Fabrication and Testing of A Middle-Ear Implanted Microphone

Abstract

Cochlear implants are devices that can restore hearing to people with sensorineural deafness. Despite their name, cochlear implants rely on an external unit which contains components such as a microphone. This work presents the design, fabrication, and testing of an implantable middle-ear microphone called the “UmboMic” that measures the displacement of the tympanic membrane at the umbo. Particular consideration is paid to the biocompatability of the microphone and its long-term durability in the body. The work discusses biocompatible materials, methods of encapsulation, and techniques for testing device robustness.

The UmboMic is a piezoelectric displacement sensor that is implanted in the middle ear cavity and contacts the umbo. As the umbo moves, it displaces the UmboMic, resulting in a charge that is amplified with a custom amplifier. The active area of the UmboMic is a triangular shaped cantilever made from two layers of piezoelectric thin film called polyvinylidene fluoride (PVDF). The bimorph design reduces common mode noise as compared to our previous microphone designs.

Extensive bench testing and experiments in fresh human cadavers demonstrates excellent microphone performance despite the use of biocompatible materials. The UmboMic sensor is well shielded against electromagnetic interference, tolerant to implantation variations, and can be repeatably fabricated with little difference between sensor performances. It demonstrates high sensitivity from 100 Hz to above 8 kHz, with a sensitivity of 58 fC/Pa at 1 kHz and 230 fC/Pa at 2 kHz when including the outer ear. The noisefloor of the UmboMic normalized over 1/3 octave bins is 10⁻² fC, and the A-weighted equivalent input noise of the UmboMic with the outer ear is 82.4 dB SPL from 100 Hz to 7 kHz. When tested in five different human cadavers, the UmboMic sensors work reliably despite anatomical differences.

Internalizing the entire cochlear implant would greatly improve the quality of life of wearers. In its current form, cochlear implants cannot be used during sleep and vigorous activity, are susceptible to noise from wind, and function poorly in loud environments. Implanting the device would mitigate these problems and provide users with the discretion of an invisible device. Our prototype demonstrates the feasibility of an implanted microphone and is an important step towards developing a totally implantable cochlear implant.