Frequency, place and temporal fine structure in cochlear implants

Reinhold Schatzer, Katrien Vermeire, Daniel Visser, Mathias Kals, Andreas Krenmayr, Clemens Zierhofer, C. Doppler Laboratory for Active Implantable Systems, Institute of Ion Physics and Applied Physics, University of Innsbruck, Austria Simulation studies in normal-hearing subjects have demonstrated the importance of fine structure (FS) for the perception of music, tonal languages and speech in competing-talker backgrounds. However, a persisting challenge in cochlear implants is how to represent fine structure information in a way that it can be perceived. Some clues to that extent may come (1) from the experience with combined electric-acoustic stimulation and (2) from pitch matching results in implant recipients with contralateral normal hearing who received a CI in the deaf ear to suppress intractable tinnitus. Experiences from combined electric-acoustic stimulation show that residual acoustic hearing at low frequencies often provides a significant benefit for the intelligibility of speech in noise if combined with electrical stimulation. These benefits may be largely due to the normal representation of low-frequency information in the F0 range. On the other hand, frequency-place maps for electrical stimulation derived from implant recipients with near-to-normal hearing in the contralateral ear suggest that place pitch resolution is leveling off in the apical region of the cochlea. In order to elicit low pitch percepts and perhaps better encode fine structure in the F0 range, a corresponding low-frequency temporal code is required at the right tonotopic place. Based on these findings, a coherent fine structure (CFS) coding strategy has been designed and evaluated so far in a small number of subjects. CFS derives a carrier signal c(t) in the F0 range containing explicit temporal fine structure information. This carrier signal is used to modulate CFS channels with slowly varying spectral envelopes in which temporal fluctuations related to F0 have been removed. Since c(t) is applied equally across all CFS channels, the resulting stimulation pattern shows bursts of pulses at the pitch period, appearing coherently across all CFS electrodes. Thus, the temporal pitch representation is not impaired by spatial channel interaction. In a processor implantation of this concept, apical channels are configured with CFS, while the remaining channels run on CIS. Preliminary results from acute comparisons to CIS show a comparable outcome for the intelligibility of sentences in a competing-talker background. Pitch discrimination of harmonic complex tones at fundamental frequencies ranging from 100 to 300 Hz was substantially better with CFS. Support provided by the C. Doppler Research Association

The contribution of temporal fine structure and envelope information to performance in multi-source, reverberant environments
Bernhard U. Seeber, MRC Institute of Hearing Research, Nottingham, UK Cochlear implants (CIs) often restore speech understanding in quiet, but most patients complain that the presence of reverberation or noise makes understanding of speech more difficult or even impossible. The precedence effect is thought to help in those situations as it shows the perceptual suppression of sound reflections while localization is maintained at the leading, the direct sound. This is thought to be based on the evaluation of temporal fine structure information which is not accurately transmitted by CIs. Because of the strong information transmission limits at the neural interface in CIs, knowledge of the essential information to function in complex environments is crucial for a targeted development of future CI-strategies. We have developed the Simulated Open Field Environment, a loudspeaker setup to reproduce the reverberant sound field of multiple sources in the free-field of an anechoic chamber, to investigate hearing mechanisms in difficult listening situations with normal-hearing participants and patients alike. In our study of the precedence effect about half of the tested bilateral CI-patients showed no precedence effect; instead a single sound source was localised in-between the leading and the lagging source. This suggests that those patients can not distinguish the binaural information in lead and lag, resulting in an average. However, selected patients showed the precedence effect even for temporally overlapping stimuli while instead the simulation of the paradigm with normal hearing listeners using a noise-band vocoder resulted in a breakdown such that the reflection was always audible. Subsequent studies were aimed at finding the information needed to evoke the precedence effect with diminished acoustic information as in CIs. Results show that it is possible to evoke the precedence effect with ongoing sounds without encoding ITDs in the carrier. The precedence effect can solely be based on interaural level and time differences conveyed in the envelope. However, carrier frequencies need to be similar in both ears, suggesting that accurate place matching of CIelectrodes would help the analysis of concurrent sounds. Further results suggest that envelope fluctuations are even needed to evoke the precedence effect.

Pitch scaling psychometrics in electric acoustic stimulation (EAS)

Uwe Baumann, Goethe-University of Frankfurt, Audiological Acoustics Combined electric acoustic stimulation (EAS) is a therapeutic option for patients with severe-toprofound high and mid frequency hearing loss but remaining low frequency hearing. Knowledge of electric and acoustic pitch transformation is very important for effective fitting which joins both kinds of stimulation into one perception. The fitting software of current cochlear implant (CI) sound processors requires that filter parameters be set to control the frequency range of acoustic and electric processing. obtained in CI recipients. To determine the electric place/pitch function in EAS, pitch adjustment experiments were conducted in 4 patients implanted with EAS and a FLEXEAS 24 mm electrode (MED-EL, Innsbruck) who had sufficient residual hearing in the opposite ear. Two subjects with single sided deafness and standard CI treatment with a PULSAR OR SONATA CI and a standard 31.7 mm electrode carrier (MED-EL, Innsbruck) served as controls. The patients task was to listen to a pure tone presented in the ear opposite the implant and then adjust a knob until the pure tone presented through a single electrode stimulation in the implanted ear matched the frequency of what was heard. Frequencies of the pure tones being presented to the non-implanted ear were randomized. Adjustments were repeated to ensure reproducibility and accuracy. Postoperative radiographs (modified Stenvers view) were analyzed to compare individual data in terms of insertion angle. Results showed considerable variation in terms of the individual pitch elicited by a certain electrode insertion angle. The lowest frequency adjustment in EAS patients was approximately 500 Hz, whereas control patients with deep electrode insertion made adjustments at or even below 200 Hz. Since the residual hearing in the EAS patient group was rather limited in the higher frequency hearing range, only a limited number of more basal electrodes could be assessed in order to determine the slope of the electrical place/pitch function. Slopes varied greatly between patients. The data was also analyzed in terms of the effect of start frequency and the effect of habituation to the electrical stimulation over time. Our outcomes suggest that in order to obtain reasonable default filter parameter settings for the acoustic/electric crossover frequency, postoperative audiograms of the implanted ear and Stenvers radiographic insertion depth control of the electrode array are necessary prerequisites. The position of the most apical electrode gives a good first approximation of the reference for setting of the crossover frequency. Recent data suggests that the frequency/place function proposed by Greenwood for normal hearing (Greenwood, JASA 1961) does not sufficiently explain results

SIMULTANEOUS PULSATILE STIMULATION WITH CHANNEL INTERACTION COMPENSATION: CONCEPT AND RESULTS
Mathias Kals, Reinhold Schatzer, Andreas Krenmayr, Katrien Vermeire, Matthias Zangerl, Daniel Visser, Paul Bader, Christian Neustetter, Clemens Zierhofer C. Doppler Laboratory for Active Implantable Systems, Institute of Ion Physics and Applied Physics, University of Innsbruck, Austria Recently, a method to compensate for the detrimental effects of electric potential summation with simultaneous pulsatile stimulation has been presented (Zierhofer, 2008). The proposed Channel Interaction Compensation (CIC) concept is based on an exponential model of spatial potential distribution within the scala tympani. Two constants, and , model the field decay towards the apex and base at each electrode. A third constant, , models the decay of placedependent field potentials from apex to base. If electrodes are activated simultaneously, CIC starts from the respective sequential pulse amplitudes and computes compensated simultaneous amplitudes by taking potential summation into account. CIC has been evaluated by measuring speech reception thresholds for sentences in speech shaped noise in seven subjects for various CIC settings and CIS as a control condition. CIC decay constants and number of simultaneous channels were varied systematically. Identical channel pulse rates and phase durations were used across conditions. Results demonstrate that simultaneous stimulation with CIC is feasible, supporting intelligibility of sentences in noise which does not differ significantly from sequential CIS stimulation. In some subjects, this is true even for settings with fully simultaneous stimulation on all available electrodes of the array. However, good speech recognition requires proper settings of decay constants. Without CIC compensation simultaneous stimulation significantly degrades performance in most subjects. Simultaneous stimulation with CIC may be applied to achieve substantially higher overall pulse rates for a precise temporal coding of fine structure information, or to increase pulse phase durations while at the same time reducing pulse amplitudes, lowering compliance voltage requirements. Support provided by the C. Doppler Research Association

Binaural Hearing Benefit in Cochlear Implant Users Binauraler Hrgewinn bei Cochlear-Implant-Trgern
Wolfgang H. Dring, HNO-Klinik des Universittsklinikums der RWTH-Aachen Einfhrung Das binaurale Hren erlaubt normalhrenden Personen ein besseres Richtungshren und Sprachverstehen in strschallerfllter Umgebung (Cocktail Party Effekt). Fr die meisten Cochlear-Implant (CI) -Trger stellen Alltags-Hrsituationen jedoch eine groe Herausforderung dar, auch wenn durch eine beidseitige Cochlea-Implantation versucht wird, zumindest eine partielle Nutzung von Fhigkeiten des binauralen Hrens zu ermglichen. Stand der Technik Aktuell werden nach bilateraler Implantation beide Ohren mit unabhngig voneinander arbeitenden CI-Systemen versorgt, deren Einstellungen mit dem sich entwickelnden Hrvermgen mehr und mehr auf einen balancierten Hreindruck hin optimiert werden. Informationen ber interaurale Schallmerkmale (insbesondere die interaurale Zeitdifferenz) knnen jedoch bisher nur partiell bertragen werden, so da ein rumliches Hren nur mit Einschrnkungen zu erreichen ist. Dies fhrt zwar in der Regel trotzdem zu einem betrchtlichen Hrgewinn fr den Patienten, dessen Hrleistungen sind aber noch weit von den binauralen Hrleistungen Normalhrender entfernt. Aktuelle Trends Die Entwicklung von Signalverarbeitungsalgorithmen und neuen Codierungsstrategien hat zum Ziel, dem CI-Trger mglichst viele Merkmale rumlicher Hrsituationen auswertbar zur Verfgung zu stellen und dadurch ein besseres Richtungshren und Sprachverstehen im Strschall zu ermglichen. Zudem kann eine geeignete bilaterale Signal-Vorverarbeitung bei bilateral versorgten CI-Trgern das Sprachverstehen in realen Hrsituationen deutlich verbessern und vereinfachen. Ausblick Die Notwendigkeit einer Datenverbindung zwischen linkem und rechtem Sprachprozessor zur Synchronisation der Verarbeitung oder zur gemeinsamen Verarbeitung beider Ohrsignale muss geprft und mit den kommerziellen Anforderungen nach hohem Tragekomfort (keine strenden Verbindungskabel am Kopf) und geringen Betriebskosten (Stromverbrauch) in Einklang gebracht werden. Wenn dies gelingt, und wenn auch die Codierungsstrategien so gestaltet werden, dass interaurale Schallmerkmale im neuronalen Erregungsmuster besser reprsentiert werden, kann auch das rumliche Hren mit bilateralen CI-Systemen noch deutlich verbessert werden.

Spectral Cues for Sound Localization with Cochlear Implants

Piotr Majdak, Bernhard Laback, and Matthew Goupell Acoustics Research Institute (ARI), Austrian Academy of Sciences Current cochlear implant (CI) systems aim to restore speech understanding in profound deaf humans, however, they can not replace all functions of the normal auditory system. While localization of sound sources in the left/right dimension has been shown to be possible by means of CI stimulation at both ears, localization in the front/back and up/down dimensions (within the sagittal planes) still remains difficult because of insufficient transmission of relevant cues by current CI systems. In normal hearing (NH), localization in the sagittal planes is achieved by the evaluation of spectral colouring of the incoming sound induced by the pinna at high frequencies (pinna cue). Current CI systems, however, do not transmit pinna cues due to the placement of the microphone behind the pinna and the restricted bandwidth of the signal analysis stage of the processor. While these technical limitations can be handled, the following basic questions remain: a) How many frequency channels are required to transmit pinna cues relevant for sound localization in the sagittal planes? b) To what extent can CI listeners learn to localize sound sources in the sagittal planes using high-frequency pinna cues that are mapped to tonotopic places with lower characteristic frequencies, as usually covered by CI electrodes? To address these questions psychoacoustic experiments have been conducted with CI and NH subjects, the latter being presented with acoustic simulations of the perception of a CI listener. The ultimate goal of the project is to enable CI listeners to localize sound sources in the sagittal planes. A series of experiments determined the effects of parameters, such as the number and tonotopic range of the stimulating electrodes on the coding of pinna cues and the speech signal. The results of these studies will be presented in the talk. In a current study (in progress), the ability of both subject groups to learn localization in the sagittal planes is studied using pinna cues that are frequency-warped to match the tonotopic range associated with the basal electrodes in CI listeners. The subjects learn to associate auditory stimuli with spatial positions by means of audiovisual stimulus presentation.

Effects of slope steepness and off time on envelope ITD sensitivity in acoustic and electric hearing
Bernhard Laback, Inge Zimmermann, and Piotr Majdak Acoustics Research Institute (ARI), Austrian Academy of Sciences Cochlear implant (CI) listeners have been shown to be sensitive to interaural time differences (ITD) in the pulse timing (fine structure) and to ITD in the envelope. However, the sensitivity is lower and varies much more between individuals compared to normal hearing (NH) listeners. With respect to envelope ITD, basic questions are still open both in normal and electric hearing. In normal hearing, the ITD sensitivity for transposed tones (TTs), a special type of amplitudemodulated high-frequency pure tones, has been shown to be better than for sinusoidally-amplitudemodulated (SAM) tones. In experiments with nine NH and nine CI listeners we addressed the question which properties of TTs cause the increased ITD sensitivity. To that end, we systematically varied the off-time and slope steepness of trapezoidally modulated (27 Hz) carrier signals (8727-Hz pure tone and 1515-pps pulse train in acoustic and electric hearing, respectively). The NH listeners showed significant effects of both off time and slope. Both effects did not interact and the best performance occurred for the longest off-time combined with the steepest slope tested. The NH data suggest that the JND-difference between SAM tones and TTs can be attributed to a combination of effects of slope and off time. The CI listeners showed only a significant effect of the off-time. Potential reasons for the lack of a slope effect in CI listeners and general implications of the results for CI coding strategies are discussed.

Stimulation of the cochlea using green laser light

Wenzel GI, Lim HH,4, Zhang K, Balster S, Lubatschowski H, Ertmer W, Reuter G, Lenarz T Department of Otolaryngology, Medical University Hannover, Laser Center Hannover, Institute of Quantum Optics, Leibniz University of Hannover, 4Department of Biomedical Engineering, University of Minnesota The hearing performance with conventional hearing aids and cochlear implants is dramatically reduced in noisy environments and for sounds more complex than speech (e.g., music) partially due to the lack of localized sensorineural activation across different frequency regions with these devices. Laser light can be focused in a controlled manner and may provide more localized activation of the cochlea. The activation of auditory nerve fibers using mid-infrared light has been demonstrated previously (Richter et al., Izzo et al., 2006-2008). We hypothesized that visible light could be used to activate the organ of Corti for the purpose of replacing the cochlear amplifier in hearing impaired patients with residual hearing. The laser parameters were selected based on our initial attempt to induce an optoacoustic effect as the energy transfer mechanism to the cochlea. Optically induced ABRs (OABRs) similar in shape to those of acoustic stimulation were elicited with single laser pulses applied with an optical fiber positioned in the round window niche and directed towards the basilar membrane in guinea pigs. The OABR peaks increased with energy levels from 0.6 to 23 J/pulse and remained consistent even after 30 minutes of continuous stimulation at 13 J, indicating minimal or no stimulation-induced damage within the cochlea. Additional local field potentials and spike activity were elicited within the central nucleus of the inferior colliculus (ICC) using an electrode array (16 linear sites) in response to the same stimulation parameters as for the OABRs. At low levels, neural activity was localized to a high frequency region of the ICC consistent with the stimulated high frequency region of the cochlea near the round window. As the energy level was increased, greater activity appeared across the tonotopic gradient of the ICC. Regarding the activation mechanism, our Laser Doppler experiments on excised cochlea demonstrated that 532 nm pulsed laser irradiation induces optoacoustic waves within the cochlea and localized vibrations of the basilar membrane in close proximity to the optoacoustic waves source. The vibration amplitude could be modulated by adjusting the laser pulse energy. These findings suggest that green laser light stimulation has the potential for a new type of auditory prosthesis that can achieve localized activation of the cochlea. Further studies are needed to investigate the effects of optical stimulation of lower frequency cochlear regions and how to safely elicit frequency-specific auditory activation with multiple fibers positioned within the cochlea.