Predicting dysphonia by measuring surface electromyographic activity of the supralaryngeal muscles

Abstract

Purpose: This study set out to investigate whether individuals with dysphonia, as determined by either self-assessment or clinician-based auditory-perceptual judg-ment, exhibited differences in perilaryngeal muscle activities using surface elec-tromyography (sEMG) during various phonatory tasks. Additionally, the study aimed to assess the effectiveness of sEMG in identifying dysphonic cases. Method: A total of 77 adults (44 women, 33 men, Mage = 30.4 years) partici-pated in this study, with dysphonic cases identified separately using either a 10-item Voice Handicap Index (VHI-10) or clinician-based auditory-perceptual voice quality (APVQ) evaluation. sEMG activities were measured from the areas of suprahyoid and sternocleidomastoid muscles during prolonged vowel /i/ pho-nations at different pitch and loudness levels. Normalized root-mean-square value against the maximal voluntary contraction (RMS %MVC) of the sEMG sig-nals was obtained for each phonation and compared between subject groups and across phonatory tasks. Additionally, binary logistic regression analysis was performed to determine how the sEMG measures could predict the VHI-10- based or APVQ-based dysphonic cases. Results: Participants who scored above the criteria on either the VHI-10 (n = 29) or APVQ judgment (n = 17) exhibited significantly higher RMS %MVC in the right suprahyoid muscles compared to the corresponding control groups. Although the RMS %MVC value from the right suprahyoid muscles alone was not a significant predictor of self-evaluated dysphonic cases, a combination of the RMS %MVC values from both the right and left suprahyoid muscles signifi-cantly predicted APVQ-based dysphonic cases with a 69.66% fair level. Conclusions: This study found that individuals with dysphonia, as determined by either self-assessment or APVQ judgment, displayed more imbalanced supra-hyoid muscle activities in voice production compared to nondysphonic groups. The combination of the sEMG measures from both left and right suprahyoid mus-cles showed potential as a predictor of dysphonia with a fair level of confidence.