머신러닝 분석을 통한 일측성 성대마비 진단
Prediction of Unilateral Vocal Cord Paralysis Patients Through Machine Learning Analysis of Acoustic Parameters: A Preliminary Study
0
조회
0
다운로드
0
인용
0
참고문헌
키워드
초록
Background and Objectives The purpose of this study is to evaluate value of diagnostic tool for vocal cord palsy utilizing artificial intelligence without laryngoscope Materials and Method A dataset consisting of recordings from patients with unilateral vocal cord paralysis (n=54) as well as normal individuals (n=163). The dataset included prolonged pronunciations of the vowels /ah/, /u/, /i/, and vocal cord data from paralyzed patients. Various acoustic parameters such as Mel-frequency cepstral coefficients, jitter, shimmer, harmonics-to-noise ratio, and fundamental frequency statistics were analyzed. The classification of vocal cord paralysis encompassed paralysis status, paralysis degree, and paralysis location. The deep learning model employed the leave-one-out method, and the feature set with the highest performance was selected using the following methods. Results Vocal Cord Paralysis Classifier: The classifier accurately distinguished normal voice from vocal cord paralysis, achieving an accuracy and F1 score of 1.0. Paralysis Location Classifier: The classifier accurately differentiated between median and paramedian vocal cord paralysis, achieving an accuracy and micro F1 score of 1.0. Breathiness Degree Classifier: The classifier achieved an accuracy of 0.795 and a mean absolute error of 0.2857 in distinguishing different degrees of breathiness. Conclusion Although the small sample size raises concerns of potential overfitting, this preliminary study highlights distinctive acoustic features in cases of unilateral vocal fold paralysis compared to those of normal individuals. These findings suggest the feasibility of determining the presence, degree, and location of paralysis through the utilization of acoustic parameters. Further research is warranted to validate and expand upon these results.
Abstract
Background and Objectives The purpose of this study is to evaluate value of diagnostic tool for vocal cord palsy utilizing artificial intelligence without laryngoscope Materials and Method A dataset consisting of recordings from patients with unilateral vocal cord paralysis (n=54) as well as normal individuals (n=163). The dataset included prolonged pronunciations of the vowels /ah/, /u/, /i/, and vocal cord data from paralyzed patients. Various acoustic parameters such as Mel-frequency cepstral coefficients, jitter, shimmer, harmonics-to-noise ratio, and fundamental frequency statistics were analyzed. The classification of vocal cord paralysis encompassed paralysis status, paralysis degree, and paralysis location. The deep learning model employed the leave-one-out method, and the feature set with the highest performance was selected using the following methods. Results Vocal Cord Paralysis Classifier: The classifier accurately distinguished normal voice from vocal cord paralysis, achieving an accuracy and F1 score of 1.0. Paralysis Location Classifier: The classifier accurately differentiated between median and paramedian vocal cord paralysis, achieving an accuracy and micro F1 score of 1.0. Breathiness Degree Classifier: The classifier achieved an accuracy of 0.795 and a mean absolute error of 0.2857 in distinguishing different degrees of breathiness. Conclusion Although the small sample size raises concerns of potential overfitting, this preliminary study highlights distinctive acoustic features in cases of unilateral vocal fold paralysis compared to those of normal individuals. These findings suggest the feasibility of determining the presence, degree, and location of paralysis through the utilization of acoustic parameters. Further research is warranted to validate and expand upon these results.
참고문헌 (14)
- Ha JF, "Unilateral vocal fold palsy & dysphagia : a review", Auris Nasus Larynx, 47(3), pp.315~334 (2020)
- Heikkinen M, "Incidence, risk factors, and natural outcome of vocal fold paresis in 920 thyroid operations with routine pre-and postoperative laryngoscopic evaluation", World J Surg, 43(9), pp.2228~2234 (2019)
- Ortega Beltrá N, "Extralaryngeal causes of unilateral vocal cord paralysis : aetiology and prognosis", Acta Otorrinolaringol Esp(Engl Ed), 73(6), pp.376~383 (2022)
- Loochtan MJ, "Vocal fold paralysis after esophagectomy for carcinoma", Otolaryngol Head Neck Surg, 155(1), pp.122~126 (2016)
- Bando H, "Vocal fold paralysis as a sign of chest diseases : a 15-year retrospective study", World J Surg, 30(3), pp.293~298 (2006)
- Rosenthal LH, "Vocal fold immobility : a longitudinal analysis of etiology over 20 years", Laryngoscope, 117(10), pp.1864~1870 (2007)
- Jo GE, "A study on user experience of artificial intelligence speaker", J Korea Converg Soc, 9(8), pp.127~133 (2018)
- Kim SO, "Analysis of human-AI dialog involving Korean versus Chinese users", J East Aisan Cult, 91, pp.61~92 (2022)
- Jeong YH, "Exploring the efficacy of an AI chatbot-based automatic speech recognition based on a comparative error analysis", Multimed Assist Lang Learn, 24(4), pp.261~289 (2021)
- Hong KH, "The vertical level of the paralyzed vocal fold in unilateral vocal fold paralysis", Korean J Otorhinolaryngol-Head Neck Surg, 43(8), pp.866~872 (2000)
- Yeo EJ, "Automatic severity classification of dysarthria using voice quality, prosody, and pronunciation features", Phon Speech Sci, 13(2), pp.57~66 (2021)
- Titze IR, "Workshop on acoustic voice analysis: summary statement", pp.1~36 (1995), National Center for Voice and Speech
- "Western Michigan University", 2000
- Hernandez A, "Prosody-based measures for automatic severity assessment of dysarthric speech", Appl Sci, 10(19), pp.6999~ (2020)