EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

CNN-Based Identification of Parkinson’s Disease from Continuous Speech in Noisy Environments

Paul Faragó iD, Sebastian-Aurelian Ștefănigă iD, Claudia-Georgiana Cordoș, Laura-Ioana Mihăilă, Sorin Hintea, Ana-Sorina Peștean, Michel Beyer, Lăcrămioara Perju-Dumbravă, Robert Radu Ileșan iD

Bioengineering · 26.04.2023 · 10 (5) · 531

DOI: 10.3390/bioengineering10050531 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Bioengineering, online erschienen am 26.04.2023. Verfasst von Paul Faragó und 8 weiteren Autor:innen. Beteiligte Einrichtung: Bases of Electronics Department, Faculty of Electronics, Telecommunications and Information Technology, Technical University of Cluj-Napoca, 400114 C…. Die Arbeit wurde bisher 38-mal zitiert (Crossref) und nennt 68 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Parkinson’s disease is a progressive neurodegenerative disorder caused by dopaminergic neuron degeneration. Parkinsonian speech impairment is one of the earliest presentations of the disease and, along with tremor, is suitable for pre-diagnosis. It is defined by hypokinetic dysarthria and accounts for respiratory, phonatory, articulatory, and prosodic manifestations. The topic of this article targets artificial-intelligence-based identification of Parkinson’s disease from continuous speech recorded in a noisy environment. The novelty of this work is twofold. First, the proposed assessment workflow performed speech analysis on samples of continuous speech. Second, we analyzed and quantified Wiener filter applicability for speech denoising in the context of Parkinsonian speech identification. We argue that the Parkinsonian features of loudness, intonation, phonation, prosody, and articulation are contained in the speech, speech energy, and Mel spectrograms. Thus, the proposed workflow follows a feature-based speech assessment to determine the feature variation ranges, followed by speech classification using convolutional neural networks. We report the best classification accuracies of 96% on speech energy, 93% on speech, and 92% on Mel spectrograms. We conclude that the Wiener filter improves both feature-based analysis and convolutional-neural-network-based classification performances.

Schlagwörter

Stichwörter der Autor:innen

Parkinson’s diseaseWiener filterartificial intelligencecontinuous speechconvolutional neural networkshypokinetic dysarthrianoisy speechpre-diagnosisspectrogramsspeech assessment

Autor:innen und Institutionen

  • Bases of Electronics Department, Faculty of Electronics, Telecommunications and Information Technology, Technical University of Cluj-Napoca, 400114 Cluj-Napoca, Romania.
  • Department of Computer Science, Faculty of Mathematics and Computer Science, West University of Timisoara, 300223 Timisoara, Romania.
  • Department of Neurology and Pediatric Neurology, Faculty of Medicine, University of Medicine and Pharmacy "Iuliu Hatieganu" Cluj-Napoca, 400012 Cluj-Napoca, Romania.
  • Clinic of Oral and Cranio-Maxillofacial Surgery, University Hospital Basel, CH-4031 Basel, Switzerland.
  • Medical Additive Manufacturing Research Group (Swiss MAM), Department of Biomedical Engineering, University of Basel, CH-4123 Allschwil, Switzerland.

ORCID-Profile: Paul Faragó, Sebastian-Aurelian Ștefănigă, Robert Radu Ileșan

Zitieren

APA

Faragó, P., Ștefănigă, S. A., Cordoș, C. G., Mihăilă, L. I., Hintea, S., Peștean, A. S., Beyer, M., Perju-Dumbravă, L., & Ileșan, R. R. (2023). CNN-Based Identification of Parkinson’s Disease from Continuous Speech in Noisy Environments. Bioengineering, 10(5), 531. https://doi.org/10.3390/bioengineering10050531

Vancouver

Faragó P, Ștefănigă SA, Cordoș CG, Mihăilă LI, Hintea S, Peștean AS, et al. CNN-Based Identification of Parkinson’s Disease from Continuous Speech in Noisy Environments. Bioengineering. 2023;10(5):531. doi:10.3390/bioengineering10050531

BibTeX
@article{Farag2023,
  author = {Faragó, Paul and Ștefănigă, Sebastian-Aurelian and Cordoș, Claudia-Georgiana and Mihăilă, Laura-Ioana and Hintea, Sorin and Peștean, Ana-Sorina and Beyer, Michel and Perju-Dumbravă, Lăcrămioara and Ileșan, Robert Radu},
  title = {CNN-Based Identification of Parkinson’s Disease from Continuous Speech in Noisy Environments},
  journal = {Bioengineering},
  year = {2023},
  volume = {10},
  number = {5},
  pages = {531},
  doi = {10.3390/bioengineering10050531}
}
Kostenlos starten →