Exploring digital speech biomarkers of hypokinetic dysarthria
in a multilingual cohort

J 2024

Exploring digital speech biomarkers of hypokinetic dysarthria in a multilingual cohort

KOVÁČ, Daniel, Jiří MEKYSKA, Vered AHARONSON, Pavol HARAR, Zoltan GALAZ et. al.

Základní údaje

Originální název

Exploring digital speech biomarkers of hypokinetic dysarthria in a multilingual cohort

Autoři

KOVÁČ, Daniel (203 Česká republika), Jiří MEKYSKA (203 Česká republika), Vered AHARONSON, Pavol HARAR, Zoltan GALAZ, Steven RAPCSAK, Juan Rafael OROZCO-ARROYAVE, Luboš BRABENEC (203 Česká republika, domácí) a Irena REKTOROVÁ (203 Česká republika, garant, domácí)

Vydání

Biomedical Signal Processing and Control, OXFORD, ELSEVIER SCI LTD, 2024, 1746-8094

Další údaje

Jazyk

angličtina

Typ výsledku

Článek v odborném periodiku

Obor

30401 Health-related biotechnology

Stát vydavatele

Velká Británie a Severní Irsko

Utajení

není předmětem státního či obchodního tajemství

Odkazy

URL

Impakt faktor

Impact factor: 5.100 v roce 2022

Organizační jednotka

Středoevropský technologický institut

DOI

http://dx.doi.org/10.1016/j.bspc.2023.105667

UT WoS

001108281100001

Klíčová slova anglicky

Hypokinetic dysarthria; Parkinson's disease; Multilingual study; Acoustic speech features; Statistical analysis; Machine learning

Štítky

rivok

Příznaky

Mezinárodní význam, Recenzováno

Změněno: 3. 3. 2024 22:42, Mgr. Eva Dubská

Anotace

V originále

Hypokinetic dysarthria, a motor speech disorder characterized by reduced movement and control in the speech -related muscles, is mostly associated with Parkinson's disease. Acoustic speech features thus offer the potential for early digital biomarkers to diagnose and monitor the progression of this disease. However, the influence of language on the successful classification of healthy and dysarthric speech remains crucial. This paper explores the analysis of acoustic speech features, both established and newly proposed, in a multilingual context to support the diagnosis of PD. The study aims to identify language-independent and highly discriminative digital speech biomarkers using statistical analysis and machine learning techniques. The study analyzes thirty-three acoustic features extracted from Czech, American, Israeli, Columbian, and Italian PD patients, as well as healthy controls. The analysis employs correlation and statistical tests, descriptive statistics, and the XGBoost classifier. Feature importances and Shapley values are used to provide explanations for the classification results. The study reveals that the most discriminative features, with reduced language dependence, are those measuring the prominence of the second formant, monopitch, and the frequency of pauses during text reading. Classification accuracies range from 67% to 85%, depending on the language. This paper introduces the concept of language robustness as a desirable quality in digital speech biomarkers, ensuring consistent behaviour across languages. By leveraging this concept and employing additional metrics, the study proposes several language-independent digital speech biomarkers with high discrimination power for diagnosing PD.

Návaznosti

LX22NPO5107, projekt VaV

Název: Národní ústav pro neurologický výzkum

Investor: Ministerstvo školství, mládeže a tělovýchovy ČR, Národní ústav pro neurologický výzkum, 5.1 EXCELES

NU20-04-00294, projekt VaV

Název: Diagnostika onemocnění s Lewyho tělísky v prodromálním stadiu založená na analýze multimodálních dat

Investor: Ministerstvo zdravotnictví ČR, Diagnostika onemocnění s Lewyho tělísky v prodromálním stadiu založená na analýze multimodálních dat

734718, interní kód MU

Název: Novel Network-Based Approaches for Studying Cognitive Dysfunction in Behavioral Neurology (Akronym: CoBeN)

Investor: Evropská unie, Novel Network-Based Approaches for Studying Cognitive Dysfunction in Behavioral Neurology, MSCA Marie Skłodowska-Curie Actions (Excellent Science)

Citovat

KOVÁČ, Daniel, Jiří MEKYSKA, Vered AHARONSON, Pavol HARAR, Zoltan GALAZ, Steven RAPCSAK, Juan Rafael OROZCO-ARROYAVE, Luboš BRABENEC a Irena REKTOROVÁ. Exploring digital speech biomarkers of hypokinetic dysarthria in a multilingual cohort. Biomedical Signal Processing and Control. OXFORD: ELSEVIER SCI LTD, 2024, roč. 88, B, s. 1-11. ISSN 1746-8094. Dostupné z: https://dx.doi.org/10.1016/j.bspc.2023.105667.

@article{2371788,
   author = {Kováč, Daniel and Mekyska, Jiří and Aharonson, Vered and Harar, Pavol and Galaz, Zoltan and Rapcsak, Steven and OrozcoandArroyave, Juan Rafael and Brabenec, Luboš and Rektorová, Irena},
   article_location = {OXFORD},
   article_number = {B},
   doi = {http://dx.doi.org/10.1016/j.bspc.2023.105667},
   keywords = {Hypokinetic dysarthria; Parkinson's disease; Multilingual study; Acoustic speech features; Statistical analysis; Machine learning},
   language = {eng},
   issn = {1746-8094},
   journal = {Biomedical Signal Processing and Control},
   title = {Exploring digital speech biomarkers of hypokinetic dysarthria in a multilingual cohort},
   url = {https://www.sciencedirect.com/science/article/pii/S174680942301100X?via%3Dihub},
   volume = {88},
   year = {2024}
}

TY  - JOUR
ID  - 2371788
AU  - Kováč, Daniel - Mekyska, Jiří - Aharonson, Vered - Harar, Pavol - Galaz, Zoltan - Rapcsak, Steven - Orozco-Arroyave, Juan Rafael - Brabenec, Luboš - Rektorová, Irena
PY  - 2024
TI  - Exploring digital speech biomarkers of hypokinetic dysarthria in a multilingual cohort
JF  - Biomedical Signal Processing and Control
VL  - 88
IS  - B
SP  - 1-11
EP  - 1-11
PB  - ELSEVIER SCI LTD
SN  - 17468094
KW  - Hypokinetic dysarthria
KW  - Parkinson's disease
KW  - Multilingual study
KW  - Acoustic speech features
KW  - Statistical analysis
KW  - Machine learning
UR  - https://www.sciencedirect.com/science/article/pii/S174680942301100X?via%3Dihub
N2  - Hypokinetic dysarthria, a motor speech disorder characterized by reduced movement and control in the speech -related muscles, is mostly associated with Parkinson's disease. Acoustic speech features thus offer the potential for early digital biomarkers to diagnose and monitor the progression of this disease. However, the influence of language on the successful classification of healthy and dysarthric speech remains crucial. This paper explores the analysis of acoustic speech features, both established and newly proposed, in a multilingual context to support the diagnosis of PD. The study aims to identify language-independent and highly discriminative digital speech biomarkers using statistical analysis and machine learning techniques. The study analyzes thirty-three acoustic features extracted from Czech, American, Israeli, Columbian, and Italian PD patients, as well as healthy controls. The analysis employs correlation and statistical tests, descriptive statistics, and the XGBoost classifier. Feature importances and Shapley values are used to provide explanations for the classification results. The study reveals that the most discriminative features, with reduced language dependence, are those measuring the prominence of the second formant, monopitch, and the frequency of pauses during text reading. Classification accuracies range from 67% to 85%, depending on the language. This paper introduces the concept of language robustness as a desirable quality in digital speech biomarkers, ensuring consistent behaviour across languages. By leveraging this concept and employing additional metrics, the study proposes several language-independent digital speech biomarkers with high discrimination power for diagnosing PD.
ER  -

KOVÁČ, Daniel, Jiří MEKYSKA, Vered AHARONSON, Pavol HARAR, Zoltan GALAZ, Steven RAPCSAK, Juan Rafael OROZCO-ARROYAVE, Luboš BRABENEC a Irena REKTOROVÁ. Exploring digital speech biomarkers of hypokinetic dysarthria in a multilingual cohort. \textit{Biomedical Signal Processing and Control}. OXFORD: ELSEVIER SCI LTD, 2024, roč.~88, B, s.~1-11. ISSN~1746-8094. Dostupné z: https://dx.doi.org/10.1016/j.bspc.2023.105667.

Podrobný výpis o publikaci