Graeme Clark Collection

Permanent URI for this collection

Search Results

Now showing 1 - 3 of 3
  • Item
    Thumbnail Image
    Future developments in speech processing for multichannel cochlear implants in children [Abstract]
    Dowell, Richard C. ; Dooley, G. ; McDermott, H. D. ; Blamey, P. ; McKay, C. ; Clark, Graeme M. ( 1992)
    The average speech perception score for adult implant patients is now about 60% on an open-set sentence test without lipreading. This is higher than the scores obtained by many profound and severe-to-profoundly impaired hearing aid users. This suggests that some hearing aid users, particularly those who use a hearing aid in one ear only, could benefit from a cochlear implant. As neither the implant nor the hearing aid will provide perfect speech recognition it is to be expected that this group should obtain maximum benefit by using the hearing aid in one ear together with the implant in the other ear. However, experience with this group of patients has shown that many people find the use of two independent devices unacceptable. Furthermore, perceptual interaction of the acoustic and electrical signals makes it desirable to be able to control the two outputs in a more co-ordinated way than is possible with two independent devices. Consequently, a "bimodal" speech processor has been developed with both acoustic and electrical outputs controlled from the same speech processing unit. Feature coding aspects of the implant processing have been applied to the acoustic signal in such a way as to enhance speech perception with the hearing aid and improve compatibility with the implant. Initial testing with the bimodal aid shows promise to help severely-to-profoundly impaired individuals. The device has also been useful as a research tool to investigate the complex interactions of simultaneous acoustic and electrical stimulation. The Spectral Maxima Sound Processor (SMSP) has also been developed at the University of Melbourne for use with the Nucleus cochlear implant. Studies with adult subjects have shown improved perception of vowels, consonants, words and sentences in quiet and sentences in background noise with the SMSP as compared with the MSP(MULTIPEAK) which is currently supplied for use with this implant. Results for four subjects showed mean scores for open set sentences at a 10 dB signal-to-noise ratio of 78.7% for the SMSP and 50.0% for the MSP. Mean scores for the same group on open set monosyllabic words in quiet were 57.4% for SMSP and 39.9% for MSP. These results suggest that future improvements in speech perception will be possible for children using the Nucleus cochlear implant.
  • Item
    Thumbnail Image
    Factors affecting speech perceptual performance for children using the 22-electrode cochlear prosthesis [Abstract]
    Dowell, Richard C. ; Blamey, Peter J. ; Clark, Graeme M. ( 1992)
    Speech perception results for all 40 children and adolescents implanted with the. Nucleus 22 electrode cochlear prosthesis in Melbourne (as of February 1, 1992) were used to categorize performance for each child into one of six hierarchical groups: 1: detection of speech including high frequencies 2: discrimination of suprasegmental features of speech in addition to 1,3: discrimination and recognition of .vowel sounds in addition to 1 and 2,.4: discrimination arid recognition of consonant sounds in addition to 1, 2 and 3,5: open set speech. recognition with scores less than 20% for unfamiliar material in addition to 1 to 4,6: open set speech recognition with scores greater than 20% for unfamiliar material in addition to 1 to 5, above.All children demonstrated discrimination of suprasegmentals (level 2) and 58% demonstrated some openset speech recognition (levels 5 and 6). The pattern of results suggested that children who can discriminate segmental features of speech tend to achieve open set speech perception after adequate experience with the prosthesis. The performance level, described above, was used as the dependent variable in a multiple regression analysis to assess the effect of various factors on speech perception performance. The duration of profound hearing impairment and the amount of experience with the prosthesis were shown to contribute significantly to the variance, in performance level. A weaker trend was evident (or recently implanted children which may suggest that those' in oral/aural educational settings progress more rapidly, in terms of speech perception, than those in total communication settings. Age at implantation, cause of deafness, hearing levels prior to implantation, and number of electrodes in use, did not contribute significantly. to the variance in speech perception performance for these children. The results showed that all children with less than seven years of profound hearing impairment and with over one year of experience with the prosthesis have achieved some open-set speech recognition. This is an encouraging result as the trend in clinical application of cochlear prostheses has been towards implanting younger children in recent years. If this sample of hearing-impaired children in Melbourne is representative of the general population, we may expect that most implanted young children will ,develop reasonable speech perception skills after adequate experience and training.
  • Item
    Thumbnail Image
    Cochlear implantation in young children: long-term effects of implantation on normal hair cells and spiral ganglion cells in the monkey model [Abstract]
    Burton, Martin J. ; Shepherd, R. K. ; Xu, S-A. ; Clark, Graeme M. ( 1992)
    Recent independent results obtained by profoundly deaf children implanted with the Melbourne 22-channel cochlear implant (1) have provided further impetus. for examining thefeasibility of implanting children under two and children with profound deafness. Safety st1,ldies, in appropriate animal models, must first establish the safety of this procedure.