List of questions
Is energy harvesting a possibility for powering/charging cochlear implants?
Cochlear implants today use externally worn rechargable batteries with an RF transcutaneous link to the implanted part. This is only 30% efficient. Future devices may move the battery inside the body, so different approaches to powering the system are interesting.
Which hermetic encapsulation materials might replace titanium packaging?
Today cochlear implants contain electronic circuitry and an RF coil to communicate with the outside world. This approach means the coil needs to be outside of the titanium package. An RF transparent encapulation material would be interesting to consider, although the circuity will need protecting from the body and from head trauma too.
How technology can help address the cocktail party problem
It is estimated that by 2050 over 900 million people – or one in every ten people – will have disabling hearing loss. Many of these people experience the cocktail party problem which makes social situations extremely difficult. Developments in technology can help these people but they have different requirements meaning that any solution needs to fit the users needs.
How to identify personalized bottle necks/barriers in the electrode neural-interface in CI.
It’s well known that CI recipients, usually achieve scores that are superior to their hearing aid scores, but there is a still a large spread of individual outcomes. Even when correcting for known factors such as duration of deafness etc a large unexplained spread remains. The first step into solving this issue (and bringing all recipients up to their maximum score) is identifying which factors determine the sub-optimal score in an individual. This could be related to neural survival, electrode placement, fitting issues, rehab, motivation, cognitive factors, or others? For optimizing treatments it’s essential to map out which factors play a role in any individual not reaching maximum speech perception. We would like to discuss ideas on how to start identifying these issues clinically.
How to treat personalized bottle necks/barriers in the electrode neural-interface in CI.
Once a clinically practical method to identify the personal factors is in place we will need a standard of care to treat each individual factor. Ideally we have the tools to setup an personalized program for each individual to help them achieve the maximum score. In one person this may be extensive fitting time, in another additional training, in yet another de-selection of ‘poor’ electrodes, etc. How will we come to a scheme like this provided we have the means to identify (see Q1).
Sensitivity of CI users to spectrally local loudness cues
In CI we adjust the ‘loudness’ per frequency channel by changing electrode stimulation levels per channel. A lot of time and care is usually dedicated to ensuring loudness balance between channels. With lateral wall placed electrodes and monopolar stimulation this usually leads to very flat responses across the array but exceptions can be found. With recent profile based fitting methods, there is the risk that individual channels may be over- or underfitted. This can be identified by running loudness balance across the channels, but this is a tedious and difficult task for users. With the improvement of electrodes we are (and will be) getting closer and closer to the neural elements and we start to see more local variation in thresholds (and C-levels), which means that, at least theoretically, this risk of under- or overfitting a channels increases. However, the question is how relevant this is. One can argue that individual loudness per channel does not matter in a world filled with largely broadband sounds. Maybe ensuring proper loudness scaling using spectrally wider signals (i.e. multiple electrodes bursts) is good enough, or even better. (for instance normal hearing will experience sharp frequency dips in rooms due to room modes/ resonances that do not affect our speech understanding much). How can we answer this question?
1. What are the challenges for the next generation of electrodes for cochlear implants? Miniaturisation? Materials? Customisation? Cost?
2. Is there a place for printed electronics in the design of new electrode arrays for cochlear implant market?
(1) What are the top ways AI will be applied to the current state of the art Cochlear Implants?
(a) Fitting dig data can be used to get generalized fittings with machine learning finding important factors. (b) outcome measures (e.g. HINT) are to be correlated with fitting data. (c) pre-processing algorithms and context aware systems (Hearing aid systems do this) (d) speech recognition and cochlear implants use cases (translation, commands).
(2) How can the rollout of IOT connectivity improve hearing devices in the next 5 years?
(a) Devices will be always connected over Or will they store and upload data (b) Gather big data on customer use (HAs already do this and FRP) (c) Device diagnostics: identify device failures either before they happen or very rapidly
(3) Acoustic hearing aids and to a lesser degree cochlear implants seem to be approaching a technological limit in terms of speech improvement over the last decades. Where are the next big steps in performance going to come from?
(a) Increasing effective channels (reducing current spread) is #1 problem holding CIs back. Current steering, phase arrays and other methods can move the site of maximum stimulation but not reduce the spread much. Need for new techniques. (b) Algorithms like the interleaved 'zebra' coding may allow the brain to do more of the processing.
(4) How are bio-technology (gene drives, stem cell research, pharmacology) and classical technology going to merge in the hearing field?
(a) Starts with drug eluting electrodes (but what next).