Objective: To analyze syllable-based speech characteristics in patients with Parkinson’s disease (PD), multiple system atrophy (MSA), and cerebellar ataxia (CA) to determine if there is discrimination value.

Background: Speech disorders differ between PD and MSA, but studies focusing on group differences based on syllables or including CA are lacking until now.

Method: From September 2021 to May 2022, patients diagnosed with MSA, PD, and CA, as well as healthy control (HC) were recruited from the Department of Neurology at Seoul National University Hospital. All participants performed four vocal tasks: producing high-pitched and low-pitched sounds for 5 Korean vowels, repeating 14 Korean consonants with the vowel /a/, raising and lowering pitch of the vowel /a/, and continuously repeating /pataka/ for 5 seconds. Acoustic analysis and exploratory analysis using artificial intelligence were conducted to identify the syllable combinations that best distinguish between disease groups and their accuracy.

Results: Ninety-eight probable MSA patients were recruited, with 40 MSA-C (24 male) and 19 MSA-P (10 male), while the remaining were diagnosed with mixed-type MSA. In addition, 100 patients with idiopathic PD (49 male) and 46 patients with CA (22 male) were enrolled. HC group consisted of 100 subjects (47 male) of comparable age (mean age 66.78, SD 7.61). CA patients were significantly younger than PD and MSA (p<0.05). MSA patients had higher age at onset (p<0.05) but shorter disease duration compared to the other two disease groups (p<0.05). Hoehn and Yahr stage was significantly lower in PD patients compared to MSA and CA patients (p=0.000). In syllable-based speech analysis using artificial intelligence, the accuracy was the highest in combination of two tasks, with the top three combinations: /aaa-hahaha/, /nanana-sasasa/, and /dadada-aaa/, all capable of distinguishing one group among the four with an accuracy of 67.50%. Also, CA patients could be distinguished with high accuracy of 86.76% using the /dadada-aaa/. Both PD and MSA patients could be distinguished with over 70% accuracy, using the /aaa-hahaha/ for PD and /nanana-sasasa/ for MSA.

Conclusion: In addition to acoustic parameters, syllable-based speech characteristics can be used to discriminate among Parkinsonian disorders and CA, indicating its potential utility as a promising diagnostic tool.

« Back to 2024 International Congress

MDS Abstracts - https://www.mdsabstracts.org/abstract/syllable-based-speech-characteristics-as-potential-biomarker-for-differential-diagnosis-of-parkinsons-disease-multiple-system-atrophy-and-cerebellar-ataxia/