control 66 102415
H Gunduz 2021 An efficient dimensionality reduction method using filter-based feature selection and variational autoencoders on Parkinson's disease classification Biomed. Signal Process. Control 66 102452
H Gunduz 2019 Deep learning-based Parkinson’s disease classification using vocal feature sets IEEE Access 7 115540 115551
D Gupta S Sundaram A Khanna AE Hassanien VHC Albuquerque De 2018 Improved diagnosis of Parkinson's disease using optimized crow search algorithm Comput. Electr. Eng. https://doi.org/10.1016/j.compeleceng.2018.04.014
D Gupta A Julka S Jain T Aggarwal A Khanna N Arunkumar VHC Albuquerque de 2018 Optimized cuttlefish algorithm for diagnosis of Parkinson’s disease Cogn. Syst. Res. https://doi.org/10.1016/j.cogsys.2018.06.006
Q Zhang 2021 StackPDB: predicting DNA-binding proteins based on XGB-RFE feature optimization and stacked ensemble classifier Appl. Soft Comput. 99 106921
DH Wolpert 1992 Stacked generalization Neural Netw. 5 2 241 259
TG Dietterich 1998 Approximate statistical tests for comparing supervised classification learning algorithms Neural Comput. 10 7 1895 1923
Funding
The authors have not disclosed any funding.
Author information
Authors and Affiliations
Corresponding author
Ethics declarations
Conflict of interest
None.
Additional information
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
About this article
Cite this article
Karan, B. Speech-Based Parkinson’s Disease Prediction Using XGBoost-Based Features Selection and the Stacked Ensemble of Classifiers. J. Inst. Eng. India Ser. B 104, 475–483 (2023). https://doi.org/10.1007/s40031-022-00851-2
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s40031-022-00851-2