Electrodes restore hearing

Perhaps the best way to understand the importance of the latest technology for the hearing impaired is to read this article out loud.

If you can hear your own voice-or the lyrics from your favorite CD, or the laughter of your child-the quality of your life is miles ahead of the more than 4,000 people in the U.S. suffering from Neurofibromatosis Type 2 (NF2).

NF2 is a life-threatening genetic condition characterized by the growth of multiple tumors along the spinal cord and on the auditory nerves. For most victims, the only cure is to remove the tumors through surgery. But, removing them from the auditory nerves entails severing those nerves. The result: total deafness in both ears.

Attached to brain. Cochlea implants don't help those patients because the nerves the implants stimulate are no longer attached to the brain. What may help, however, is new technology known as auditory brainstem implants (ABI).

Developed at the Los Angeles-based House Ear Institute, and manufactured by Cochlear Corp., of Englewood, CO, the implants incorporate a new design in electrodes. Sound signals from the electrodes bypass the severed auditory nerves and travel directly to the brain.

The technology restores enough hearing to rid patients of the feeling of being cut off from the world around them, says Robert V. Shannon, Ph.D. and director of auditory implant research at the House Ear Institute.

"Patients can hear and modulate their own voices, and that helps them feel more in contact with the world," he says. They can also hear such things as a knock on the door, traffic, and running water.

And, with the help of lip reading, they can understand some amount of conversational speech.

Key to the technology is the design of the electrodes.

While the electrodes in a cochlea implant fit into the snail-shaped inner ear and stimulate a malfunctioning hearing nerve, the ABI electrodes sit on the jello-like, tapered-down brain stem near the top of the spinal cord.

The cochlea is encapsulated in bone. The brain stem is soft. Traditional electrodes could float around in it or actually sink through it. Engineers and researchers at the House Institute collaborated with colleagues at the Pasadena, CA-based Huntington Medical Research Institute to eliminate that risk by designing electrodes with a Dacron(TM)-mesh backing for stability.

Surgeons have implanted ABIs with the Dacron-enhanced electrodes in 27 patients since clinical trials began. "Our best patient understood 60- to 80% of speech without lip reading in a recent test," Shannon says.

Familiar packaging. Like cochlea implants, the packaging of ABIs consists of both external and internal parts. The external parts include a tiny headset containing a microphone and transmitter coil that fits behind the ear, and a speech-processing box the size of a cigarette pack that analyzes sound and translates it to electronic pulses. The external speech processor uses a single AA battery, which lasts through 10-12 hours of use. Internal parts should last for the patient's lifetime.

The pulses travel along the transmitter coil, which contains a small magnet that works with an internal magnet in the implanted receiver/stimulator to hold the transmitter firmly in place above and behind the ear. Also inside the skull is the array of eight 1-mm disk electrode contacts mounted on a 2mm x 8mm silastic substrate.

The electrodes attach to the brain stem at the lateral recess of the fourth ventricle, or channel where fluid circulates nutrients to the brain. That ventricle exits the brain at the point where the auditory nerves enter the brain.

"Since the brain stem is soft, any hard surface such as a platinum plate could cut into it or sink through it," says Shannon. He and his team took advantage of several years of National Institutes of Health-funded research at the Huntington Institute on surface cortical electrode arrays. The research showed that adding cloth to the electrodes gave them enough surface area so that the body would build a capsule around the electrodes, holding them in place.

"Fibrous tissues from the brain stem actually grow into our loose-weave Dacron fabric, developing a 0.5-mm-thick encapsulation," says Shannon.

The mesh is tacked at four corners to the silastic electrode carrier. Engineers at first tried casting it into the silastic, but surgeons thought that made the assembly too stiff to handle well during surgical insertion.

Tradeoffs. The choice of eight electrodes versus the 22 used in cochlea implants represents a tradeoff between the small size of the lateral recess of that fourth ventricle and the desire to stimulate as many regions as possible. Smaller electrodes might allow more stimulation points, but would produce a high charge density that could damage tissue, Shannon says.

The current ABI design is the culmination of more than 15 years of research. The first generation consisted of two ball electrodes and a speech processor that was a modified body-worn Bosch hearing aid. The first patient received auditory sensations daily for several months, but began to experience twitching in the leg on the same side of the body as the implant. Researchers assumed the electrodes migrated away from the cochlea nucleus area.

They then replaced the ball electrode with one that had the Dacron mesh backing and used the same Bosch hearing aid as the sound processor. Later refinements included:

Changing lead wires from single strand to braided, which is more flexible and more resistant to breakage

Location is everything. Much of the research involved finding the right place on the brain stem to implant the electrodes. Put them in the wrong place, and the electrodes could stimulate reactions other than hearing. The researchers found that the lateral recess of the fourth ventricle was the place that would produce the least amount of unintended reactions, but even there one or two of the electrodes could stimulate tingling in an arm or a feeling of jittering in the eye.

"We have taken a true engineering approach to solving a biomedical problem," says the House Ear Institute's Shannon. "And the process has been unique in the degree we have involved several disciplines, such as electrical engineering, physiology, surgery, anatomy, psychophysics, and psychology, in this important and vital solution."

Among the improvements they're working on now are modifying the algorithms for processing speech and developing penetrating electrodes, both of which could aid speech intelligibility.

One spinoff application for the technology could be pain control, though Shannon says that possibility is several years away from reality.

For now, Shannon and his team will be happy if the ABI produces the life-quality improvements early recipients have experienced.

"I want to hear, and anything would be better than total deafness," says Marilyn Davidson, the first to receive an ABI implant. "It was so strange to talk and not hear my own voice. With the implant, I could hear myself and other people's voices."

Adds college student Jenny Hendricks, "It helps me a lot in lip reading. When I go dancing, I can hear the beat of the music. I thought I would never go dancing again."

And, Chris Arcia, another college student, says that with the implants, "I know if someone in the room is speaking to me."

DSP improves hearing aid design

Hearing aids are remarkable devices, but they can degrade the wearer's ability to hear directionally and tune out interfering or distracting background noise. Work at the House Ear Institute, Minneapolis-based Starkey Laboratories, and QSound, Inc., Calgary, Canada, seeks to solve that problem with a binaural hearing aid.

Researchers have developed a method for measuring the arrival time and level of signals at the ear drum, and use the measurements to design a digital filter that corrects for hearing loss while preserving binaural time and sound level cues.

Traditional hearing aids have analog circuits. The key to the new binaural aids, says Sigfrid Soli, director of the program for the House Institute, is a new digital signal processor that can go inside the ear. Digital processing is required to control the time of arrival and level of the signal at different frequencies. Engineers are designing a digital circuit that requires only a 1.2V hearing aid battery. The circuitry will perform one million digital operations per second, which makes it relatively slow compared to the 75-100 million operations per second of 486 computer chips.

Researchers are now completing their third round of field tests. They expect commercialization sometime in 1996.