電腦輔助聽能訓練系統對於重度及極重度學語前聽力喪失的助聽器使用者語音聽辨力及主觀獲益的影響

Effect of Computer-Assisted Speech Training (CAST) on Speech Recognition and Subjective Benefit for Hearing Aid Users with Severe to Profound Pre-Lingual Hearing Loss

謝敦理(Duen-Lii Hsieh)

指導教授 ： 許權振 ； 劉殿楨

國立臺灣大學/醫學院/臨床醫學研究所/碩士(2008年)

https://doi.org/10.6342/NTU.2008.01469

摘要

研究背景: 聽障患者已在小聲或無聲的環境中許久，突然接觸聲音，是全新的體驗。儘管現今的助聽器比起以前已經有長足的進步，配戴之後的復健仍是非常重要的。電腦輔助聽能訓練系統 (computer-assisted speech training, CAST) 是近期發展出來的聽能訓練系統。這是由美國House Ear Institute研發出來的聽能訓練軟體，本來主要是針對人工耳蝸植入的患者所設計，病患只需將軟體安裝在電腦當中即可在家進行聽能訓練。與助聽器使用者相較，人工耳蝸使用者畢竟屬於少數，倘若此系統能夠應用於配戴助聽器患者的復健，將使患者可以在家進行聽能訓練，從而解決目前復健資源不足的問題。
研究方法: 以15名重度至極重度學語前失聰的聽障助聽器使用者為訓練對象。另有6名相同背景者作為對照組。所有受試者在訓練前均接受純音聽力檢查、語音聽辨力測驗，包括單字詞、雙字詞、母音聽辨、子音聽辨和聲調聽辨，及填寫主觀助益問卷COSI 。15名受試者將軟體安裝到自己的桌上型或筆記型電腦當中。受試者被要求以CAST每星期訓練3次，每次1小時，如此持續12週。此後在訓練4週，8週及12週後重複語音聽辨力測驗，訓練結束時再評估COSI。對照組也接受同樣的測驗和評估，比較訓練前後的結果及和對照組間的差異。
結果: 15名受試者中有12名完成為期12週的訓練和測驗，有3名於訓練期間退出。對照組6名皆完成4次測驗。比較實驗組和對照組的語音聽辨力成績，單字詞聽辨測驗方面，實驗組在訓練後8週的測驗成績優於訓練前，達到統計學顯著意義(P<0.0001, GEE model, SAS 9.1.3) ，但在訓練後12週則和訓練前沒有差別。子音聽辨測驗方面，實驗組在訓練後12週的測驗成績優於訓練前，達到統計學顯著意義(P=0.0139, GEE model, SAS 9.1.3) 。雙字詞聽辨測驗、母音聽辨測驗和聲調聽辨測驗，實驗組和對照組均和訓練前無差別(GEE model, SAS 9.1.3) 。COSI 問卷方面，實驗組improvement score 平均為12.67，final score 平均為14.83，對照組improvement score 平均為11，final score 平均為14.83，兩者未達統計學顯著意義(multiple linear regression, SAS 9.1.3) 。
結論: 使用CAST系統訓練重度至極重度學語前失聰聽障助聽器使用者，在單字詞及子音聽辨測驗方面有進步。主觀獲益問卷方面，雖然實驗組普遍認為有進步，但和對照組比較仍未達統計學顯著意義。

關鍵字

電腦輔助聽能訓練系統 ； 助聽器 ； 人工耳蝸

並列摘要

Background: Hearing impairment is a common complaint in ENT clinic, especially sensorineural hearing loss. If the hearing loss is mild, closely followup of the hearing is suggested. If the hearing loss is too severe to communicate, hearing aid would be suggested. However, the efficacy of speech recognition with hearing aid varies from one to another. Therefore, the rehabilitation after wearing a hearing aid is important. Currently, the rehabilitation resources for hearing aid users are far from sufficient. Thus, in addition to develop new generation hearing aids, seeking for better rehabilitation quality is another important issue.

Purpose: Computer-assisted speech training (CAST) is a speech training system developed by House Ear Institute for cochlear implant users. The training targets on vowel, consonant, tone and pure tone can be performed after the software program is installed into a personal computer. The supervisors can monitor the time and score of self testing on internet. The purpose of this study is to apply this system to hearing aid users with prelingual severe to profound hearing loss. If positive results can be achieved, there will be a promising method of rehabilitation for hearing aid users to improve speech performance.

Methods: Fifteen hearing aid users with prelingual severe to profound hearing loss were included into this study. Another 6 hearing aid users with similar background were included for control group. Pure tone audiogram, speech recognition tests including Mandarin monosyllabic word, Mandarin disyllabic spondee-word, vowel, consonant and Chinese tone recognition were performed before training. Subjective benefit questionnaire, or client oriented score of improvement (COSI) was filled out. The software was installed into the personal computer and the subjects were asked to perform the training program 3 days a week, 1 hour per day, for a period of 12 weeks. Every 4 weeks the subjects were called back for speech recognition tests. COSI was completed after the whole training period. The control group received speech recognition tests and filled out COSI without performing the training program. Compare the scores of the tests before and after the training program and the difference between subject and control groups.

Result: Twelve of 15 subjects completed the twelve-week training period, speech recognition tests and COSI whereas 3 of them dropped out due to poor compliance. Tone recognition test was completed in only 10 subjects because this test was not introduced in the very first of the study. All 6 people in the control group completed four speech recognition tests and COSI. Results showed the training group had significant improvement in monosyllabic word recognition test after training 8 weeks with CAST (p < 0.0001). However, the improvement failed to present after training 12 weeks. In addition, the training group also had significant improvement in consonant recognition test after training 12 weeks with CAST (p = 0.0139). Otherwise, no significant improvement was noted in disyllabic spondee-word, vowel and tone recognition tests, comparing with pre-training scores. The result of COSI showed higher improvement score in training group but didn’t have statistical significance. The final score of COSI between training and control groups didn’t show statistical difference, either. The data of speech recognition test was analyzed by multiple marginal regression models for repeatedly measured response variable using the generalized estimation equations (GEE) method. The data of COSI was analyzed by multiple linear regression model. These regression analyses were performed using the SAS statistical software (Version 9.1.3, SAS Institute Inc., Cary, NC, USA). Two-sided p < 0.05 was considered statistically significant.

Conclusion: The result showed computer assisted speech training (CAST) system have some benefit in training hearing aid users with prelingual severe to profound hearing loss. However, the effect is not significant compared with previous studies. Thus, when applying CAST to this kind of hearing aid users, carefully assess the functional gain of hearing aid and make sure the compliance is good should be mandatory.

並列關鍵字

CAST ； hearing aid ； cochlear implant

參考文獻

1. Burk MH, Humes LE, Amos NE, Strauser LE: Effect of training on word-recognition performance in noise for young normal hearing and older hearing-impaired listeners. Ear Hear 2006;27:263-278.

2. Busby PA, Roberts SA, Tong YC, Clark GM: Results of speech perception and speech production training for three prelingually deaf patients using a multiple-electrode cochlear implant. Br J Audiol 1991;25:291-302.

4. Dawson PW, Clark GM: Changes in synthetic and natural vowel perception after specific training for congenitally deafened patients using a multichannel cochlear implant. Ear Hear 1997;18:488-501.

6. Effects of computer-assistive speech training (CAST) on mandarin speaking hearing impaired children: a preliminary report. Abstract presented in the 5th Asian Pacific Congress of Cochlear Implant and Related Science. Hong Kong 2005.

7. Fu QJ, Galvin J, Wang X, Nogaki G: Effects of auditory training on adult cochlear implant patients: a preliminary report. Cochlear Implants International 2004;5 Suppl 1:84-90.