Eric Javel1, Iain L Grant, Kai Kroll. 1. Department of Otolaryngology, University of Minnesota Medical School, Minneapolis, MN 55455, USA. javel001@umn.edu
Abstract
BACKGROUND: Piezoelectric bimorph transducers may be used at the input stage of implantable hearing aids to convert ossicle vibrations into electrical waveforms, and at the output stage to convert electrical signals into mechanical motion that drives the ossicles. This study assessed transducer performance in anesthetized, acutely implanted cats using computer-averaged, laser-Doppler vibrometer measures and cochlear potentials. METHODS: Measures of output linearity and distortion for a transducer placed on the umbo were obtained from averaged laser-Doppler vibrometer outputs. Frequency response and equivalent sound pressure level for transducers placed against the stapes were estimated by comparing compound action potentials and cochlear microphonics elicited preoperatively by acoustic signals with responses elicited postoperatively by signals presented through transducers. RESULTS: The transducer placed on the umbo exhibited an effective bandwidth that exceeded 8 kHz, linear response behavior for driving voltages up to 2 Vrms, and harmonic distortion of -40 dB or better at all frequencies greater than 250 Hz. Except for a shorter latency, transducer-elicited cochlear potentials were indistinguishable from acoustically elicited responses. Frequency response varied widely across transducers, ranging from reasonably flat to possessing a bandpass characteristic with a peak at 2 to 4 kHz; 1-Vrms signals applied to transducers with various geometries yielded equivalent intensities of 62 to 108 dB sound pressure level at 4 kHz, 51 to 98 dB sound pressure level at 2 kHz, and 55 to 80 dB sound pressure level at 1 kHz. Differences in frequency response and equivalent sound pressure level stemmed from different resonance frequencies in transducers with dissimilar lengths and, more importantly, from variation in transducer-stapes contact force. CONCLUSIONS: Appropriately designed piezoelectric transducers can provide the cochlea with high-fidelity, wide-bandwidth signals. However, using them in implantable hearing aids requires that geometry and contact force be optimized to reduce variability in output level. Recording cochlear potentials is a cost-effective means of assessing transducer performance intraoperatively, but care must be exercised to take into account any temporary, drill-induced sensitivity loss.
BACKGROUND: Piezoelectric bimorph transducers may be used at the input stage of implantable hearing aids to convert ossicle vibrations into electrical waveforms, and at the output stage to convert electrical signals into mechanical motion that drives the ossicles. This study assessed transducer performance in anesthetized, acutely implanted cats using computer-averaged, laser-Doppler vibrometer measures and cochlear potentials. METHODS: Measures of output linearity and distortion for a transducer placed on the umbo were obtained from averaged laser-Doppler vibrometer outputs. Frequency response and equivalent sound pressure level for transducers placed against the stapes were estimated by comparing compound action potentials and cochlear microphonics elicited preoperatively by acoustic signals with responses elicited postoperatively by signals presented through transducers. RESULTS: The transducer placed on the umbo exhibited an effective bandwidth that exceeded 8 kHz, linear response behavior for driving voltages up to 2 Vrms, and harmonic distortion of -40 dB or better at all frequencies greater than 250 Hz. Except for a shorter latency, transducer-elicited cochlear potentials were indistinguishable from acoustically elicited responses. Frequency response varied widely across transducers, ranging from reasonably flat to possessing a bandpass characteristic with a peak at 2 to 4 kHz; 1-Vrms signals applied to transducers with various geometries yielded equivalent intensities of 62 to 108 dB sound pressure level at 4 kHz, 51 to 98 dB sound pressure level at 2 kHz, and 55 to 80 dB sound pressure level at 1 kHz. Differences in frequency response and equivalent sound pressure level stemmed from different resonance frequencies in transducers with dissimilar lengths and, more importantly, from variation in transducer-stapes contact force. CONCLUSIONS: Appropriately designed piezoelectric transducers can provide the cochlea with high-fidelity, wide-bandwidth signals. However, using them in implantable hearing aids requires that geometry and contact force be optimized to reduce variability in output level. Recording cochlear potentials is a cost-effective means of assessing transducer performance intraoperatively, but care must be exercised to take into account any temporary, drill-induced sensitivity loss.