AU2011265489B2 - Pulsatile cochlear implant stimulation strategy - Google Patents

Abstract

Abstract The invention provides an implantable device having a multi-channel electrode array, each electrode in the array being associated with a signal processing output channel, 5 the device comprising: an audio processing stage for processing an input audio signal to produce a plurality of output channel signals each representing an associated band of audio frequencies; a timing and envelope detector for processing the output channel signals in a sequence of sampling intervals, wherein for each sampling interval, the processing includes determining: one or more pulse timing requests, and corresponding envelope 10 signals representing pulse magnitude for pulse timing requests; a pulse selection amplitude definition stage for determining for the requested pulse timings: output pulses, if any, at specified times and amplitudes as selected from the requested pulse timings based on a pulse selection inhibition function that selects fewer pulse timings at higher frequencies; and the multi-channel electrode array for applying the output pulses at their associated 15 stimulation amplitudes to surrounding tissue. 30390231 (GHMatters) P83976.AU.1 cj~ 4.) * * * '0 4) 4) 4) o I 1-I 4) Cd, Li cj*~ 4) C.)

Description

AUSTRALIA Patents Act 1990 COMPLETE SPECIFICATION Standard Patent Applicantss: MED-EL Elektromedizinische Geraete GmbH Invention Title: Pulsatile cochlear implant stimulation strategy The following statement is a full description of this invention, including the best method for performing it known to me/us: -2 Pulsatile Cochlear Implant Stimulation Strategy Related application This application is a divisional application of Australian application no. 5 2008323718 the disclosure of which is incorporated herein by reference. Most of the disclosure of that application is also included herein, however, reference may be made to the specification of application no. 2008323718 as filed or accepted to gain further understanding of the invention claimed herein. 10 Field of the Invention The present invention relates to implantable medical devices, and more specifically, to techniques for coding stimulation pulses in such devices. Background Art 15 Cochlear implants are implantable systems which can provide hearing to profoundly deaf or severely hearing impaired persons. Unlike conventional hearing aids which mechanically apply an amplified sound signal to the middle ear, a cochlear implant provides direct electrical stimulation to multiple implant electrodes that excite the acoustic nerve in the inner ear. Most current cochlear implant electric stimulation coding strategies 20 represent a sound signal by splitting it into distinct frequency bands and extracting the envelope (i.e., energy) of each of these bands. These envelope representations of the acoustic signal are used to define the stimulation amplitude of each electrode. One current approach, the Fine Structure Processing (FSP) coding strategy, 25 commercially available in the Med-El OPUS 1 and OPUS 2 speech processors, analyzes the phase of the band pass signals and synchronizes the stimulation pulses with specific events in the phase of the corresponding electrode. In FSP coding, time events are defined using the zero crossings of the band pass signal where all system channels are stimulated sequentially in a predetermined order (a "stimulation frame"). The stimulation rate or grid 30 respectively of each channel is generally defined by the sum of the pulse durations and the pauses between consecutive stimulation pulses. The frame rate (i.e. the repetition rate) of one stimulation frame equals the stimulation rate or grid of each channel, typically 1000 2000 Hz. 3039023_ (GHMatters) P83975 AU.I -3 FSP coding uses Channel Specific Sampling Sequences (CSSS), described, for example, in U.S. Patent 6,594,525 (incorporated herein by reference) to represent the temporal information in the band pass signal. After a zero crossing in the band pass signal, 5 a specific CSSS is started at the assigned electrode. The temporal accuracy is determined by the grid that is equal to the frame rate in FSP coding. This accuracy allows for coding temporal fine structure information up to several hundred Hertz. The temporal accuracy of CSSS in FSP is mainly defined by the pulse durations, i.e. at high pulse duration the accuracy of CSSS is low and the maximum frequency coded temporally is low as well. 10 Higher temporal accuracy of stimulation pulses can be achieved in an temporal fine structure coding strategy using CSSS together with the use of selected channel stimulation groups, as described, for example, in U.S. Patent 7,283,876 (incorporated herein by reference). Different types of channels are defined (e.g. CSSS channels and 15 envelope channels) and certain channels have to be grouped. For example, all the CSSS channels are placed into one or more groups in which some of the groups are repeated more often during a given stimulation frame. And within a given group, one or more of the channels can be stimulated simultaneously. This results in a temporal grid of CSSS stimulation which is a multiple of the frame rate. Improved temporal accuracy of the CSSS 20 allows coding of phase information (up to about 1000 Hz) based on a high temporal grid using short pulse durations. With high pulse durations, temporal accuracy and frame rate (i.e. the rate of high frequency envelope channels) are reduced again. Most feasible combinations of CSSS, selected groups, and simultaneous stimulation, will have some mismatch between average CSSS rates of the highest CSSS channel and neighboring 25 envelope channels. In such temporal fine structure coding strategies, a certain number of requested stimulation pulses are deselected. The number of deselected stimulation pulses (mainly within CSSS channels) is higher with higher pulse durations, which might lead to a loss of temporal information. 30 The current literature describes three other approaches that provide some temporal fine structure information. Peak Derived Timing (PDT) was described in Vandali et al., Pitch Ranking Ability Of Cochlear Implant Recipients: A Comparison Of Sound Processing Strategies, J Acoust Soc Am. 2005 May; 117(5):3126-38 (incorporated herein 3039023_1 (GHMatlers) P83978AU.1 -4 by reference). The PDT coding was experimentally used in cochlear implant users and derived the timing of stimulation pulses from the positive peaks in the band pass signals. Timing of the pulses was managed by an arbitration scheme which delayed or advanced simultaneously requested stimulation pulses. No refractory behavior was implemented in 5 this algorithm. Asynchronous Interleaved Sampling (AIS) was described in Sit et al., A Low Power Asynchronous Interleaved Sampling Algorithm For Cochlear Implants That Encodes Envelope And Phase Information, IEEE Trans Biomed. Eng. 2007 Jan; 10 54(l):138-49 (incorporated herein by reference). The AIS strategy used asynchronous extraction of time events from band pass signals, but lacked any handling of interleaved stimulation pulses, which are a necessary part of a usable cochlear implant sound coding strategy. 15 Spike-based Temporal Auditory Representation (STAR) strategy is based upon an auditory model as described, for example, in Grayden et al., A Cochlear Implant Speech Processing Strategy Based On An Auditory Model, Proceedings of the 2004 Intelligent Sensors, Sensor Networks and Information Processing Conference, 14-17 Dec. 2004: 491 - 496 (incorporated herein by reference). The STAR approach, somewhat like CSSS, 20 extracted the pulse timing from the zero crossings of the band pass signals. In this strategy 'spike timing contentions' are resolved by systematically shifting stimulation pulses to different time instances around the zero crossing. No details about the algorithm are given. The average stimulation rate on high frequency channels is restricted, but no details about the mechanism are given in the publication. 25 Summary of the Invention In an aspect, the invention provides an implantable device having a multi-channel electrode array, each electrode in the array being associated with a signal processing output channel, the device comprising: 30 an audio processing stage for processing an input audio signal to produce a plurality of output channel signals each representing an associated band of audio frequencies; a timing and envelope detector for processing the output channel signals in a sequence 30390231 (GHMatters) P83976 AU.1 of sampling intervals, wherein for each sampling interval, the processing includes determining: i. one or more pulse timing requests, and ii. corresponding envelope signals representing pulse magnitude for pulse 5 timing requests; a pulse selection amplitude definition stage for determining for the requested pulse timings at specified times and amplitudes as selected from the requested pulse timings based on a pulse selection inhibition function that selects fewer pulse timings at higher frequencies; and 10 the multi-channel electrode array being for applying the output pulses at their associated stimulation amplitudes to surrounding tissue. In more specific embodiments, the inhibition function may be constant at times so as to define an absolute inhibition state, and/or may be changing at times so as to define a 15 relative inhibition state. The inhibition function may be output channel dependent and/or pulse magnitude dependent, for example, reflecting a ratio of inhibition state to pulse magnitude. The output pulses may be selected based on length of an inhibition state defined by the inhibition function; for example, the output pulses may be selected preferentially based on shortness of the inhibition state. 20 In specific embodiments, the input audio signal includes temporal structure characteristics which are represented in the output channel signals by channel specific sampling sequences (CSSS) and/or are reflected in the specified times of the output pulses. The electrode array may specifically be a cochlear implant electrode array. 25 Brief Description of the Drawings In order that the invention may be more clearly ascertained, embodiments will now be described, by way of example, with reference to the accompanying drawings, in which: 30 Figure 1 shows an example of a typical acoustic signal. Figure 2 shows an acoustic signal decomposed by band pass filtering by a bank of 3383010_1 (GHMatters) P83976 AU.11405 12 -6 filters into a set of signals (input channels). Figure 3 shows an example of a Timing and Energy Detector according to an embodiment of the present invention. 5 Figure 4 shows examples of the envelopes extracted from the set of input signals. Figure 5 shows timing of requested stimulation pulses extracted from the set of input signals. 10 Figure 6 shows an example of a Pulse Selection and Amplitude Definition Stage (PSADS) according to one embodiment of the present invention. Figure 7 shows an example of a possible implementation of a pulse selection 15 algorithm within a PSADS. Figure 8 shows an example of timing of the selected stimulation pulses according to an embodiment. 20 Figure 9 shows an example of biphasic stimulation pulses in a sequential implementation of the system. Figure 10 shows further details of the biphasic stimulation pulses in a sequential implementation of the system. 25 Detailed Description of Specific Embodiments Embodiments of the present invention are directed to a signal coding strategy for an implantable device having a multi-channel electrode array. The signal coding strategy encodes temporal characteristics of sound signals at a higher temporal accuracy and higher 30 pulse durations than previously. Temporal fine structure is encoded and stimulation of refractor nerve populations can be inhibited by an inhibition function that determines whether or not stimulation pulses are delivered to one or more electrodes. Such signal coding is no longer based upon fixed stimulation rates and channel orders, and the 3039023_1 (GHMatters) P83976 AU.1 -7 encoding of the temporal fine structure is more accurate, even at long pulse durations. Figure 1 shows a typical sound signal in which the overall amplitude varies over a short period of time. Such a sound signal inherently contains specific timing information 5 that characterizes the signal. The sound signal in this form as an audio electrical signal is typically pre-processed into multiple output channel signals. For example, one common approach is to pre-process the initial audio signal with a bank of filters where it is decomposed by band pass filtering to form a set of output channel signals such as the example shown in Figure 2, each of which represents an associated band of audio 10 frequencies. Alternatively, in another embodiment, the initial audio signal can be processed by one or more non-linear filters which provide multiple output channel signals. Unlike prior fine structure processing approaches where different types of output channels have to be defined (e.g., CSSS fine structure channels and envelope channels), 15 embodiments of the present invention treat all output channels equally. Figure 3 shows an example of a Timing and Envelope Detector (TED) that receives as an input a set of output channel signals such as the sound signals shown in Figure 2 from a bank of band pass filters. The TED processes these output channel signals in a continuing sequence of sampling intervals which are sampled at a given rate that, for example, may be defined by 20 the pulse durations used for the electrical stimulation (e.g. the inverse of the maximum pulse duration). The TED extracts certain time events from each sampling interval, e.g., zero crossings, signal maxima, adaptive threshold levels, etc. as well as envelope information. The TED outputs a set of envelope signals (shown, for example, in Figure 4) serving for the calculation of pulse magnitudes, and a set of time event signals (shown, for 25 example, in Figure 5) that flag requested stimulation pulses. As shown in Figure 6, the TED outputs are provided to a Pulse Selection / Amplitude Definition Stage (PSADS) that selects a reduced set of time events (output stimulation pulses) and calculates stimulation amplitudes for the selected output pulses. 30 The PSADS uses an inhibition function to calculate and analyze an inhibition state for each output channel. Within each sampling interval, the requested pulses are identified, and based on the inhibition states and envelopes of identified channels, at least one channel requesting a pulse is selected. For example, one way for the PSADS to select 3039023_1 (GHMatters) P83970.AU.1 -8 pulses might be to select one or more pulse requests within each sampling interval which have the shortest associated inhibition states. More complex selection algorithms can take into account the envelopes of the requested pulses so that the ratio of the inhibition state to the pulse amplitude serves as a selection criterion. Figure 7 illustrates the selection process 5 using a simple prototype of an inhibition function where the asterisks represent the timing of requested pulses per channel, solid lines depict the selected pulses, and dashed lines depict the inhibition states. Once a pulse request is selected as an output pulse, a channel- and amplitude 10 specific inhibition function is triggered on the selected output channel. In the specific embodiment shown in Figure 7, the inhibition function is constant for some hundreds of microseconds (e.g., 500 is) during a maximum or absolute inhibition phase, and then decreases towards zero over another period of several hundred microseconds (e.g., 1500 ps) which defines a relative inhibition phase. In this embodiment, requested pulses that 15 occur within the absolute inhibition phase are not selected as output pulses for stimulation. Thus, in this example, inhibition times can be used to define the maximum channel specific stimulation rate of the system. Figure 8 illustrates the reduced set of time events (selected output pulses) resulting from this pulse selection. When compared to the initially requested pulse timings shown in Figure 5, the number of selected output pulse timings 20 produced by the PSADS, especially at higher frequencies, is markedly reduced. Figure 9 shows the resulting biphasic stimulation pulses and their pulse amplitudes which are applied to the different channel electrodes. Figure 10 shows a detailed expansion of a portion of the time illustrated in Figure 9. 25 Such stimulation timing approaches can provide very accurate representation of low frequency temporal fine structure. For example, the time grid of stimulation per output channel can be defined by the maximum pulse duration required for electrical stimulation. With typical biphasic pulse durations of 50 ps, a time grid of up to 20 kHz may be realized. Even with relatively long electrical pulses, a higher timing accuracy of timing 30 can be achieved. For example, with a 100 pts pulse duration on each output channel, a 10 kHz stimulation grid may be achievable. In an embodiment in which only one output pulse is selected for each sampling interval, the time grid is twice as fast as the fastest possible combination of just CSSS with selected electrode channel groups also applying one 3039023.1 (GHMatters) P83976.AU.1 -9 stimulation pulse at any time instance. Compared to the fastest possible combinations of CSSS with selected electrode channel groups, the number of deselected pulses can be largely reduced with specific embodiments. For example, at a time grid of 20 kHz and an absolute inhibition phase of 500 is, a negligibly low number of requested stimulation 5 pulses has to be deselected from output channels carrying a temporal fine structure of up to above 1000 Hz. Specific embodiments can be implemented in a system with relatively low supply voltage such as a fully implantable cochlear implant system. In such a system, low 10 compliance voltages would need relatively long stimulation pulses to achieve comfortable loudness. Embodiments enable presentation of temporal fine structure even at low supply voltages and can therefore be used in patients with low compliance voltages. A specific embodiment could be implemented with cochlear implants capable of either simultaneous stimulation or sequential stimulation. 15 Embodiments could also be useful for patients suffering from facial stimulation due to the cochlear implant stimulation. Such applications need relatively long pulse durations, and under such conditions, embodiments of the present invention are able to exactly transmit temporal fine structure. 20 The type and form of a channel-/amplitude-specific inhibition function can be used to define channel specific stimulation rates. The above described system with the above described prototype inhibition function with an absolute/maximum inhibition of 500 ps and the channel selection based exclusively on the inhibition function would allow 2000 25 Hz stimulation per channel at a maximum. Longer durations of absolute inhibition could be used to drastically reduce power consumption through electrical stimulation. Some embodiments may also better mimic the natural neural behavior of the human ear. In particular, neural populations stimulated by electrodes connected to low 30 frequency channels would be stimulated at lower rates. Stimulation on these electrodes will be relatively deterministic for a broad range of inhibition functions and times. High frequency channels can apply pseudo-stochastic timing to stimulation pulses. For a given TED and selection of inhibition time constants and pulse widths (time grid), the channel 3039023_1 (GHMalters) P83976.AU.1 - 10 specific stimulation rate would be nearly equal to the characteristic frequency of the corresponding electrode channel, whereas for higher frequency electrode channels, a "natural" saturation of stimulation rate may be obtainable. The inhibition function may allow for the refractory behavior of stimulated nerve populations within an electrode 5 channel, which would also lead to a reduction of stimulation power at a constant loudness percept. Specific embodiments of the PSADS may well provide additional functionality useful for selecting and defining the output pulses. Accordingly, the PSADS may typically 10 contain hardware and/or software modules to that end, such as non-linear circuitry for defining patient- and electrode-specific stimulation amplitudes. For example, specific implementations of an inhibition function algorithm may take into account such goals. As a result, in specific embodiments the PSADS could be more complex than as described above. 15 Embodiments of the invention may be implemented in any conventional computer programming language. For example, preferred embodiments may be implemented in a procedural programming language (e.g., "C") or an object oriented programming language (e.g., "C++", Python). Alternative embodiments of the invention may be implemented as 20 pre-programmed hardware elements, other related components, or as a combination of hardware and software components. Embodiments can be implemented as a computer program product for use with a computer system. Such implementation may include a series of computer instructions 25 fixed either on a tangible medium, such as a computer readable medium (e.g., a diskette, CD-ROM, ROM, or fixed disk) or transmittable to a computer system, via a modem or other interface device, such as a communications adapter connected to a network over a medium. The medium may be either a tangible medium (e.g., optical or analog communications lines) or a medium implemented with wireless techniques (e.g., 30 microwave, infrared or other transmission techniques). The series of computer instructions embodies all or part of the functionality previously described herein with respect to the system. Those skilled in the art should appreciate that such computer instructions can be written in a number of programming languages for use with many 30390231 (GHMfteis) P83976.AU.1 - 11 computer architectures or operating systems. Furthermore, such instructions may be stored in any memory device, such as semiconductor, magnetic, optical or other memory devices, and may be transmitted using any communications technology, such as optical, infrared, microwave, or other transmission technologies. It is expected that such a 5 computer program product may be distributed as a removable medium with accompanying printed or electronic documentation (e.g., shrink wrapped software), preloaded with a computer system (e.g., on system ROM or fixed disk), or distributed from a server or electronic bulletin board over the network (e.g., the Internet or World Wide Web). Of course, some embodiments of the invention may be implemented as a combination of both 10 software (e.g., a computer program product) and hardware. Still other embodiments of the invention are implemented as entirely hardware, or entirely software (e.g., a computer program product). Although various exemplary embodiments of the invention have been disclosed, it 15 should be apparent to those skilled in the art that various changes and modifications can be made which will achieve some of the advantages of the invention without departing from the true scope of the invention. In the claims which follow and in the preceding description of the invention, 20 except where the context requires otherwise due to express language or necessary implication, the word "comprise" or variations such as "comprises" or "comprising" is used in an inclusive sense, i.e. to specify the presence of the stated features but not to preclude the presence or addition of further features in various embodiments of the invention. 25 It is to be understood that references to prior art referred herein do not constitute an admission that the publication forms or formed a part of the common general knowledge in the art, in Australia or any other country. 30 30390231 (GHMatters) P83976.AU.1

Claims (12)

1. An implantable device having a multi-channel electrode array, each electrode in the array being associated with a signal processing output channel, the device comprising: 5 an audio processing stage for processing an input audio signal to produce a plurality of output channel signals each representing an associated band of audio frequencies; a timing and envelope detector for processing the output channel signals in a sequence of sampling intervals, wherein for each sampling interval, the processing 10 includes determining: i. one or more pulse timing requests, and ii. corresponding envelope signals representing pulse magnitude for pulse timing requests; a pulse selection amplitude definition stage for determining for the requested pulse 15 timings at specified times and amplitudes as selected from the requested pulse timings based on a pulse selection inhibition function that selects fewer pulse timings at higher frequencies; and the multi-channel electrode array being for applying the output pulses at their associated stimulation amplitudes to surrounding tissue. 20

2. An implantable device according to claim 1, wherein the inhibition function is constant at times so as to define an absolute inhibition state.

3. An implantable device according to claim 1, wherein the inhibition function is 25 changing at times so as to define a relative inhibition state.

4. An implantable device according to claim 1, wherein the inhibition function is pulse magnitude dependent. 30

5. An implantable device according to claim 4, wherein the inhibition function reflects a ratio of inhibition state to pulse magnitude.

6. An implantable device according to claim 1, wherein the inhibition function is output 3383010_1 (GHMatters) P83978.AU.1 I105/2012 channel dependent.

7. An implantable device according to claim 1, wherein output pulses are selected based on length of an inhibition state defined by the inhibition function. 5

8. An implantable device according to claim 7, wherein output pulses are selected preferentially based on shortness of the inhibition state.

9. An implantable device according to claim 1, wherein the input audio signal includes 10 temporal structure characteristics which are represented in the output channel signals by channel specific sampling sequences (CSSS).

10. An implantable device according to claim 1, wherein the input audio signal includes temporal structure characteristics which are reflected in the specified times of the output 15 pulses.

11. An implantable device according to claim 1, wherein the electrode array is a cochlear implant electrode array. 20

12. An implantable device as claimed in any one of the preceding claims, and substantially as herein described with reference to the accompanying drawings.

3363010.1 (GHMtIers) P83970.AU.1 14/05/2012