Cochlear implant system with improved temporal feature selection

Abstract

This application discloses a cochlear implantation system with improved time feature selection, comprising: a microphone unit; an electrode array; a processor unit including: a filter bank; a first time feature extractor configured to determine an initial set of time features of a bandpass audio signal; a second time feature extractor configured to window-filter the bandpass audio signal and determine a secondary set of time features of the window-filtered bandpass audio signal; a stimulation generator configured to determine a final set of time features based on the initial set of time features and the secondary set of time features, and to determine a similarity measure value between the initial set of time features and the secondary set of time features for the final set of time features; the processor unit configured to determine, based on the similarity measure value, whether the final set of time features will be encoded into a stimulation pulse; the encoded stimulation pulse is transmitted to an electrode configured to apply electrical stimulation to a group of auditory nerve fibers based on the encoded stimulation pulse.

Description

Technical Field

[0001] This application relates to cochlear implantation systems. More specifically, this application relates to a processor unit configured to determine whether a set of temporal features will be encoded into a stimulation pulse. Background Technology

[0002] Known multichannel cochlear implant systems directly encode ambient sound into multiple stimulation pulses, which are then encoded into one or more stimulation frames using specific temporal feature encoding strategies such as Temporal Fine Structure (TFS) frames. TFS encoding strategies consider many aspects of normal hearing physiology. A key aspect of this type of encoding strategy (distinguishing it from other strategies) is the timing of the electrode pulses; for example, their initiation is used to transmit temporal features that would be lost in standard fixed-rate strategies. For instance, an audio signal can be divided into multiple bandpass signals, each constrained by one electrode of the cochlear implant system, and the phase of each bandpass signal is typically ignored in fixed-rate strategies, while in TFS encoding strategies, this phase information is attempted to be encoded as electrode timing. For this purpose, each bandpass signal and its phase are typically obtained using time-domain based analyses such as filter banks, zero-intersections, and / or Hilbert methods. Including temporal features such as phase in the encoding strategy without quantifying their association with the auditory brain is not optimal. This is because, due to implant or CI user limitations, frequent trade-offs are necessary; for example, the same electrode cannot be stimulated simultaneously at two different rates. This problem is becoming increasingly important because there are many conflicting stimulus impulses, which requires a TFS strategy to select and prioritize one impulse, causing other impulses to deviate or be removed to make room.

[0003] Therefore, a solution is needed that addresses at least some of the problems mentioned above. Summary of the Invention

[0004] One aspect of the present invention is to provide an improved option for encoding temporal characteristics into stimulation pulses by a cochlear implantation system.

[0005] Another aspect of the present invention is to reduce the computing power of the processor unit of the cochlear implant system.

[0006] This aspect of the invention is implemented by a cochlear implantation system, which includes a microphone unit configured to receive acoustic signals and transmit audio signals based on the acoustic signals, and an electrode array including a plurality of electrodes. Furthermore, the cochlear implantation system includes a processor unit comprising a filter bank, the filter bank including a plurality of bandpass channels configured to divide the audio signals into a plurality of bandpass audio signals. Each of the plurality of bandpass channels may be limited to one of the plurality of electrodes. Additionally, the cochlear implantation system may include: a first time feature extractor configured to determine an initial set of time features of a bandpass audio signal from the plurality of bandpass audio signals; and a second time feature extractor configured to perform window filtering on the bandpass audio signal and determine a secondary set of time features of the window-filtered bandpass audio signal.

[0007] A window filter is provided to generate an envelope-like output signal, in which time features that are more robust to filtering pass through the window filter, thus the window-filtered bandpass audio signal differs from the unfiltered bandpass audio signal because it includes fewer time features.

[0008] A window filter may include a low-pass filter or a band-pass filter configured to filter a band-pass audio signal. The window-filtered band-pass audio signal may include an oscillating envelope, which primarily comprises the fundamental frequency, as harmonic frequencies are highly attenuated due to filtering. The cutoff frequency of the window filter may be adapted to provide the center frequency of a band-pass channel that will filter the band-pass audio signal by the window filter. This band-pass channel is part of a plurality of band-pass channels.

[0009] Window filters can include any linear or nonlinear filter function.

[0010] For example, the higher the bandpass channel number, the higher the center frequency, and thus the higher the cutoff frequency.

[0011] Therefore, the cutoff frequency of the window filter can be determined by the processor unit based on the center frequency of the bandpass channel that provides the bandpass audio signal among multiple bandpass channels.

[0012] The cutoff frequency of the window filter can be determined by the processor unit based on feedback from the second amplitude envelope detector. This feedback may include information about whether the window-filtered bandpass audio signal includes an envelope from which the amplitude can be determined.

[0013] The similarity measure between an initial set of time features and a secondary set of time features can be determined by a value function between the initial set of time features and the secondary set of time features.

[0014] The similarity measure between an initial set of time features and a secondary set of time features can be determined by a value function between a first initial time feature in the initial set of time features and a second initial time feature in the secondary set of time features, where the first and second initial time features belong to the same type.

[0015] The first and second initial time features belong to the same type, such as the start event time or amplitude. The first generation time feature in the initial set of time features and the second generation time feature in the secondary set of time features belong to the same type, such as the start event time or amplitude.

[0016] If the similarity between the first initial time feature and the second initial time feature increases, the value function decreases, thereby increasing the similarity measure value.

[0017] The similarity measure between an initial set of time features and a secondary set of time features can be determined as follows: determine the similarity measure between the first amplitude and the second amplitude and determine the time difference measure between the first start event time and the second start event time, and then determine the similarity measure value through a value function that includes the similarity measure between amplitudes and the time difference measure between start event times.

[0018] Therefore, the similarity measure defines the overall similarity level of the initial event time and amplitude of the initial set of time features and the secondary set of time features.

[0019] Similarity measures may include determining the cross-correlation between the first and second amplitudes and determining the cross-covariance between the first and second amplitudes. The similarity measure can then be determined by a linear combination of the cross-correlation and the cross-covariance. This linear combination can be the sum or multiplication of the cross-correlation and the cross-covariance.

[0020] Additionally, the processor unit includes a stimulus generator configured to determine a final set of time features based on an initial set of time features and a secondary set of time features. Furthermore, the stimulus generator can be configured to determine a similarity measure between the initial set of time features and the secondary set of time features for the final set of time features. The processor unit can be configured to determine, based on the similarity measure, whether the final set of time features will be encoded into stimulus impulses or to prioritize stimulus impulses encoded based on the final set of time features based on the similarity measure.

[0021] Encoded stimulation pulses can be transmitted to one of a plurality of electrodes, which can be configured to apply electrical stimulation to a set of auditory nerve fibers of the recipient of the cochlear implant system based on the encoded stimulation pulses.

[0022] The processor unit is configured to determine a final set of temporal features for each of a plurality of bandpass audio signals, and to determine a similarity metric value for each final set of temporal features. Furthermore, the processor unit is configured to determine, based on the similarity metric value, whether the final set of temporal features will be encoded into a stimulation pulse, wherein each of the encoded stimulation pulses can be transmitted to a set of electrodes among a plurality of electrodes.

[0023] An initial set of time features may include a first start event time of the initial event detected in the initial event sequence of the bandpass audio signal and a first amplitude of the initial event at the first start event time. A secondary set of time features may include a second start event time of the secondary event detected in the secondary event sequence of the window-filtered bandpass audio signal and a second amplitude of the secondary event at the second start event time. A final set of time features may include a final start event time and a final amplitude, wherein the final start event time may be determined based on the first start event time and the second start event time, and the final amplitude may be determined based on the first amplitude and the second amplitude.

[0024] The final start event time defines the timing of encoding temporal characteristics into the stimulus pulse. The final start event time can define the timing of adjusting the power required to transmit the encoded stimulus pulse to the auditory nerve fiber.

[0025] The final amplitude definition encodes the amplitude level and / or pulse width of the stimulus pulse.

[0026] A first temporal feature extractor may include: a first event detector configured to determine a first start event time of a bandpass audio signal among a plurality of bandpass audio signals; and a first envelope detector configured to determine a first amplitude of the bandpass audio signal among the plurality of bandpass audio signals at the first start event time. A second temporal feature extractor may include: a second event detector configured to determine a second start event time of a windowed bandpass audio signal among the plurality of bandpass audio signals; and a second envelope detector configured to determine a second amplitude of the windowed bandpass audio signal among the plurality of bandpass audio signals at the second start event time.

[0027] The event detector can be configured to determine a phase signal over time based on a bandpass audio signal (or a windowed bandpass audio signal), wherein the start time of the event is determined in the phase signal.

[0028] The envelope detector can be configured to determine the amplitude envelope over time based on a bandpass audio signal (or a windowed bandpass audio signal), wherein the amplitude is determined at a time equal to the start event time within the amplitude envelope.

[0029] The event detector can be configured to generate a sequence of events for a bandpass audio signal based on the detection that the phase of the bandpass audio signal exceeds a phase threshold. Each event in the event sequence is generated based on the detection that the phase of the bandpass audio signal exceeds a phase threshold, which can be any value between 0 and 2π.

[0030] When the difference between the first initial time feature in the initial set of time features and the second initial time feature in the secondary set of time features is lower than the threshold level of the starting event time feature or the threshold level of the amplitude feature, the final time feature in the final set of time features can be determined by the first weight of the first initial time feature in the initial set of time features and the second weight of the first initial time feature in the secondary set of time features. Wherein, when the first weight decreases, the second weight increases, and vice versa. The sum of the first weight and the second weight is always 1.

[0031] When the difference between the first initial time feature in the initial set of time features and the second initial time feature in the secondary set of time features is higher than the threshold level of the starting event time feature or the threshold level of the amplitude feature, the final time feature in the final set of time features can be equal to the first initial time feature in the initial set of time features.

[0032] The final time feature in the final set of time features can be determined by the following formula:

[0033]

[0034] Wherein, 1-W(Fc(ch),BW(ch)) is a first weighting factor depending on the center frequency (Fc(ch)) and / or bandwidth of the bandpass channel ch among multiple bandpass channels, W(Fc(ch),BW(ch)) is a second weighting factor, F1(ch) is a first initial time feature (such as the first start event time T1(ch) or the first amplitude A1(ch)), F2(ch) is a second initial time feature (such as the second start event time T2(ch) or the second amplitude A2(ch)), TFHR is the threshold level for the initial event time feature, and AFHR is the threshold level for the amplitude feature. AFHR can vary between 0 and above 0.

[0035] The above equation is a generalized equation for the final time feature and the final amplitude feature, but the weighting factor values ​​are different between the final time feature (TF) and the final amplitude feature (AF).

[0036] The second weighting factor increases with decreasing center frequency and decreases with increasing center frequency. Attached Figure Description

[0037] Various aspects of the invention will be best understood from the following detailed description taken in conjunction with the accompanying drawings. For clarity, these drawings are schematic and simplified, showing only the details necessary for understanding the invention while omitting other details. Throughout the specification, the same reference numerals are used for the same or corresponding parts. Features of each aspect may be combined with any or all features of other aspects. These and other aspects, features, and / or technical effects will be apparent from and illustrated in the following figures, wherein:

[0038] Figure 1 An example of a cochlear implantation system;

[0039] Figure 2 Another example of a cochlear implantation system;

[0040] Figure 3A and 3B Here is an example of a temporal feature extractor;

[0041] Figure 4A and 4B Here is an example of a window filter;

[0042] Figure 5 An example of a stimulus generator;

[0043] Figure 6A and 6B Example of a processor unit;

[0044] Figure 7A and 7B Another example of a processor unit;

[0045] Figures 8A-8D A graphical example of similarity values;

[0046] Figures 9A-9D Another graphical example of similarity values. Detailed Implementation

[0047] The detailed description below, taken in conjunction with the accompanying drawings, serves as a description of various different configurations. This detailed description includes specific details to provide a thorough understanding of several different concepts. However, it will be apparent to those skilled in the art that these concepts can be implemented without these specific details. Several aspects of the apparatus and method are described by various different blocks, functional units, modules, elements, circuits, steps, processes, algorithms, etc. (collectively, “elements”). Depending on the specific application, design constraints, or other reasons, these elements may be implemented using electronic hardware, computer programs, or any combination thereof.

[0048] Hearing aids can be or include devices adapted to improve or enhance a user's hearing ability by receiving sound signals from the user's environment, generating corresponding audio signals, possibly modifying the audio signals, and providing the possibly modified audio signals as audible signals to at least one ear of the user. Improving or enhancing a user's hearing ability may include compensating for the specific hearing loss of an individual user. "Hearing aid" may also mean a device adapted to electronically receive audio signals, such as a wearable device, headset, or earphone, which may modify the audio signals and provide the possibly modified audio signals as audible signals to at least one ear of the user. The audible signals may be provided as acoustic signals radiated into the user's outer ear, or as acoustic signals transmitted as mechanical vibrations through the bone structures of the user's head and / or through the user's middle ear to the user's inner ear, or as electrical signals transmitted directly or indirectly to the user's cochlear nerve and / or auditory cortex.

[0049] Hearing aids are suitable for wearing in any known manner. This may include: i) placing the hearing aid unit behind the ear (having a tube to guide acoustic signals into the ear canal or having a receiver / speaker positioned close to or within the ear canal and connected via a wire (or wirelessly) to the behind-the-ear unit), such as behind-the-ear hearing aids; and / or ii) placing the hearing aid wholly or partially within the user's auricle and / or ear canal, such as in-the-ear (ITE) or in-the-canal (ITC) / deep-in-the-canal (DIC) hearing aids; or iii) positioning the hearing aid unit to be connected to a fixation device implanted into the skull, such as a bone-anchored hearing aid or a cochlear implant; or iv) positioning the hearing aid unit as a wholly or partially implanted unit, such as a bone-anchored hearing aid or a cochlear implant system. Hearing aids may be implemented in a single unit (shell) or in multiple units individually connected to each other.

[0050] A “hearing system” refers to a system comprising one or two hearing aids, and a “binaural hearing system” refers to a system comprising two hearing aids, wherein the hearing aids are adapted to provide audio signals to both ears of a user in a cooperative manner. A hearing system or a binaural hearing system may also include one or more assistive devices that communicate with at least one hearing aid, which affect the operation of the hearing aid and / or benefit from its functionality. A wired or wireless communication link is established between at least one hearing aid and the assistive device to allow the exchange of information (such as control and status signals, possibly audio signals). The assistive device may include at least one of the following: a remote control, a remote microphone, an audio gateway device, a wireless communication device such as a mobile phone (e.g., a smartphone) or a tablet computer or another device (e.g., including a graphical interface), a broadcasting system, a car audio system, a music player, or a combination thereof. The audio gateway device may be adapted to receive multiple audio signals, such as from an entertainment device such as a TV or music player, a telephone device such as a mobile phone, or a computer such as a PC. The assistive device may also be adapted (e.g., enabling the user) to select and / or combine appropriate signals from the received audio signals (or combinations of signals) to transmit to at least one hearing aid. The remote control is suitable for controlling the functions and operation of at least one hearing aid. The functions of the remote control can be implemented in a smartphone or other (e.g., portable) electronic device, which may run an application (APP) to control the functions of at least one hearing aid.

[0051] Generally, a hearing aid includes i) a receiving unit, such as a microphone, for receiving acoustic signals from the user's surroundings and providing a corresponding input audio signal, and / or ii) a receiving unit for electronically receiving the input audio signal. The hearing aid also includes a signal processing unit for processing the input audio signal and an output unit for providing an audible signal to the user based on the processed audio signal.

[0052] The receiving unit may include multiple input microphones, for example, for providing direction-dependent audio signal processing. Such directional microphone systems are adapted to (relatively) amplify a target acoustic source among a large number of acoustic sources in a user's environment and / or attenuate other sound sources (such as noise). In one aspect, the directional system is adapted to detect (e.g., adaptively detect) the direction from which a specific portion of the microphone signal originates. This can be achieved using methods conventionally known. The signal processing unit may include an amplifier adapted to apply a frequency-dependent gain to the input audio signal. The signal processing unit may also be adapted to provide other related functions such as compression, noise reduction, etc. The output unit may include an output converter, such as a speaker / receiver for providing airborne acoustic signals transdermally or percutaneously to the skull, or a vibrator for providing structure-borne or fluid-borne acoustic signals. In some hearing aids, the output unit may include one or more output electrodes, such as those in a cochlear implant, for providing electrical signals.

[0053] Cochlear implants typically include: i) an external portion for picking up and processing sound from the environment and determining a pulse sequence for electrode stimulation based on the current input sound; ii) a (typically wireless, such as inductive) communication link for simultaneously transmitting information about the stimulation sequence and transmitting energy to the implant portion; and iii) an implant portion that enables stimulation to be generated and applied to multiple electrodes, which may be implanted at different locations in the cochlea, thereby enabling stimulation of different frequencies within the auditory range. Such systems are described, for example, in US 4,207,441 and US 4,532,930.

[0054] On the one hand, hearing aids include multi-electrode arrays, for example, in the form of a carrier containing multiple electrodes adapted to be located in the cochlea and close to the user's auditory nerve. This carrier is preferably made of a flexible material to enable proper positioning of the electrodes in the cochlea, allowing the electrodes to be inserted into the recipient's cochlea. Preferably, the individual electrodes are spatially distributed along the length of the carrier, thereby providing a corresponding spatial distribution along the cochlear nerve in the cochlea when the carrier is inserted into the cochlea.

[0055] Now for reference Figure 1 The illustration shows a cochlear implant system 1, which includes an external unit 2 capable of percutaneous communication with the user's skin and an implantable unit 4. The implantable unit 4 is connected to an electrode array 5, which is configured to be inserted into the user's cochlea 6. The electrode array may include a plurality of electrodes 7.

[0056] Figure 2 A cochlear implant system 1 is shown, comprising a microphone unit 20 configured to receive acoustic signals and transmit audio signals based on the acoustic signals. The cochlear implant system 1 includes a processor unit 10, which includes a filter bank 22 configured to receive audio signals. The filter bank 22 includes multiple bandpass channels (ch1-chN) configured to divide the audio signals into multiple bandpass audio signals (BAS), each BAS(ch) of the multiple bandpass audio signals being passed to a first time feature extractor 24 and a second time feature extractor 26 of the processor unit 10. The first time feature extractor is configured to determine an initial set of time features of one bandpass audio signal from the multiple bandpass audio signals, and the second time feature extractor is configured to perform window filtering on the bandpass audio signal and determine a secondary set of time features of the window-filtered bandpass audio signal. In this specific example, the initial set of time features includes the first starting event time T1(ch) and the first amplitude A1(ch) of the initial event at the first starting event time T1(ch). In this specific example, the secondary set of time features includes the second starting event time T2(ch) of the secondary event and the second amplitude A2(ch) of the secondary event at the second starting event time.

[0057] Two sets of temporal features (A1(ch), T1(ch), A2(ch), T2(ch)) are passed to the stimulus generator 28, which is configured to determine a final set of temporal features (AF(ch), TF(ch)) based on the initial set of temporal features (A1(ch), T1(ch)) and the secondary set of temporal features (A2(ch), T2(ch)). The stimulus generator determines a similarity measure (SEM(ch)) for the final set of temporal features (AF(ch), TF(ch)) based on the initial set of temporal features and the secondary set of temporal features.

[0058] In this specific example, the final set of time features includes the final start event time TF(ch) and the final amplitude AF(ch), wherein the final start event time TF(ch) is determined based on the first start event time T1(ch) and the second start event time T2(ch), and the final amplitude AF(ch) is determined based on the first amplitude A1(ch) and the second amplitude A2(ch).

[0059] The processor unit 10 is configured to determine, based on a similarity measure (SEM(ch)), whether a final set of temporal features (AF(ch), TF(ch)) will be encoded into a stimulation pulse, or to prioritize stimulation pulses encoded based on the final set of temporal features (AF(ch), TF(ch)) based on the similarity measure (SEM(ch)), wherein the encoded stimulation pulses are transmitted to electrodes 7 of the electrode array 5, which are configured to apply electrical stimulation to a set of auditory nerve fibers of the recipient of the cochlear implant system based on the encoded stimulation pulses 61.

[0060] Figure 3A and 3B Examples of temporal feature extractors (24, 26) are shown. Both examples of temporal feature extractors include an event detector and an envelope detector. The event detector is configured to generate a sequence of events for a bandpass audio signal, consisting of multiple bandpass audio signals, based on the detection of a bandpass audio signal whose phase exceeds a phase threshold. Each event in the event sequence is generated based on the detection of the bandpass audio signal's phase exceeding the phase threshold, which can be any value between 0 and 2π. The envelope detector is configured to determine the amplitude of the bandpass audio signal among the multiple bandpass audio signals at a given time when the event is detected, i.e., the event start time. Figure 3AAn example of a first temporal feature extractor 24 is shown, configured to receive a bandpass audio signal BAS(ch) from a bandpass channel ch. This bandpass audio signal is divided into two parts. The first part is received by a first event detector 40, configured to determine a first start event time T1(ch) when the phase of the bandpass audio signal exceeds a phase threshold. The second part is received by a first envelope detector 41, configured to determine the amplitude A1(ch) of the bandpass audio signal at the first start event time T1(ch).

[0061] The first event detector 40 can be configured to determine a phase signal based on the bandpass audio signal over time, wherein a first start event time T1(ch) is determined in the phase signal.

[0062] The first envelope detector 41 can be configured to determine the amplitude envelope based on the bandpass audio signal over time, wherein, in the amplitude envelope, the first amplitude A1(ch) is determined at a time equal to the first start event time T1(ch).

[0063] Figure 3B An example of a second temporal feature extractor 26 is shown, configured to receive a bandpass audio signal BAS(ch) from a bandpass channel ch, which is passed through a window filter 44 before being split into two parts. The first part is received by a second event detector 42, configured to determine a second start event time T2(ch) when the phase of the windowed bandpass audio signal FBAS(ch) exceeds a phase threshold. The second part is received by a second envelope detector 43, configured to determine a second amplitude A2(ch) of the windowed bandpass audio signal FBAS(ch) at the second start event time T2(ch).

[0064] The second event detector 42 can be configured to determine a phase signal based on a windowed bandpass audio signal over time, wherein the second start event time T2(ch) is determined in the phase signal.

[0065] The second envelope detector 43 can be configured to determine the amplitude envelope based on the windowed bandpass audio signal over time, wherein the second amplitude A2(ch) is determined in the amplitude envelope at a time equal to the second start event time T2(ch).

[0066] Figure 4A and 4B An example is shown of how the cutoff frequency y of window filter 44 is determined by processor unit 10. Figure 4AIn this embodiment, processor unit 10 is configured to adjust the cutoff frequency y of the window filter 44 based on the center frequency 45 and / or bandwidth 45 of the bandpass channel ch that provides the bandpass audio signal BAS(ch) to be filtered by window filter 44. Window filter 44 may be a bandpass filter, and processor unit 10 is configured to determine the cutoff frequency of the bandpass filter 44 based on the center frequency 45 and / or bandwidth 45 of the bandpass channel ch that provides the bandpass audio signal BAS(ch) to be filtered by window filter 44. If window filter 44 is a low-pass filter, processor unit 10 is configured to determine the cutoff frequency of the low-pass filter 44 based on the center frequency 45 and / or bandwidth 45 of the bandpass channel ch. For example, if the center frequency of the bandpass channel is 1000 Hz and the bandwidth is 200 Hz, the cutoff frequency of the low-pass filter 44 may be set to approximately 1100 Hz, or the cutoff frequency of the bandpass filter 44 may be set to approximately 900 Hz and approximately 1100 Hz.

[0067] exist Figure 4B In this configuration, processor unit 10 is configured to adjust the cutoff frequency y of window filter 44 based on feedback 45 from second amplitude envelope detector 43. The feedback includes information about whether the window-filtered bandpass audio signal FBAS(ch) includes an envelope from which the amplitude can be determined. If the feedback 45 includes information that the amplitude cannot be determined, processor unit 10 adjusts the cutoff frequency y of window filter 44 until the amplitude can be determined.

[0068] Figure 5 An example is shown where the stimulus generator 28 is configured to determine a 50 similarity measure value SEM(ch) based on a value function between an initial set of time features (T1(ch), A1(ch)) and a secondary set of time features (T2(ch), A2(ch)).

[0069] The similarity measure (SEM) between an initial set of time features and a secondary set of time features can be determined by a value function between the first initial time feature (e.g., T1) in the initial set of time features and the second initial time feature (e.g., T2) in the secondary set of time features.

[0070] The similarity measure (SEM) between an initial set of time features and a secondary set of time features can be determined by a value function between the first-generation time feature (e.g., A1) in the initial set of time features and the second-generation time feature (e.g., A2) in the secondary set of time features.

[0071] The final time feature (TF, AF) in the final set of time features is determined by a weighting function in sections 51 and 52 when the difference between the first initial time feature in the initial set of time features and the second initial time feature in the secondary set of time features is lower than the threshold level TFHR for the initial event time feature or the threshold level AFHR for the amplitude feature. TFHR and AFHR are different. The weighting function includes a first weight for the first initial time feature in the initial set of time features and a second weight for the first initial time feature in the secondary set of time features. The second weight increases when the first weight decreases, and vice versa. The sum of the first and second weights is always 1. If the difference is equal to or higher than the time or amplitude feature threshold, the final time feature is equal to the first initial time feature in the initial set of time features. More specifically, in 51, when the difference between the first starting event time T1(ch) and the second starting event time T2(ch) is lower than the time feature threshold level TFHR, the final starting event time TF(ch) is determined by a weighting function comprising a first weight and a second weight of the first starting event time, wherein the sum of the first weight and the second weight is always 1. In 51, when the difference between the first starting event time T1(ch) and the second starting event time T2(ch) is equal to or higher than the starting event time feature threshold TFHR, the final starting event time TF(ch) is equal to the first starting event time T1(ch).

[0072] In step 52, when the difference between the first amplitude A1(ch) and the second amplitude A2(ch) is lower than the amplitude characteristic threshold level AFHR, the final amplitude AF(ch) is determined by a weighting function comprising a first weight of the first amplitude and a second weight of the second amplitude, wherein the sum of the first weight and the second weight is always 1. In step 52, when the difference between the first amplitude A1(ch) and the second amplitude A2(ch) is equal to or higher than the amplitude time characteristic threshold AFHR, the final amplitude AF(ch) is equal to the first amplitude A1(ch).

[0073] Figure 6A and 6B A more specific example is shown of how processor unit 10 determines the 50 similarity metric values ​​SEM(ch). Figure 6A In this process, the similarity measure between an initial set of time features and a secondary set of time features can be determined as follows: A similarity measure 61 is determined between a first amplitude A1(ch) and a second amplitude A2(ch), and a time difference measure 60 is determined between a first starting event time T1(ch) and a second starting event time T2(ch). Then, the similarity measure value SEM(ch) is determined through a value function 62 that includes the output 63 from the similarity measure 61 and the output 64 from the time difference measure 60. Figure 6BIn this process, the similarity measure 61 may include determining the cross-correlation 65 between the first amplitude A1(ch) and the second amplitude A2(ch) and determining the cross-covariance 66 between the first amplitude A1(ch) and the second amplitude A2(ch). Then, the similarity measure 61 may be determined by a linear combination 67 of the cross-correlation 65 and the cross-covariance 66. The linear combination 67 may be the summation or multiplication of the cross-correlation 65 and the cross-covariance 66.

[0074] Figure 7A and 7B An example is shown where the processor unit 10 is configured to determine, based on the similarity measure SEM(ch), whether the final set of temporal features (AF(ch), TF(ch)) will be encoded into the stimulus pulse, or to prioritize the stimulus pulse encoded based on the final set of temporal features (AF(ch), TF(ch)) based on the similarity measure SEM(ch). Figure 7A An example is shown where the processor unit 10 has determined that the final set of time characteristics of the bandpass channels ch1 and chN (30, 70) will be encoded into the stimulation pulse 71 and transmitted to the electrode array 5. Figure 7B An example is shown where processor unit 10 encodes multiple sets of final time features of multiple bandpass channels (ch1, ch2, and chN) into stimulation pulses (P1, P2, PN) and prioritizes each of the multiple stimulation pulses (30, 72). In this example, stimulation pulses P2 and PN are simultaneously activated by the final start event times TF(ch2) and TF(chN), respectively. This creates a timing conflict between P2 and PN because PN has a higher priority than P2, and stimulation pulse P2 is discarded and not transmitted to electrode array 5. In another example, stimulation pulse P2 is delayed slightly so that there is no timing conflict between P2 and PN.

[0075] Figures 8A to 8D A graphical example is shown showing how the similarity measure value changes over time for a low-frequency bandpass audio signal. Figures 8A-8D Align the time frame of each curve in the image. Figure 8A The figure shows the envelope of the bandpass audio signal BAS and the envelope of the window-filtered bandpass audio signal FBAS. Furthermore, the figure illustrates the variations of the first amplitude A1 and the second amplitude along the envelopes of the bandpass audio signal BAS and the window-filtered bandpass audio signal FBAS, respectively. Additionally, the first start event time T1 and the second start event time T2 are shown on the same curve. Figure 8B The curves for the similarity measure of the first amplitude A1 and the second amplitude A2 are shown, where, in this example, the similarity measure includes a combination of the cross-correlation and cross-covariance of the first and second amplitudes. Figure 8C The curve showing the time difference φ between the first and second starting event times is illustrated. Figure 8D The curves showing the changes in the similarity measure SEM relative to the amplitude (A1, A2) and the time of the initial event (T1, T2) are shown. It can be seen that the similarity measure SEM decreases as the time difference metric φ increases, and the similarity measure SEM also decreases if the similarity metric cross+cov decreases.

[0076] Figures 9A-9D It shows the relationship with Figures 8A-8D A similar example is shown, where the bandpass audio signal corresponds to a higher frequency range.

[0077] Unless explicitly stated otherwise, the singular forms “a” and “the” used herein include the plural forms (i.e., meaning “at least one”). It should be further understood that the terms “having,” “comprising,” and / or “including” as used in the specification indicate the presence of the stated features, integers, steps, operations, elements, and / or components, but do not preclude the presence or addition of one or more other features, integers, steps, operations, elements, components, and / or combinations thereof. It should be understood that, unless explicitly stated otherwise, when an element is referred to as “connected” or “coupled” to another element, it can be a direct connection or coupling to the other element, or there may be intermediate inserting elements. The term “and / or” as used herein includes any and all combinations of one or more of the listed related items. Unless explicitly stated otherwise, the steps of any method disclosed herein do not necessarily have to be performed in the exact order disclosed.

[0078] It should be understood that references to "an embodiment," "an embodiment," "an aspect," or "may" in this specification mean that a particular feature, structure, or characteristic described in connection with that embodiment is included in at least one embodiment of the invention. Furthermore, specific features, structures, or characteristics may be suitably combined in one or more embodiments of the invention.

[0079] The foregoing description is provided to enable those skilled in the art to implement the various aspects described herein. Various modifications will be apparent to those skilled in the art, and the general principles defined herein may be applied to other aspects. Unless expressly stated otherwise, an element referred to in the singular does not mean "one and only one," but rather "one or more." Unless expressly stated otherwise, the term "some" means one or more.

[0080] Therefore, the scope of this invention should be determined based on the claims.

Claims

1. A cochlear implantation system, comprising: - A microphone unit configured to receive acoustic signals and transmit audio signals based on those acoustic signals; - An electrode array comprising multiple electrodes; - Processor unit, including --Filter bank, which includes multiple bandpass channels configured to divide an audio signal into multiple bandpass audio signals; --First-time feature extractor, configured to determine an initial set of time features for one of a plurality of bandpass audio signals; --A second time feature extractor is configured to perform window filtering on the bandpass audio signal and determine a secondary set of time features of the window-filtered bandpass audio signal among the plurality of bandpass audio signals; --Stimulus generator, configured to determine a final set of time features based on an initial set of time features and a secondary set of time features, and to determine similarity measures between the initial set of time features and the secondary set of time features for the final set of time features. The processor unit is configured to determine whether the final set of temporal features will be encoded into the stimulus pulse based on the similarity metric value, or to prioritize the stimulus pulse encoded based on the final set of temporal features based on the similarity metric value. In this process, an encoded stimulation pulse is transmitted to one of a plurality of electrodes, which is configured to apply electrical stimulation to a group of auditory nerve fibers of the recipient of the cochlear implant system based on the encoded stimulation pulse.

2. The cochlear implant system of claim 1, wherein, The processor unit is configured to determine a final set of time features for each of a plurality of bandpass audio signals and to determine a similarity measure for each final set of time features, and to determine whether each final set of time features will be encoded into a stimulation pulse based on the similarity measure, wherein each of the encoded plurality of stimulation pulses is transmitted to a set of electrodes among a plurality of electrodes.

3. The cochlear implant system of claim 1 or 2, wherein, An initial set of time features includes a first start event time of the initial event detected in the initial event sequence of the bandpass audio signal and a first amplitude of the initial event at the first start event time. A secondary set of time features includes a second start event time of the secondary event detected in the secondary event sequence of the window-filtered bandpass audio signal and a second amplitude of the secondary event at the second start event time. A final set of time features includes a final start event time and a final amplitude, wherein the final start event time is determined based on the first start event time and the second start event time, and the final amplitude is determined based on the first amplitude and the second amplitude.

4. The cochlear implant system according to claim 3, wherein, The first-time feature extractor includes: - A first event detector, configured to determine a first start event time of the bandpass audio signal among a plurality of bandpass audio signals; - A first envelope detector, configured to determine the first amplitude of one of a plurality of bandpass audio signals at a first initiation event time; The second temporal feature extractor includes: - A second event detector, configured to determine the second start event time of a windowed bandpass audio signal among a plurality of bandpass audio signals; - A second envelope detector is configured to determine the second amplitude of a window-filtered bandpass audio signal among a plurality of bandpass audio signals at the second start event time.

5. The cochlear implant system according to claim 1, wherein, The window filter includes a low-pass filter or a band-pass filter configured to filter the bandpass audio signal.

6. The cochlear implant system according to claim 1, wherein, The window-filtered bandpass audio signal includes the oscillation envelope.

7. The cochlear implant system according to claim 6, wherein, The cutoff frequency of the window filter is determined by the processor unit, such that the oscillation envelope appears in the window-filtered bandpass audio signal.

8. The cochlear implant system according to claim 7, wherein, The cutoff frequency of the window filter is determined by the processor unit based on the center frequency and / or bandwidth of the bandpass channel that provides the bandpass audio signal among multiple bandpass channels.

9. The cochlear implant system according to claim 1, wherein, The similarity measure between the initial set of time features and the secondary set of time features is determined by a value function between the initial set of time features and the secondary set of time features.

10. The cochlear implant system according to claim 3, wherein, The similarity measure between the initial set of temporal features and the secondary set of temporal features is determined as follows: - Determine a similarity measure between the first and second amplitudes; - Determine the time difference measure between the first initiation event time and the second initiation event time; The similarity measure value is determined by a value function that includes the similarity measure and the time difference measure.

11. The cochlear implant system according to claim 10, wherein, The similarity measure includes: - Determine the cross-correlation between the first amplitude and the second amplitude; - Determine the cross-covariance between the first amplitude and the second amplitude; The similarity measure is determined by a linear combination of cross-correlation and cross-covariance.

12. The cochlear implant system according to claim 2 or 9, wherein, The first time feature is the time of the first starting event, and the second time feature is the time of the second starting event.

13. The cochlear implant system according to claim 1, wherein, The similarity measure between the initial set of time features and the secondary set of time features is determined by a real-valued function between the first time feature in the initial set of time features and the second time feature in the secondary set of time features.

14. The cochlear implant system according to claim 2 or 13, wherein, The first time feature is the first amplitude, and the second time feature is the second amplitude.

15. The cochlear implant system according to claim 1, wherein, The final set of time features is determined based on a function comprising a first weighted sum of the initial set of time features and a second weighted sum of the secondary set of time features, wherein the first and second weights are determined based on the center frequency and / or bandwidth of the corresponding bandpass channel of the bandpass audio signal in the plurality of bandpass channels.

16. The cochlear implant system according to claim 4, wherein, An event detector is configured to generate an event sequence of one of a plurality of bandpass audio signals, wherein each event in the event sequence is generated when the phase of the bandpass audio signal exceeds a phase threshold, wherein the phase threshold is any value between 0 and 2π.

17. The cochlear implant system according to claim 1, wherein, The final time feature FF(ch) in the final set of time features is related to the first initial time feature in the initial set of time features and the second initial time feature in the secondary set of time features in the following way: Wherein, 1-W(Fc(ch),BW(ch)) is the first weighting factor that depends on the center frequency Fc(ch) and / or bandwidth BW(ch) of the bandpass channel ch among multiple bandpass channels, W(Fc(ch),BW(ch)) is the second weighting factor, F1(ch) is the first initial time feature, F2(ch) is the second initial time feature, TFHR is the threshold level of the initial event time feature, and AFHR is the threshold level of the amplitude feature.

18. The cochlear implant system according to claim 17, wherein, The second weighting factor increases with decreasing center frequency and decreases with increasing center frequency.

Images