Audiology and Speech Pathology - Theses
Permanent URI for this collection
Search Results
1 - 7 of 7
-
ItemAn evidence-based pathway to early cochlear implantation in infants: reducing the reliance on the behavioural audiogram( 2023-03)ABSTRACT Background and Aims Cochlear implantation has emerged as a leading hearing intervention for infants born with severe to profound hearing loss. Research has shown that early intervention with a cochlear implant (CI) can significantly improve speech perception and oral language development in infants, providing them with access to the sounds that are critical for language acquisition. However, studies investigating the relationship between initial electrophysiological hearing thresholds and subsequent behavioural thresholds have demonstrated uncertainty as to whether initial results are able to provide an accurate estimation of an infant’s functional hearing. This estimation must be accurate if a CI is to be considered before behavioural audiometry has been obtained, as CI surgery is likely to remove any remaining natural hearing. This study aims to explore the variance between initial electrophysiological thresholds and subsequent behavioural hearing thresholds in young infants with suspected severe to profound hearing loss in a clinical setting, to determine under which conditions initial electrophysiological results can be used to support a CI recommendation in the absence of behavioural audiology. It also aims to propose evidence-based recommendations for implementation of a paediatric CI candidacy pathway for cases in which reliance on behavioural audiometry can be reduced. Method A prospective cohort study of 63 infants with suspected severe to profound hearing loss, referred to the Cochlear Implant Clinic, Melbourne, before 12 months of age. Assessment protocol included auditory brainstem response, auditory steady-state response, otoacoustic emissions, tympanometry, cortical auditory evoked response, and behavioural audiometry. Data was explored using correlational analysis, linear regression, and clinical decision analysis. Results 90.8% of all individuals who met CI candidacy criteria with initial electrophysiological testing (3FAHL>85dB) also met criteria upon behavioural testing. This proportion increased to 95% once certain exclusion criteria were placed on the group, excluding children born prematurely and children with auditory neuropathy. This result indicates that, under certain conditions, initial electrophysiological thresholds may be reliable enough to form the basis of a CI recommendation. Sensitivity and specificity of diagnostic electrophysiological testing was also found to increase once exclusion criteria were applied. Conclusion For most infants undergoing CI candidacy evaluation, initial electrophysiological thresholds were an accurate reflection of their subsequent behavioural hearing thresholds. An alternative CI candidacy pathway is recommended for infants who show profound sensorineural hearing loss on initial electrophysiological testing in the absence of middle ear effusion or auditory neuropathy features, to provide CI recommendations before behavioural thresholds are obtained. This would lead to reduced timeframes between hearing loss identification, diagnosis, and implantation, which can improve oral speech and language outcomes.
-
ItemGoing digital: enabling caregivers to optimise the linguistic environment for children with significant hearing loss( 2019)Early auditory exposure and linguistic interactions are fundamental in shaping the trajectory of typical language development in hearing children. These early linguistic experiences occur within the context of adult-child exchanges, through daily experiences, care routines and play. For children with prelingual hearing loss, early access to sound through hearing devices enables them to participate in these early linguistic experiences and can mitigate the potential negative effects resulting from hearing loss. In addition to early device fitting, research has demonstrated a strong association between the manner, content and amount of spoken language which infants and toddlers are exposed to at home, to enhance their learning, and their later language outcomes. While there is growing research awareness with regard to measuring the amount of talk directed to children, establishing how best to provide feedback to caregivers on their language contributions to their children in the context of everyday life, has been a challenge in traditional service delivery models. To this end, automated speech and language processing technology, also known as the Language ENvironment Analysis (LENA) system, offers clinicians the possibility of measuring and imparting quantified information to parents about their child’s auditory-language environment. Additionally, the use of mobile applications has been gaining traction as an accessible method of promoting health. The Daily Vroom App (DVA) for example, was created as an outreach initiative to communicate developmental research evidence to parents, presented as practical and implementable content. The DVA content takes the form of daily tips aimed at turning everyday moments for parents and their young children into opportunities for shared interaction. This study evaluated these two technology-driven interventions, using LENA technology and the Daily Vroom App, as ways to increase the amount of conversational turns and words spoken by adults to children. The present study firstly investigated relationships between family, maternal and child characteristics and language counts (measured using the LENA system) recorded in the children’s natural home environments. Secondly, changes in the participants’ adult word count and adult-child conversational turn count from baseline were examined after mothers received feedback based on LENA generated language estimates and/or, after mothers had used the Daily Vroom App. Thirdly, survey findings regarding the use of LENA technology and the DVA from the perspective of clinicians and mothers were explored. There were 10 mother-child dyads (six hearing children and four children with bilateral severe-to-profound hearing loss) who participated in a prospective intervention trial. After establishing each dyad’s baseline LENA-generated language counts (adult words and adult-child conversational turns), mean percentage changes in their language counts were analysed for three study conditions: LENA-Only recordings of adult-child verbal interactions in the child’s natural environments (Condition LO); adult-child verbal interactions following specific, individualised feedback from a clinician regarding the quantity of talk (LENA Feedback- Condition LF); and DVA use which provided tips for shared parent-child interactions in addition to Condition LF (LENA Feedback and App- Condition LFA). Maternal and child measures were administered at the beginning of the study. Demographic information and responses to a survey on the use of LENA and mobile technology were collected from participating clinicians and mothers. Key group findings suggested that the mean percentage increases in adult word and conversational turn counts from baseline were significant after the interventions (i.e. providing participants with LENA feedback and mother’s use of the mobile app), compared to Condition LO where only LENA recordings were completed. Significant correlations were found between mothers’ occupational skill level, maternal receptive vocabulary and their ratings of eHealth literacy. Participants assigned the DVA an overall mean quality rating of 3.85 out of 5. Significant positive relationships were found between objective app quality indicators (engagement, aesthetics, information) and participants’ perceived impact (on help-seeking, intention to change and behaviour change) of using the DVA. These two contemporary technologies (LENA and DVA) were demonstrated to support adult language input and conversational turns in the home environments for mothers of children with normal hearing, and children with bilateral significant hearing loss. This study addressed the question of whether providing structured information and feedback to mothers would be associated with a change in maternal language behaviours. This study also contributed to current knowledge regarding the practical applications of the LENA system and the Daily Vroom App as a complement to current service models, extending intervention practices beyond the therapy room.
-
ItemA study of electrical stimulation levels over 10 years for adults using Nucleus cochlear implants( 2017)Cochlear implants have improved sound perception for thousands of people with severe and profound sensorineural hearing loss. To ensure a good quality sound signal, the implant must be individually programmed throughout the user’s lifetime. Programming determines the electrical stimulation level requirements for each electrode of the implant. The frequency with which programming occurs has so far been based primarily on clinical experience and resources available to the clinic for programming purposes. To develop an evidence-based schedule for the frequency of programming, a comprehensive investigation of the change in electrical stimulation level requirements over time for a large group of adults is necessary. The overall purpose of this retrospective study was therefore to investigate the change in electrical stimulation levels up to 10 years postimplantation for a large group of adults using Nucleus cochlear implants. In addition, this study also aimed to investigate whether demographic characteristics and electrode array segments were predictive of the change in electrical stimulation levels. The Cochlear Implant Clinic of the Royal Victorian Eye and Ear Hospital, Victoria, Australia, has been providing programming services for cochlear implant users for over 25 years. The electrical stimulation data obtained at the programming sessions were extracted for 680 participants who used a Nucleus® cochlear implant over the course of 10 years. For each implant user, programming data for the following time points were extracted from Cochlear Limited’s TM Custom Sound 3.2® fitting software: 2, 3, 6, 9 months postimplantation and biennial time points from 1 year up to 10 years postimplantation. For each time point, the mean T level, C level, and dynamic range (DR) were calculated separately for four electrode array segments: apical (mean of electrodes 22, 20, 18), medial (16, 14, and 10) upper-basal (7, 6), and lower-basal (4, 3). The degree of change (DC) in levels between consecutive time points was also calculated. Long-term changes in electrical stimulation levels were also investigated for 128 participants from the same group who had adequate programming data up to 10 years postimplantation. The trends and amount of change in programming levels were investigated separately for the four electrode array segments using regression analyses. The effects of several demographic characteristics (e.g., aetiology and onset of hearing loss) were also investigated. For the 680 participants, the average T levels and C levels were consistent from 3 months and 6 months postimplantation, respectively. In terms of the degree of change between time points, the majority of participants showed an average of ≤ 20% change in levels as a function of DR after the 3- to 6-month time point comparison postimplantation. Long-term patterns of change for 128 individual participants showed that, for 42% of these participants, the electrical stimulation levels remained relatively stable without any significant trends for either the T levels or C levels, from 6 months up to 10 years postimplantation. Some significant changes in levels were evident up to 10 years postimplantation; however, changes were small, equating to less than 6% per year as a function of DR for 75% of the participants and a maximum of 10% change in levels for the remaining 25% of participants. Additional noteworthy findings include the following: 1. For the participants who showed more change in levels in the first 3 to 6 months postimplantation, approximately 70% of these participants showed a non-significant long-term trend in levels up to 10 years postimplantation, 2. The mean T levels and C levels were found to differ significantly for the four electrode array segments in the first 2 years postimplantation, but were relatively consistent in the long-term, and 3. The mean degree of change in levels as a function of DR was significantly greater for the Otosclerosis pathological group (n = 72) compared to all other groups, and for the prelingual onset of deafness group compared to the postlingual group. This thesis provided strong evidence that electrical stimulation levels change in the first 6 months but expected to remain relatively stable up to 10 years postimplantation for the majority of adults using Nucleus cochlear implants. Based on this evidence, a programming protocol for 10 years postimplantation has been proposed. The proposed protocol recommends more frequent programming sessions in the first 6 months compared to the number of sessions thereafter. The reduced number of sessions over an implant user’s lifetime will assist in managing the increasing caseload with the amount of resources currently available for programming. Given that the levels differed across the segments in the first 2 years postimplantation, but not in the long-term, the proposed protocol recommends measuring levels separately for the four segments in the early postimplantation period. This can then be limited to one or two segments beyond 2 years postimplantation. The protocol also suggests that more programming sessions may be required in the first two years for implant users with Otosclerosis and those with a prelingual onset of hearing loss.
-
ItemExploring the variance in cochlear implant outcomes as a function of information-processing ability( 2016)The modern cochlear implant (CI) is a medical prosthesis that restores to its hearing-impaired users the ability to hear speech. Most CI recipients are generally able to achieve a good level of speech understanding, but there remains substantial variability in overall speech perception outcomes that cannot be explained by demographic and audiologic factors alone. The purpose of this study was to explore top-down or cognitive processing ability as an additional source of variability in outcomes for postlingually-deafened adult CI users. Psychophysical measures of bottom-up processing ability were also used to assess the amount of useful speech information discriminable by CI recipients. Taken together, it was hypothesised that these measures would account for more of the overall variance in CI speech perception, adding to current knowledge about predictor variables that have traditionally been considered to be important for good auditory outcomes. Adult CI candidates were recruited prior to surgery from the Melbourne and Sydney Cochlear Implant Clinics (n=37) and assessed using a cognitive test battery. These tests were designed to assess domains related to working memory, processing speed, IQ, mental flexibility, and verbal learning. Bottom-up processing ability was assessed approximately three months after surgery using tests of gap detection threshold and electrode discrimination. Postoperative auditory outcomes were monosyllabic word perception scores at three months post-implantation in the monaural and binaural listening conditions. The findings from this study revealed that age at implantation was the most important factor in both listening conditions. In addition, information-processing speed, residual hearing in the contralateral ear, and binaural hearing aid use were also found to be significant factors for binaural speech perception outcomes. Together, the four factors in the binaural model accounted for approximately 48% of the variance in binaural word perception scores, while the monaural model accounted for approximately 16% of the variance in word perception scores using the CI alone. Gap detection threshold was also found to be significantly correlated with speech perception outcomes, indicating that temporal resolution ability is important for successful speech recognition using the CI. These findings have implications for future clinical practice, adding further knowledge to the evidence-based guidelines that inform clinical decision-making and CI candidacy criteria. Most importantly, the significant finding of age at implantation strongly suggests that CI recipients will have better outcomes if they are implanted earlier in the progression of their hearing loss.
-
ItemFactors influencing post-implantation hearing ability of adult cochlear implant candidates with pre-operative acoustic hearing( 2016)It is becoming increasingly common for adults with significant residual acoustic hearing in one or both ears to attend cochlear implant centres for candidature assessment. The clinical decision-making and provision of recommendations regarding this candidate population is complex, because there are only limited clinical data currently available to assist with counselling of candidates regarding expected outcomes. Pre- and post-implantation (3, 6 and 12-month) measures were obtained for 65 subjects who were implanted unilaterally in the poorer ear, and who had significant pre-operative acoustic hearing in one or both ears. This research first investigated outcomes in relation to the potential risks associated with implantation of such candidates; (1) the risk to loss of acoustic hearing in the implanted ear for candidates with pre-operative bilateral acoustic hearing, (2) the potential for less post-implantation benefit in candidates with better contralateral hearing, and (3) the unknown influence of predictive factors in this expanded population. In (3) were included the influence of demographic and hearing configuration, as well as an aspect of post-operative device fitting relating to the degree of match between the assigned acoustic input frequency in the sound processor and the electrical pitch percept. Finally, the factors that influenced the electrical pitch percept were examined. Significant group mean post-implantation improvement was observed on a sentence-in-coincident babble test and for spatial release from masking advantage when the noise location favoured the implanted side. Self-reported group ratings of functional benefit were improved post-implantation for all questionnaires administered. Objective assessment revealed poorer group mean localisation ability post-operatively for subjects with pre-operative bilateral acoustic hearing, however subjective ratings in real-world environments were more variable. Degree of post-implantation SRT benefit on the coincident babble test and on self-reported ratings of perception of soft speech and sounds in the environment was greater for subjects with less contralateral hearing. Factors that predicted better unilateral word scores in quiet (and accounted for 34.1% of the variance) were a shorter duration of severe-to-profound hearing loss in the implanted ear and poorer pure-tone-averaged thresholds in the contralateral ear. A shorter duration of severe-to-profound hearing loss in the implanted ear, a lower age at the time of implantation, and better contralateral hearing thresholds were associated with higher bilateral word recognition in quiet and SRT in noise (and accounted for 36.0% and 30.9% of the variance respectively). Degree of match between the initial (pre-activation) pitch percept elicited from stimulation on the most apical channel and the assigned frequency affected rate of post-operative improvement in the unilateral but not bimodal device configuration. Post-experience measures revealed higher unilateral word in quiet and sentence in babble scores for those subjects with closer match between the assigned acoustic input frequency and electrical pitch percept. No significant correlation was observed between degree of match and bilateral speech recognition ability. Subjects with poorer hearing in the implanted ear tended to have a lower pitch and a shallow electrical pitch function than predicted by the spiral ganglion frequency-position model. There was no significant group effect of listening experience.
-
ItemCommunication outcomes for children who receive a cochlear implant before 2 1/2 years of age( 2015)Background and aims: This study examined the long-term benefits of cochlear implantation for children who received their first cochlear implant before 2.5 years of age. Using a comprehensive battery of speech perception, speech production and language measures over six years, the study compared communication outcomes for children using cochlear implants and children with normal hearing. The relative influence of age at cochlear implantation and other covariates on outcome measures was examined. Method: The study design involved the prospective longitudinal evaluation of 32 children with no additional disabilities, who received their first cochlear implant before 2.5 years. Speech perception skills were quantified using the CAP pre-operatively and at one, two and three years’ post-implantation. When the children entered primary/elementary school (five years of age), open-set word testing (CNC) was completed. Speech production was evaluated using the DEAP at school entry. Language was evaluated pre-implantation, at one, two and three years post-implantation, and at school entry using the RI-TLS, PPVT and the CELF or PLS when relevant. Results: All children in the study developed significant open-set speech perception skills with a group mean monosyllabic word score of 56% and phoneme score of 81%. Speech production skills were delayed compared to normally hearing peers at school entry. Rate of language development for the children (M 1.04) and standard scores (M 86) at school entry were equivalent to hearing peers. Younger age at implantation was found to significantly reduce language delay and was associated with optimum speech production at school entry. There was no significant relationship between age at implantation and language growth, language standard score at school entry or speech perception. Measures of family participation and child non-verbal IQ were significantly associated with speech perception, speech production and language. Conclusion: Children with cognitive development within the typical range who receive a cochlear implant before 2.5 years of age, can demonstrate speech perception skills sufficient to communicate using audition alone. This level of speech understanding however, does not lead to age appropriate speech production in all cases. In the present study, speech production was delayed compared to hearing peers. Children who receive their first cochlear implant before 2.5 years can demonstrate rates of language acquisition equal to their hearing peers, but may retain a language delay approximately equal to their age at implantation. This data supports the provision of cochlear implants as early as possible to minimise any language delay resulting from an initial period of auditory deprivation.
-
ItemOptimisation of rate-pitch perception in cochlear implant hearing( 2013)It is well established that Cochlear implants (CIs) enable moderate-to-profoundly deaf people to understand speech without the aid of lip reading. However, the perception of voice and musical pitch by implantees is far from satisfactory. The limitations stem from an inability to code and perceive detailed information about the fundamental frequency (F0) in signals, which in normal hearing provides the principle cue to pitch. Present CI systems instead focus on conveying information about the signal’s envelope. In these systems, F0-pitch (albeit somewhat weaker than that of normal hearing) is coded over a limited range by way of amplitude modulation in the temporal (time) envelope. However, the depth and shape of this modulation is highly dependent on acoustic properties of the signal and characteristics of the sound coding strategy. Thus the salience and accuracy of coded F0 information can vary substantially across signals, acoustic environments, and coding techniques. Furthermore, variations in spectral (frequency) envelope information (e.g., differences amongst vowels) which produce changes in the place of electrical stimulation, can adversely affect CI recipients’ judgement of pitch derived from temporal envelope information. The main aims of the research were: (1) to develop a rate-pitch coding strategy that enhances the salience (pitch-strength) and accuracy (pitch-height) of coded F0 temporal information without adversely affecting recognition of speech; and (2) to explore novel methods in which individual’s ability to attend to F0 information exclusively as a cue to pitch in the presence of spectral envelope variations are improved through the use of a training program. To address the first aim, psychophysics experiments were conducted with CI recipients using modulated electrical pulse trains to examine the effect of modulation depth, rate, and shape on pitch-height and loudness relative to that of unmodulated electrical pulse trains. Unmodulated pulse trains were used as a reference because previous studies had shown that accurate identification of pitch intervals could be obtained on the basis of changes in pulse rate, at least for low rates. Results of the present psychophysics experiments demonstrated that deep sinusoidal amplitude modulation was needed to elicit an accurate percept of pitch for low F0 signals. For a sharper modulation function, similar performance could be obtained at a shallower modulation depth. However, application of deep or sharp modulation reduced loudness. These results, together with outcomes of previous studies, provided input necessary for the development of an experimental strategy (eTone) that enhanced coding of F0 modulation in the stimulus envelope. Pitch and speech perception using that experimental strategy were compared to that of the clinical Advanced Combinational Encoder (ACE) strategy in six adult CI recipients. Significant improvements in discrimination of pitch were observed with no reductions in recognition of speech in quiet or noisy conditions. The second aim of the research was to determine whether improvements in pitch perception could be obtained through specific training that primarily directed listeners to attend to F0 information exclusively as a cue to pitch, and to resonant frequency as a cue to spectral timbre. Outcomes demonstrated significant improvements in CI recipient’s F0 discrimination thresholds after training whereas no improvements were observed for control subjects that did not participate in training. This outcome generalised across different stimuli including natural sung vowels that embodied small variations in spectral timbre that could adversely affect judgement of F0 pitch. The improvement was also maintained several months after completion of training.