由超音波觀察成人發音時之舌頭運動

Ultrasound Investigation of Tongue Movement During Vowels Articulation

葉為昌(Frank Wei-Chang Yeh)

指導教授 ： 彭建綸 ； 蔡吉陽

臺北醫學大學/口腔學院/牙醫學系碩博士班/碩士(2000年)

摘要

發音時舌部之運動功能為許多牙醫師、齒顎矯正醫師、口腔顎面外科醫師、耳鼻喉科醫師、神經生理學家及語言病理治療師等很感興趣的研究課題。許多發音不正或言語困難的患者，常與舌部運動功能失常有關；然而不論是先天性或後天性的功能失常，大部分皆可經由舌部復健手術、舌部復健運動及發音矯正治療來加以改善；因此若要進一步了解舌部發音功能之生理、病理及復健機制，實有必要使用一適合診斷及研究的工具來對舌頭的發音運動功能做更深一層的瞭解。 過去使用的研究技術，如動態X光、磁氣測量及肌電圖等，應用在研究舌頭運動方面有不少的缺點。超音波測量則為一摩登的方法，它有操作簡單、可顯示真時間 (real time) 動態二度平面圖像及沒有生物傷害性之優點，再加上近來口腔超音波之防震測量技術(cushion scanning technique, CST) 的發展，可提供定量的舌頭運動訊息，因此逐漸受到相關研究者的重視。本研究之目的即為使用此超音波之技術 (CST)，來觀察評估發音時舌部之運動模式，同時嘗試量化舌頭之發音運動，以做為未來相關研究之參考。 本研究選取32名成人，年齡在19至27歲之間，並且無吞嚥困難(dysphagia)、言語障礙症(dysphasia)、或是發聲困難(dysphonia)的病史為研究對象。每一位受測者皆依文字示意，以其自己原本的發音習慣發出[a]、[e]、[i]、[o]、[u]之母音，並各重覆三次。應用口腔超音波之防震測量技術(CST)，以3.5 MHz之探頭(transducer)在受測者發音時截取舌頭中央矢狀面(mid-sagittal plane)影像，同時以B+M模式觀察，並以高畫質錄影機錄取超音波影像，再經由影像截取卡轉存入電腦中，經由分析、計算可得定量訊息。 研究結果顯示，健康成人男性在發母音時的舌頭運動時間平均值[a]為2.31 ±0.39秒，[e]為2.32 ±0.41秒，[i]為2.25 ±0.46秒，[o]為2.48 ±0.34秒，[u]為2.37 ±0.39秒。而成人男性在發母音時的舌體上下運動最大位移的平均值[a]為9.13 ±3.21釐米，[e]為5.48 ±1.78釐米，[i]為6.02 ±2.92釐米，[o]為13.26 ±4.60釐米，[u]為7.17 ±2.43釐米。 在健康成人女性在發母音時的舌頭運動時間平均值[a]為2.47 ±0.50秒，[e]為2.63 ±0.41秒，[i]為2.52 ±0.45秒，[o]為2.44 ±0.34秒，[u]為2.51 ±0.47秒。而成人女性在發母音時的舌體上下運動最大位移的平均值[a]為6.29 ±2.07釐米，[e]為8.96 ±3.63釐米，[i]為7.22 ±3.98釐米，[o]為9.25 ±3.97釐米，[u]為6.04 ±2.57釐米。 經由矢狀面所截取的舌頭超音波及時影像，加上B模式與M模式的輔助，使得口腔超音波在語音學及口腔生理學上，變成更有用、更具診斷價值的利器。

關鍵字

舌頭 ； 母音 ； 超音波

並列摘要

Abstract Many dentists, orthodontists, oral surgeons, ENT doctors, neurologists, physiologists and speech pathologists are interested in the research field of tongue movement during articulation. One of the common reasons for speech distortion or dysphasia in patients is abnormal tongue movements. Irregardless of whether the reasons for the abnormal tongue movements are natural or acquired, most patients’ pronunciation is improved by tongue rehabilitation surgery, tongue myofunctional exercises and speech therapies. In order to understand the mechanisms of physiological, pathological and rehabilitated tongue functions, it is necessary to develop a more suitable method for diagnosis and research of tongue movement during pronunciation. Traditional research techniques, such as cineradiography, as well as electrical magnetic and electromyography, have many draw backs in researching tongue movements. Ultrasonic investigation is a modern method, and it has the advantages of easy operation, 2-D dynamic frames and non-invasiveness. Furthermore, by the recently developed oral cushion scanning technique (CST), ultrasound can provide quantitative information on tongue movement, and it has been noticed by the relative researchers. The purpose of this study is to use the cushion scanning technique to investigate and quantify movement patterns of the tongue, as well as to serve as the basis for future research. To obtain normative data of tongue motion with this system, 32 normal-hearing native adult speakers were chosen. Those with no dysphagia, dysphasia or dysphonia, as well as having no oral-motor, dental, articulatory, structural, or rhythmic impairments that could alter tongue function were selected as subjects. Totally 16 males and 16 females spanning young adulthood from 19 to 27 years of age were included in this study. Each individual was asked to pronounce each vowel according to the word on the paper shown, and to read each [a], [e], [i], [o] and [u] three times in their habitual phonation. A 3.5 MHz mechanical sector transducer was used as the ultrasonic scanning probe under CST condition, to capture mid-sagittal images of the tongue during vowel articulation. The images, observed in ultrasonic B+M mode, were recorded with a high quality video recorder and stored in computer through a frame grabber card for analysis. The results show that the average duration of tongue movement in vowel articulation by normal male adults was [a], 2.31 ± 0.39; [e], 2.32 ± 0.41; [i], 2.25 ± 0.46; [o], 2.48 ± 0.34; and [u], 2.37 ± 0.39 seconds. The average amplitude of tongue movement in vowel articulation by normal male adults was [a], 9.13 ± 3.21; [e], 5.48 ± 1.78; [i], 6.02 ± 2.92; [o], 13.26 ± 4.60; and [u], 7.17 ± 2.43 mm. The average duration of tongue movement in vowel articulation by normal female adults was [a], 2.47 ± 0.50; [e], 2.63 ± 0.41; [i], 2.52 ± 0.45; [o], 2.44 ± 0.34; and [u], 2.51 ± 0.47 seconds. The average amplitude of tongue movement in vowels articulation by normal female adults was [a], 6.29 ± 2.07; [e], 8.96 ± 3.63; [i], 7.22 ± 3.98; [o], 9.25 ± 3.97; and [u], 6.04 ± 2.57 mm. Real time mid-sagittal ultrasound images of tongue, with the help of the B+M mode, make the oral ultrasound investigation a more powerful and more precisely diagnostic method in phonology and speech pathology.

並列關鍵字

Tongue ； Vowel ； Ultrasound

參考文獻

Baer T, Ggore JC, Boyce S and Nye PW (1987). Application of MRI to analysis of speech production. Magn Reson Imaging 5:1-7.

Baer T, Ggore JC, Gracco LC and Nye PW (1991). Analysis of vocal tract shape and dimensions using magnetic resonance imaging: Vowels. J Acoust Soc Am 90:799-828.

Beach JL and Kelsey CA (1969). Ultrasonic Doppler monitoring of vocafold velocity and displacement. J Acoust Soc Am 46(4):1045-7.

Bell-Berti F (1976). An electromyographic study of velopharyngeal function in speech. J Speech Hear Res 19(2):225-40.

Boyce S and Espy-wilson CY (1997). Coarticulatory stability in American English /r/. J Acoust Soc Am 101(6):3741-53.