CI Research

CI Research Study

Shortly after my January 17 blog post I was contacted by some folks involved in CI research. It seems my interest at comparing and gauging my pitch offset between CI-ear and normal ear, and my experienced attention to sonic detail, could be handy in a research setting. I was asked to be a test subject and I complied like a good Borg. Within a week we had scheduled a one-day trip to the Cochlear Implant Laboratory at Arizona State University.

On the day in mid-February, I had a 6:30 am flight out. So it was up at 4:00; to the airport about 5:00; hit the security area about 5:15 – my first time through with a CI. I had brought along my spare processor and I remembered reading that it should not be x-rayed (something about x-rays damaging the microphone element). After filling 4 trays with shoes, belt, pocket contents, laptop, and other contents from my bag, I pulled out the spare processor and flagged a TSA person. Upon explaining the x-ray issue, and showing my handy little CI card provided by Advanced Bionics, I was rewarded with a trip to secondary inspection so my processor could be tested for explosive residue. It only took a moment and the TSA people were friendly. I quickly re-robed and repacked and headed up to the departure area.

I located the gate for my flight and slid into a nearby seat. Most people were deep in their laptops, cell phones or mp3 players, so I felt a little the Luddite cracking open a hardback book. I mostly read, but also spent some time listening to PA announcements and even did a little eavesdropping – both of which were surprisingly successful with my decidedly non-Luddite CI.

At about 6:15 we got the call to board. It was a sold-out flight, but even with that we were ready to go on-time; something else wasn’t however so we were held on the ground for another half-hour. The pilot explained something about the delay but the crappy quality of the aircraft’s intercom put the answer just out of my reach. I could have asked my seat-mate, but that would have opened the door to an entire flight of conversation and I had reading to do; plus old hard-of-hearing anti-social habits die hard. The pilot made up some time in the air and we touched down only about 20 minutes late. I grabbed a cab and was deposited at ASU a bit after 9:00.

The CI Lab is in a shiny new building at the edge of the campus. I made my way up to the lab and was greeted in the corridor by Dr. Michael Dorman, the head of the lab. We stepped into his office and began chatting about CI’s. Dr. Dorman has been involved in CI research for quite a long time. While I was digging up CI research articles over the past few months I ran across many on which he was co-author. Soon we were joined by Dr. Tony Spahr, who would be directing my activities for the day. The nut of the conversation was that research often is full of promising ideas that go nowhere; occasional lines that lead to minor improvements; and the rare find that can be significant. ASU and Advanced Bionics are in the process of studying some new techniques to learn if they warrant testing in a full stimulation strategy and can ultimately be implemented with a software upgrade.

The focus of the current study is testing the effects of different electrical stimulation patterns on tonal quality. It is hoped that developing full processing strategies that deliver purer tones will improve overall perceived sound quality, aiding both speech discrimination and music appreciation. But the first step is finding a way to improve individual tones.

Most current CI strategies deliver sound to the cochlea as sequential pulses, hundreds to thousands of times per second. For a complex source sound, most of the electrode contact points on the array will be fired to convey pieces of the entire range of frequencies in the source. When a CI user listens to a pure tone, such as a tuning fork or sine wave tone, the pulse output goes to a small group of electrodes sequentially, with the center of the group at the highest amplitude (loudness). In lab testing, a single electrode can be pulsed. In either case, the sound perceived is not a pure tone – it is accompanied to a degree by the noise and distortion of unwanted frequencies on either side of the actual tone. It can be described as slightly fuzzy or buzzy. This may be, at least partially, due to the fatness of the electrical stimulation pattern of each pulse delivered to the cochlea affecting an area larger than the target. Speech is made up of relatively noisy components, so the effect on basic speech discrimination is likely not large. Where it may make a difference is in more difficult listening situations such as noisy environments. But the tonal quality of speech – its timbre – and of course the quality of music should improve with a more accurate tonal representation.

Most of my day at the lab was spent repeating a series of tone quality rating tests, the balance undergoing standard audiology tests for my test subject record. For the tone quality tests Dr. Spahr set me up at a test station comprised of a laptop, interface box and headpiece. A software application ran test routines which issued pure tones delivered by both standard and experimental electrical stimulation patterns. My job was to click a button to play a tone and then enter a value from 0 to 10 as a rating (0 – noise, 10 – pure tone). This would repeat about 40 to 50 times, usually randomizing 4 to 6 different tone/strategy combinations. At first it was hard to establish a reference point for myself, and judge the range of quality between various tones, but after a few times through it got easier. It was a blind test of course – I had no idea what method was used for any particular tone – but I was often detecting a fairly significant difference. On average I was rating the best as an 8, the worst as a 6. Some tone sequences the difference was more noticeable, some less. The largest spread in any test was 5 to 9, the least 7 to 8. For me, a 9 was very close to a pure tone, a 5 was a quite fuzzy tone.

At the end of the day I learned that my tone quality rating test results reinforced findings with other test subjects, and that the tones I was rating highly were the result of some of the new strategies being studied. There is no guarantee it will work well in a full sound processing strategy, but even so it is very encouraging. I was glad to be of help and spend time with individuals that are working to improve the technology I now enjoy. It was a pleasure meeting and working with them.

Research Conference

The 10^th International Conference on Cochlear Implants and Other Implantable Auditory Technologies will be held in San Diego in early April. Dr. Spahr is scheduled to give a presentation, “Perception of monopolar & tripolar stimulation by cochlear implant listeners.” The Agenda page on the website has a full listing of presentations – and there are a lot. It is an interesting window into the breadth of ongoing CI research.

A note on CI Brands

I think it is important to point out that this research is specific to the Advanced Bionics CII and HiRes90k implants. At least some of the strategies under study depend on the capability to fire multiple electrodes simultaneously, which is possible since these specific implants have an individual independent output circuit for each of 16 electrodes.

Every CI candidate has to decide what brand of implant to get. In the U.S. we have 3 choices: Advanced Bionics, Cochlear, and Med-El. It can be a difficult decision because there are many things to consider: technology; reliability; surgeon and audiologist available familiar with the brand; upgradeability, etc. But when it comes to the sound quality there is no way to do a comparison, one has to make that part of the choice on anecdotal information on blogs and other resources. And it seems everyone’s ears are different in how well they accept and function with a CI. Fortunately, all the brands seem to provide a similar level of performance. So with current publicly available stimulation strategies most users would likely be happy with any choice. However, when considering advances that might be coming, choosing the Advanced Bionics implant leaves the door open to more options. The Cochlear brand has a very good record of maintaining compatibility between their implants and upgrades to the external processor, but the implant has only a single output circuit that does not allow for simultaneous electrode firing. While this does not rule out improved stimulation strategies based on new sequential electrode firing techniques, it does rule out the use of any new strategies that require simultaneous electrode firing. The un-tapped potential of the Advanced Bionics implant is one of the main reasons why I chose it.