Towards the development of a prognostic tool for cochlear implant outcome using functional near-infrared spectroscopy (fNIRS).

Background: In the UK, four babies are born deaf each day. Children with hearing loss not only have delayed speech and language development, but also have lower educational achievements compared with children who have normal hearing. A cochlear implant is a device that can restore a sensation of hearing to children who are born or become deaf. All newborns are now tested for a hearing loss. Subsequently, children who are suitable for a cochlear implant can be identified in the first few days or months of life.

Overall, speech understanding in children improves after cochlear implantation. However, some children's speech and language abilities are much worse than we would otherwise expect and we don't fully understand why this happens. The current tests for assessing levels of hearing and speech understanding are unreliable in very young children. Since children that are born deaf often receive their cochlear implants within the first year or two of life, years can often pass before parents and healthcare professionals become aware of poor speech and language skills. Currently, it is extremely difficult to distinguish between very young children with cochlear implants who are performing well and those who are performing not so well. We want to understand why some children can hear well with a cochlear implant and others cannot. We would also like to predict and identify at the earliest possible stage those children with a cochlear implant who have poor speech understanding.

Aims of this research: At the Nottingham Hearing Biomedical Research Centre, we propose to use a non-invasive brain scanning method called functional near-infrared spectroscopy, or fNIRS for short, to measure brain activity in deaf children before and immediately after they receive a cochlear implant. We want to know if this brain scanning technique can be used to test how well a child can hear and understand speech instead of having to rely on existing hearing tests that are only suitable for older children. Although we have considered other methods for measuring brain activity, they are either not safe for use in patients with a cochlear implant or are associated with potential harmful effects. fNIRS is completely safe for children both before and after cochlear implantation and does not have any side effects or risks to health.

Expected benefits of this research: If we are able to assess how well a child can hear using fNIRS, clinical professionals could measure speech abilities in much younger children than is presently possible. Subsequently, we will be able to identify and treat children who are not hearing so well at the earliest age. In so doing, we will identify those children with a cochlear implant who struggle with their hearing and need extra speech and language support. At present, without any idea of the abilities of our very young children with cochlear implants, valuable NHS resources for speech and language support are provided to every single child, so that some children may receive more support than they require, whilst others receive too little. fNIRS may help us to tailor and more appropriately direct speech and language support to those children who need it the most. We would also be able to give parents a more accurate explanation of how likely their child is to improve with their cochlear implant.

Cochlear implants also need to be 'fine tuned' or programmed regularly so that they provide the best possible level of hearing to meet the needs of an individual. fNIRS has the potential to guide and improve this programming process. This is because it may be able to inform us on how well speech and sound is understood by the brain, years before a child is old enough to tell us, and enable us to make the appropriate adjustments to their cochlear implant. We believe that fNIRS has the potential to allow every child with a cochlear implant to have the best possible treatment that is tailored to their individual needs.

Background: Cochlear implants (CIs) allow many children to develop speech at an age-appropriate level when deafness is recognised early in life. However, the ability to perceive speech in children with CIs still varies considerably. Early identification of children at risk of developing poor speech perception is crucial to maximise the benefit of any rehabilitation approach. However, current measures of speech perception ability require behavioural assessments. Since very young children are difficult to assess with behavioural techniques, there is often a delay of many years between implantation and the detection of poor speech perception. In order to predict speech performance in children with a CI, it is necessary to find a suitable objective measurement of speech perception at an early age.
Aims & objectives: The aim of the proposed work is to evaluate whether we can predict good and poorer CI performance using brain imaging before and shortly after cochlear implantation. We will measure brain activity in infants using fNIRS, an optical brain-imaging technique that is fully compatible with a CI. We hypothesise that the benefit of a CI will be associated with cortical plasticity to auditory and visual speech, which may form the basis for a prognostic marker of CI outcome.
Methodology: Paediatirc CI candidates (age at implantation 6 months to 2 years old) will be undergo fNIRS measurements i) 1 month before CI surgery ('pre-implantation'), ii) 1 month after activation of CI ('early post-implantation') iii) 12 months after activation of CI in addition to behavioural speech perception measurements ('follow up').
Scientific and medical opportunities of the study: A prognostic tool could better inform parents on the likely treatment outcome for their child and also enable clinicians to tailor rehabilitation and signal processing strategies to an individual patient's needs. This study will inform us of mechanisms of cortical plasticity in children after implantation.

The goal of this research is to predict and monitor speech outcomes in profoundly-deaf children with a cochlear implant (CI) using fNIRS, which is a relatively novel optically-based neuroimaging technique. An ideal prognostic tool would allow clinicians to record outcomes both before and after cochlear implantation in order to tailor rehabilitation and signal processing strategies to the needs of an individual child. Furthermore it would also inform parents on the likely outcome from their child's treatment and would monitor their progress during a critical speech and language development period following surgery. Unlike other imaging techniques, fNIRS satisfies most of these clinical requirements. Specifically, it is possible to safely measure brain function with fNIRS repeatedly in a child both before and after cochlear implantation and it is unaffected by electrical and magnetic artefacts associated with the implant. Compared with fMRI, fNIRS is quick to set-up and is relatively silent; therefore fNIRS measurements can be made in a sound attenuated or quiet room in a clinical outpatient setting using a flexible 'infant-friendly' recording paradigm that involves an awake child sitting on their parent's knee.

In the absence of a reliable prognostic measure of outcome, similar rehabilitation strategies are generically provided to all CI recipients. The cost of outpatient visits, including rehabilitation, is estimated to be over £11,000 for each CI recipient over the first four post-operative years, with an additional cost of approximately £600 per annum thereafter [1]. A clinical, speech perception measure of CI success would ensure that intensive and valuable NHS rehabilitation resources are better tailored and more appropriately directed to those children who need them the most. Once the appropriate predictive outcome has been determined, it should be possible to translate these results into a clinical tool. This is likely to take about 6-10 years to define the neuroimaging marker, automate acquisition and analysis of neuroimaging data and to modify techniques to suit children. Further clinical evaluation to assess cost and clinical effectiveness across multi-centres will also be required. Therefore we hope to be having an impact on clinical practice in this time.

A pre-operative diagnostic tool would be used:
(1) to inform pre-operative decisions to provide patients with realistic expectations of their outcome; (2) by clinicians to stratify training and rehabilitation. Additional post-operative measures will help to monitor the outcome of patients and may serve to increase the sensitivity and specificity of the prognostic indicator. Furthermore, in the longer-term such a tool might even be used to personalise the care of each patient through software changes to their signal processor and/or changes to their rehabilitation strategy. Specifically, accurate behavioural measures of hearing are difficult to obtain in individuals with complex needs and very young children, making it challenging for clinicians to tailor the signal from a CI (the processing strategy) to the needs of those individuals. This tool may provide a useful objective measure of audiometric thresholds in such individuals.

As the Research Fellow, this study would enable me to significantly advance my i) generic and applied research skills ii) research governance skills iii) communication and education skills and iv) professional skills. Specifically, further experience with experimental design and conduct, data interpretation and analysis, writing and presentation skills and effective networking and collaboration will further my progress towards the ultimate goal of becoming an independent researcher.

References:
1. Toner, J., et al., Criteria of candidacy for unilateral cochlear implantation in postlingually deafened adults - II: Cost-effectiveness analysis. Ear Hear, 2004. 25: p.336-360.