System and method for indicating neural response

Abstract

A computing system includes a processing unit implementing an artificial neural network. The artificial neural network generates, at a single output node, an output indicating whether a measurement performed after stimulating a neural region of an individual includes a neural response. A method includes receiving, at an artificial neural network in a computing system, a value indicative of a measurement performed after providing stimulation to a neural region of an individual, and generating, at a single output node of the artificial neural network, an output indicative of whether the measurement includes a neural response.

Description

[0001] Cross - reference to related applications

[0002] This patent application claims the benefit of U.S. Provisional Patent Application No. 63 / 421,339, filed on November 1, 2022, which is incorporated herein by reference in its entirety. Technical field

[0003] The present disclosure relates to systems and methods for indicating the neural response of an individual in a computing system. Background art

[0004] In recent decades, medical devices have provided a wide range of therapeutic benefits to recipients. Medical devices can include internal or implantable components / devices, external or wearable components / devices, or combinations thereof (e.g., devices having external components that communicate with implantable components). Medical devices, such as traditional hearing aids, partially or fully implantable hearing prostheses (e.g., bone conduction devices, mechanical stimulators, cochlear implants, etc.), pacemakers, defibrillators, functional electrical stimulation devices, and other medical devices, have been successful for many years in performing life - saving and / or lifestyle - improving functions and / or recipient monitoring.

[0005] Over the years, the types of medical devices and the range of functions performed by them have increased. For example, many medical devices sometimes referred to as "implantable medical devices" now typically include one or more instruments, devices, sensors, processors, controllers, or other functional mechanical or electrical components permanently or temporarily implanted in a recipient. These functional devices are generally used for diagnosing, preventing, monitoring, treating, or managing diseases / injuries or their symptoms, or for researching, replacing, or modifying anatomical structures or physiological processes. Many of these functional devices utilize power and / or data received from an external device, which is part of or operates in cooperation with the implantable component. Summary of the invention

[0006] According to a first embodiment disclosed herein, a computing system includes at least one processing unit that implements an artificial neural network, wherein the artificial neural network generates, at a single output node, an output indicating whether a measurement performed after stimulating a neural region of an individual includes a neural response.

[0007] According to a second embodiment disclosed herein, a method includes: receiving, at an artificial neural network in a computing system, a value indicating a measurement performed after providing a stimulus to a neural region of an individual; and generating, at a single output node of the artificial neural network, an output indicating whether the measurement includes a neural response.

[0008] According to a third embodiment disclosed herein, a non-transitory computer-readable storage medium includes computer-readable instructions stored thereon that, when executed by a computing system, cause the computing system to: receive, at an input node of an artificial neural network in the computing system, pixels of an image of a locus of measurements performed after stimulating an individual's auditory nerve; and generate, using the artificial neural network, an indication of whether the measurements include a neural response based on the pixels.

[0009] According to a fourth embodiment disclosed herein, a method includes: sampling a signal indicative of measurements performed after stimulating an individual's auditory nerve to generate a sampled signal; performing a Fourier transform on the sampled signal to extract frequency components of the sampled signal; receiving, at an input node of an artificial neural network in a computing system, the frequency components of the sampled signal; and generating, using the artificial neural network, an output indicative of whether the measurements include a neural response. BRIEF DESCRIPTION OF THE DRAWINGS

[0010] Figure 1A FIG. 1 depicts a schematic diagram of an exemplary cochlear implant system that can be configured to implement aspects of the techniques presented herein, in accordance with some exemplary embodiments.

[0011] Figure 1B FIG. 2 depicts Figure 1A a block diagram of the cochlear implant system of FIG. 1.

[0012] Figure 2 FIG. 3 depicts an example of an artificial neural network (ANN) that can be used to determine whether measurements performed after stimulating a recipient's neural region include a neural response to the stimulation.

[0013] Figure 3 FIG. 4 depicts an example of a flowchart of operations that can be performed to train an artificial neural network (ANN) to determine whether measurements of neural activity performed after stimulating a recipient's neural region include a neural response.

[0014] Figure 4 FIG. 5 depicts an example of a flowchart of operations that can be performed to use an artificial neural network (ANN) trained according to the operations of FIG. 4 to determine whether measurements of neural activity performed after applying a stimulation to a recipient's neural region include a neural response to the stimulation. Figure 3 FIG. 6 depicts an example of a flowchart of operations that can be performed to determine a stimulation level that elicits a neural response in a recipient's auditory nerve within a search range of stimulation levels.

[0015] Figure 5 FIG. 7 depicts an example of a flowchart of operations that can be performed to determine a stimulation level that elicits a neural response in a recipient's auditory nerve within a search range of stimulation levels.

[0016] Figure 6A flowchart depicting an example of operations that can be performed to determine whether a measurement performed after applying a stimulus to a neural region of a recipient includes a neural response by providing pixels of a measured trajectory to an artificial neural network (ANN).

[0017] Figure 7 A flowchart depicting an example of operations that can be performed to determine whether a measurement performed after applying a stimulus to a neural region of a recipient includes a neural response by providing frequency components of a sample of the measurement to an artificial neural network (ANN).

[0018] Figure 8 A flowchart depicting an example of operations that can be performed to determine whether a measurement performed after applying a stimulus to a neural region of a recipient includes a neural response by providing samples of a signal indicative of the measurement to an artificial neural network (ANN).

[0019] Figure 9 An example of a suitable computing system that can perform any of the operations or functions disclosed herein is shown. DETAILED DESCRIPTION

[0020] There can be many different causes of an individual's hearing loss. Sensorineural hearing loss is the cause of deafness in many people. Sensorineural hearing loss is caused by the absence or damage of hair cells in the cochlea that convert sound signals into nerve impulses. Due to the impairment or absence of the mechanism for naturally generating nerve impulses from sound, individuals with sensorineural hearing loss cannot obtain suitable benefits from conventional hearing aids. Cochlear implant systems are an auditory prosthesis that has been developed to potentially address sensorineural hearing loss. Cochlear implant systems bypass the hair cells in the cochlea and directly deliver electrical stimuli to the auditory nerve fibers via an implanted electrode assembly. The electrical stimuli enable the brain to perceive an auditory sensation similar to natural hearing that is normally delivered to the auditory nerve fibers.

[0021] Cochlear implant systems have traditionally included an external speech processor unit worn on the recipient's body and a receiver / stimulator unit implanted in the recipient. The external speech processor unit detects external sounds and converts the detected external sounds into coded signals through a speech processing strategy. The coded signals are sent via a transcutaneous link to the implanted receiver / stimulator unit. The receiver / stimulator unit processes the coded signals to generate a series of stimulation sequences, which are then directly applied to the auditory nerve via a series arrangement or array of electrodes positioned within the cochlea.

[0022] An external speech processor unit and an implanted receiver / stimulator unit can be combined to produce a fully implantable cochlear implant system capable of operating without an external device for at least a certain period of time. In such an implant, the microphone is implanted within the recipient's body, for example, in the ear canal or within the stimulator unit. The detected sound is directly processed by the speech processor within the stimulator unit, and subsequent stimulation signals are delivered without any percutaneous transmission of the signals.

[0023] Data is obtained from the components of the cochlear implant system to enable detection and confirmation of the normal operation of the cochlear implant system. The data can also be obtained from the cochlear implant system to allow optimization of the stimulation parameters to meet the needs of different recipients, and the data includes data related to the response of the auditory nerve to stimulation. The cochlear implant system typically has the ability to communicate with an external device, for example, to receive program upgrades, perform implant queries, and read and / or change the operating parameters of the cochlear implant system.

[0024] Determining the response of the auditory nerve to stimulation has been addressed in conventional systems, but with limited success rates. Typically, after surgically implanting the implantable components of the cochlear implant system, the cochlear implant system is adapted or customized to meet the specific recipient's needs. The customization process can involve collecting and determining patient-specific parameters, such as the threshold level (T-level) and maximum comfort level (C-level) for each stimulation channel in the cochlear implant system. In previously known systems, the customization process was performed manually by applying stimulation pulses to each stimulation channel and receiving an indication from the recipient regarding the level and comfort of the resulting sound. For cochlear implant systems with a large number of stimulation channels, the customization process is time-consuming and subjective because the customization process relies heavily on the recipient's subjective impression of the stimulation rather than objective measurements.

[0025] The manual execution of the customization process is further limited for children and prelingually or congenitally deaf patients who cannot provide an accurate impression of the resulting hearing. For these recipients, the adaptation of the cochlear implant system may be suboptimal. An incorrectly adapted cochlear implant system may result in the recipient not obtaining the optimal benefit from the cochlear implant system. For example, an incorrectly adapted cochlear implant system in a child may directly impede the child's language and hearing development. Therefore, there is a need to obtain objective measurements of patient-specific data, such as the minimum threshold level (T-level) and maximum comfort level (C-level) of the stimulation channels in the cochlear implant system, especially in cases where accurate subjective measurements are not possible.

[0026] One technique for querying the performance of a cochlear implant system and making objective measurements of patient-specific data (e.g., T-levels and C-levels) is to directly measure the response of the auditory nerve to electrical stimulation. Direct measurement of the neural response, often referred to as electrically evoked compound action potentials (ECAPs) in the context of cochlear implant systems, provides an objective measurement of the response of the auditory nerve to electrical stimulation. After electrical stimulation, the neural response is caused by the superposition of neural responses at the outside of the axon. Measurements can be made from within the cochlea in response to various stimuli. Measurements are made to determine whether a neural response has occurred. This measurement is an objective measurement of neural activity. Typically, neural activity of the auditory nerve caused by stimulation given at one electrode in the implantable component of the cochlear implant system is measured at another electrode of the implantable component (e.g., at an adjacent electrode). The measurement is typically transmitted to a system located externally.

[0027] Cochlear implant systems typically have the ability to generate stimulation using one electrode and measure neural activity after stimulation at an adjacent electrode. When the stimulus is large enough to cause an electrically evoked compound action potential (ECAP) in the auditory nerve, the waveform of the measured potential presents a unique shape that can be seen by the human eye. The minimum stimulation amplitude required to generate an ECAP can be called the threshold of the neural response. The conventional technique for determining the neural response of a recipient of a cochlear implant system is a manual process that involves using electrodes in an implantable component to provide electrical stimulation to the recipient's auditory nerve with an increased amplitude, and then analyzing the measurements taken after the electrical stimulation for the ECAP. This manual process for determining neural responses is time-consuming and is affected by different people's expertise and experience. Therefore, it is desirable to provide an automated system for detecting neural responses. It is also desirable to provide a simplified and efficient system that can be used in a clinic with a large patient base.

[0028] According to some embodiments disclosed herein, systems and methods are provided that are configured to receive, at an artificial neural network (ANN) in a computing system (e.g., in an electrophysiological response measurement system), values indicative of measurements performed after providing a stimulus to a neural region of an individual, and generate, at a single output node of the artificial neural network, an output indicative of whether the measurement includes a neural response. For example, the values may include values of signals indicative of measurements of neural activity performed after delivering an electrical stimulus from an electrode in an implant system (e.g., a cochlear implant system) to the auditory nerve of a recipient of the implant system. According to other embodiments disclosed herein, systems and methods are provided that are configured to receive, at an input layer of an artificial neural network (ANN), pixels, samples, or frequency components of a signal indicative of a measurement of neural activity performed after stimulating a neural region of an individual, and use the ANN to generate an indication of whether the measurement includes a neural response. Advantageously, the present technology is capable of providing a binary output indicative of whether a neural response has been evoked, and thus, the present technology can be more generally used across a range of different types of patients while also simplifying the clinical process. Additional details of these and other embodiments are disclosed below.

[0029] For ease of description only, the techniques presented herein are described herein primarily with reference to an illustrative medical device (i.e., a cochlear implant system). However, it should be understood that the techniques presented herein may also be used with a variety of other medical devices that may benefit from the teachings used herein while providing a wide range of therapeutic benefits to recipients, patients, or other users. For example, any of the techniques described herein for one type of hearing prosthesis (e.g., a cochlear implant system) corresponds to the disclosure of another embodiment of using such teachings with another hearing prosthesis and also using such teachings with other electrical analog auditory prostheses (e.g., auditory brain stimulators), etc., the other hearing prosthesis including bone conduction devices (transcutaneous, active transcutaneous, and / or passive transcutaneous), middle ear auditory prostheses, direct acoustic stimulators. The techniques presented herein may also be used with vestibular devices (e.g., vestibular implants), visual devices (i.e., bionic eyes), sensors, pacemakers, drug delivery systems, defibrillators, functional electrical stimulation devices, catheters, seizure devices (e.g., devices for monitoring and / or treating seizure events), sleep apnea devices, electroporation devices, etc.

[0030] Although the teachings detailed herein are mainly described with respect to a hearing prosthesis, in accordance with the above, it should be noted that any disclosure herein regarding a hearing prosthesis corresponds to a disclosure of another embodiment that utilizes the associated teachings with respect to any other prosthesis mentioned herein (whether a hearing prosthesis or a sensory prosthesis, such as a retinal prosthesis). In this regard, unless explicitly indicated and / or unless the art is unable to achieve this, any disclosure herein regarding inducing a hearing perception corresponds to a disclosure of inducing other types of neural perceptions (e.g., visual / vision perception, tactile perception, olfactory perception, or gustatory perception) in other embodiments. Any disclosure herein of an apparatus, system, and / or method for or causing stimulation of the auditory nerve corresponds to a disclosure of a similar stimulation of the optic nerve using similar components, methods, and systems.

[0031] Figure 1A is a schematic diagram of an exemplary cochlear implant system 100 configured to implement aspects of the technology presented herein. Figure 1B is Figure 1A a block diagram of the cochlear implant system 100. For ease of illustration, Figure 1A and Figure 1B are described together herein. The cochlear implant system 100 includes an external component 102 and an internal / implantable component 104. The external component 102 is directly or indirectly attached to the recipient's body and generally includes an external coil 106 and a magnet ( Figure 1A-1B not shown in Figure 1B ) that is generally fixed relative to the external coil 106. The external component 102 also includes one or more input elements / devices 113 ( Figure 1B shown) for receiving input signals at the sound processing unit 112. In this example, the one or more input devices 113 include a sound input device 108 (e.g., a microphone, a pick-up coil, etc. positioned by the recipient's auricle 110) configured to capture / receive input signals, one or more auxiliary input devices 109 (e.g., an audio port such as a direct audio input (DAI), a data port such as a universal serial bus (USB) port, a cable port, etc.), and a wireless transmitter / receiver (transceiver) 111, each located in, on, or near the sound processing unit 112.

[0032] The sound processing unit 112 also includes, for example, at least one power supply 107, a radio frequency (RF) transceiver 121, and a processing module 125. The processing module 125 includes a number of elements, including an environment classifier 131, a sound processor 133, and an individualized own voice detector 134. Each of the environment classifier 131, the sound processor 133, and the individualized own voice detector 134 may be formed by one or more processors (e.g., one or more digital signal processors (DSPs), one or more processing cores, etc.), firmware, software, etc. arranged to perform the operations described herein. That is, the environment classifier 131, the sound processor 133, and the individualized own voice detector 134 may each be implemented as a firmware element, partially or fully implemented with digital logic gates in one or more application specific integrated circuits (ASICs), partially or fully implemented with software, etc.

[0033] In Figure 1A and Figure 1B example, the sound processing unit 112 is a behind-the-ear (BTE) sound processing unit configured to be attached to and worn adjacent to the recipient's ear. However, it should be understood that the sound processing unit 112 may have other arrangements, such as an over-the-ear (OTE) processing unit (e.g., having a generally cylindrical shape and configured to be magnetically coupled to a component of the recipient's head), etc., a mini or micro BTE unit, an in-the-ear canal unit configured to be located in the recipient's ear canal, a body-worn sound processing unit, etc.

[0034] In Figure 1A and Figure 1B exemplary embodiment, the implantable component 104 includes an implant body (main module) 114, a lead region 116, and an intracochlear stimulation assembly 118, all configured to be implanted beneath the recipient's skin / tissue (tissue) 105. The implant body 114 generally includes an airtight sealed housing 115 in which an RF interface circuit system 124 and a stimulator unit 120 are provided. The implant body 114 also includes an internal / implantable coil 122, which is generally outside the housing 115 but is connected to the RF interface circuit system 124 via an airtight feedthrough ( Figure 1B not shown in

[0035] The stimulation assembly 118 is configured to be at least partially implanted in the recipient's cochlea 137. The stimulation assembly 118 includes a plurality of longitudinally spaced intracochlear electrical stimulation contacts (e.g., electrodes) 126, which together form a contact or electrode array 128 for delivering electrical stimulation (current) to the recipient's cochlea. The stimulation assembly 118 extends through an opening (e.g., a cochlear fenestration, round window, etc.) in the recipient's cochlea and has an airtight feedthrough via the lead region 116 and ( Figure 1B(not shown in the figure) is connected to the proximal end of the stimulator unit 120. The lead region 116 includes a plurality of conductors (wires) that electrically couple the stimulation contacts 126 to the stimulator unit 120.

[0036] As noted, the cochlear implant system 100 includes an external coil 106 and an implantable coil 122. The coils 106 and 122 are generally wire antenna coils each including multiple turns of electrically insulated single-strand or multi-strand wire. Generally, the magnets are fixedly positioned relative to each of the external coil 106 and the implantable coil 122. In some embodiments, the external component 102 and / or the implantable component 104 may include magnet assemblies each having more than one magnetic component. The magnets fixed relative to the external coil 106 and the implantable coil 122 assist in the operational alignment of the external coil with the implantable coil. This operational alignment of the coils 106 and 122 enables the external component 102 to transmit data and possibly power to the implantable component 104 via a tightly coupled wireless link formed between the external coil 106 and the implantable coil 122. In certain examples, the tightly coupled wireless link is a radio frequency (RF) link. However, various other types of energy transfer (e.g., infrared (IR), electromagnetic, capacitive, and inductive transfer) may be used to transfer power and / or data from the external component to the implantable component, and thus, Figure 1B only one exemplary arrangement is shown.

[0037] As described above, the sound processing unit 112 includes a processing module 125. The processing module 125 is configured to convert an input audio signal into a stimulation control signal 136 for stimulating the recipient's first ear (i.e., the processing module 125 is configured to perform sound processing on the input audio signal received at the sound processing unit 112). In other words, the sound processor 133 (e.g., one or more processing elements implementing firmware, software, etc.) is configured to convert the captured input audio signal into a stimulation control signal 136 representing electrical stimulation to be delivered to the recipient. The input audio signal that is processed and converted into the stimulation control signal 136 may be an audio signal received via the sound input device 108, a signal received via the auxiliary input device 109, and / or a signal received via the wireless transceiver 111.

[0038] In Figure 1BIn an embodiment, a stimulation control signal 136 is provided to an RF transceiver 121, which transcutaneously transmits the stimulation control signal 136 (e.g., in an encoded manner) to an implantable component 104 via an external coil 106 and an implantable coil 122. The stimulation control signal 136 is received at an RF interface circuit system 124 via the implantable coil 122 and provided (e.g., as N signals) to a stimulator unit 120. The stimulator unit 120 is configured to generate an electrical stimulation signal (e.g., a current signal) using the stimulation control signal 136 for delivery to a recipient's cochlea via one or more stimulation contacts 126 in an array 128. In this way, the cochlear implant system 100 electrically stimulates the recipient's auditory nerve cells in a manner that enables the recipient to perceive one or more components of an input audio signal, bypassing missing or defective hair cells that normally convert sound vibrations into neural activity.

[0039] Figure 1B Also shown is an electrophysiological response measurement system 160 communicatively coupled to the sound processor 133 via a connection (e.g., a cable). In some embodiments, the electrophysiological response measurement system 160 is a processor-based system such as a personal computer, server, workstation, etc., having one or more processors that execute software programs to perform the techniques disclosed herein. For example, the system 160 can generate signals that are used by the cochlear implant system 100 as stimuli to stimulate the recipient's auditory nerve via one or more stimulation contacts 126, receive measurements of neural activity in response to the stimuli from the cochlear implant system 100, and generate an indication of whether the measurements of neural activity include or do not include a neural response of the recipient's auditory nerve.

[0040] According to some embodiments disclosed herein, the electrophysiological response measurement system 160 includes a computer system that implements an artificial neural network (ANN). The ANN receives a representation (e.g., a vision-based or frequency-based representation) of measurements of neural activity performed after providing a stimulus to the recipient's auditory nerve and classifies the representation as including or not including a neural response to the stimulus. For example, if the measurement includes only noise, the ANN can determine that the measurement does not include a neural response. The ANN can be incorporated into a search algorithm that generates signals that are provided as stimuli at varying stimulus levels to the recipient's auditory nerve, receives measurements of neural activity in response to the stimuli, and determines whether the measurements include a neural response.

[0041] An artificial neural network (ANN) can include an input layer, one or more hidden layers, and an output layer. The input layer of the ANN includes input nodes. The number of input nodes in the input layer of the ANN can be selected based on the data provided to the input layer, as described in more detail below. The ANN can have any number of one or more hidden layers. The number of hidden layers in the ANN can be selected according to user preferences. Each of the hidden layers has one or more hidden nodes. The output layer of the ANN can include only a single output node.

[0042] Figure 2 FIG. depicts an example of an artificial neural network (ANN) that can be used to determine whether a measurement performed after stimulating a neural region of a recipient includes a neural response to the stimulation. ANN 200 includes an input layer, a hidden layer, and an output layer. Although Figure 2 only a single hidden layer is shown as an example, it should be understood that the ANN for implementing the techniques disclosed herein can include any number of hidden layers. In the example of ANN 200, the input layer includes four input nodes 201-204, and the hidden layer includes five hidden nodes 211-215. Figure 2 The number of nodes in the input layer and the hidden layer shown in FIG. are provided only as examples. It should be understood that each of the input layer and the hidden layer in the ANN for implementing the techniques disclosed herein can have any number of nodes (e.g., hundreds or thousands of nodes). The output layer of ANN 200 has only a single output node 220. Inputs 1-4 are the values provided to input nodes 201-204, respectively. The input provided to each of the hidden nodes (e.g., hidden nodes 211-215) and the input provided to output node 220 is a weighted sum s of the outputs of the nodes in the previous layer, as shown in the following equation (1), where w 0 is a constant, and w 1 , w 2 ... are the weights applied to the outputs x 1 , x 2 ... of the nodes in the previous layer.

[0043] s = w 1 x 1 + w 2 x 2 + w 0 (1)

[0044] For example, the input to hidden node 211 is the weighted sum s of the outputs of nodes 201-204, 211 i.e., s 211 = w 1 x 1 + w 2 x 2 + w 3 x3 +w 4 x 4 , where x 1 、x 2 、x 3 、x 4 are the outputs of nodes 201 - 204, which are equal to inputs 1 - 4 respectively. The output of each of the hidden nodes in hidden nodes 211 - 215 and the output of output node 220 are the transfer function f(s). For example, the transfer function f(s) can be a differentiable function, such as the Sigmoid function or the hyperbolic tangent function (i.e., tanh), as shown in the following equation (2). In equation (2), s is the output of equation (1), and e is the mathematical constant known as Euler's number.

[0045]

[0046] Figure 3 depicts a flowchart showing an example of operations that can be performed to train an artificial neural network (ANN) to determine whether a measurement of neural activity performed after stimulating a neural region of a recipient includes a neural response. Figure 3 The operations are performed by a computer system. ANN 200 is an example of an ANN that can be trained using Figure 3 the operations. Additionally, Figure 3 the operations can also be used to train other ANNs to determine whether a measurement of neural activity performed after stimulating a neural region of a recipient includes a neural response. In operation 301, the weights of the ANN are initially set to random values. In Figure 2 the example, in operation 301, the weights w 1 、w 2 …… applied to the outputs of nodes 201 - 204 in the input layer and the outputs of nodes 211 - 215 in the hidden layer of ANN 200 are initially set to random values.

[0047] Figure 3The operations can be used to train an ANN using training data. The training data includes measurements of neural activity induced in a recipient's neural region (e.g., the auditory nerve) in response to providing a stimulus to the neural region of the recipient (e.g., of a cochlear implant system), and labels or classifications indicating whether each of the measurements includes a neural response or does not include a neural response. For example, the label or classification can be an assessment made by a human expert indicating whether each of the measurements includes a neural response. As an example, the stimulus can be an electrical stimulation of the auditory nerve generated by one or more electrodes (e.g., stimulation contact 126) in a cochlear implant system in response to a signal generated by an electrophysiological response measurement system, and the measurements can be sensed by one or more electrodes in the cochlear implant system. The measurements can be provided to the electrophysiological response measurement system. For example, the measurements can be processed by the electrophysiological response measurement system (e.g., to generate a trace on a display screen). Using training data with a large number of labeled or classified measurements helps train the ANN to subsequently determine whether an unlabeled or unclassified measurement includes a neural response. Generally, if the ANN has been trained using training data with a larger number of measurements, the ANN can more accurately determine whether a measurement includes a neural response.

[0048] In operation 302, a forward pass is performed on the ANN to compute the output of each node of the ANN using a sample from the training data discussed above, ending with the output node of the ANN. The sample includes the values provided to the input nodes in the input layer of the ANN in operation 302. For example, as described above, each sample used in operation 302 can include one or more measurements of neural activity induced in a recipient's neural region (e.g., the auditory nerve) in response to a stimulus provided to the neural region of the recipient. As a more specific example, each sample used in operation 302 can include values from measurements performed by a cochlear implant system. The output of operation 302 is the value generated by the output node of the ANN, which indicates whether the sample represents a neural response or does not represent a neural response.

[0049] Then, the output of operation 302 is compared with a target value obtained from the training data to determine an error. The error is determined based on the difference between the target value and the output of operation 302 (e.g., (target value - output of operation 302) 2) Then, in operation 303, the error is used in backpropagation to adjust all the weights of the ANN. In operation 303, backpropagation of the error (i.e., backpropagation) is performed on the ANN to adjust all the weights of the ANN based on the error and the contribution of each weight to the error, so as to reduce the error. For example, operation 303 can be performed using the delta rule, which is an example of a backpropagation algorithm. The delta rule is a gradient descent learning rule for updating the weights of the inputs of nodes in the ANN. Using a differentiable function for the transfer function f(s) in the node, such as the Sigmoid function or the hyperbolic tangent function, can help reduce the error during backpropagation.

[0050] Then, in decision operation 304, it is determined whether the ANN can be further trained using another sample in the training data. If the training data includes additional samples that have not been used to train the ANN, operations 302 and 303 are repeated using this additional sample in the training data. After decision operation 304, operations 302 and 303 are repeated for each additional sample in the training data that has not been used to train the ANN until operations 302 - 303 have been performed for each sample in the training data. If it is determined at decision operation 304 that each sample in the training data has been used to train the ANN in operations 302 - 303, then Figure 3 the process ends. Then, the training of the ANN is completed. The ANN can be retrained at any time using a different set of training data with labels or classifications. Additionally, the number of input nodes and the number of hidden nodes in the ANN can be changed at any time.

[0051] Figure 4 depicts a flowchart of an example of an operation that can be performed to determine whether a measurement of neural activity performed after applying a stimulus to a neural region of a recipient includes a neural response using an artificial neural network (ANN) trained using the operations according to Figure 3 In operation 401, an electrophysiological response measurement system (e.g., Figure 1B the electrophysiological response measurement system 160 of Figure 1A-1BAs shown in [reference], the implantable component has an electrode array implanted in the cochlea of the implant recipient, and the cochlear implant system stimulates each electrode in the array in response to corresponding signals generated by and received from an electrophysiological response measurement system during multiple iterations of operations 401 - 406.

[0052] The electrophysiological response measurement system then receives one or more objective measurements (e.g., as one or more signals) from the stimulation system. In operation 404, the electrophysiological response measurement system measures or extracts values indicative of one or more objective measurements from, for example, one or more signals received from the stimulation system. For example, the values indicative of one or more objective measurements can be displayed as a signal trace on a display screen. The electrophysiological response measurement system includes an ANN trained according to the operations disclosed herein Figure 3 The electrophysiological response measurement system provides values indicative of one or more objective measurements to the input layer of the ANN. In operation 405, the ANN receives at the input layer of the ANN values indicative of one or more objective measurements. In operation 406, the output nodes of the ANN generate an output indicative of whether one or more objective measurements include a neural response to the stimulation or do not include a neural response to the stimulation. For example, one or more objective measurements can include noise that does not indicate a neural response of the auditory nerve to the stimulation.

[0053] Figure 5 FIG. [reference] depicts a flowchart of an example of operations 501 - 505 that can be performed to determine a stimulation level that elicits a neural response in a neural region of the recipient within a search range of stimulation levels. The operations 501 - 505 are performed using an artificial neural network (ANN) trained according to the operations of Figure 3 The operations of Figure 5 perform a binary search algorithm for stimulation levels within the search range of stimulation levels. For example, the operations 501 - 505 can be performed by a computer system in the electrophysiological response measurement system. The computer system implements the ANN. For example, the operations 501 - 505 can be performed for each stimulation contact (e.g., each electrode) in the cochlea of the implant recipient in the implantable component of a cochlear implant system (e.g., cochlear implant system 100 of Figure 5 disclosed herein). Figure 5 disclosed herein). Figure 1A-1B The operations 501 - 505 can be performed for each stimulation contact (e.g., each electrode) in the cochlea of the implant recipient in the implantable component of a cochlear implant system (e.g., cochlear implant system 100 of Figure 5 disclosed herein).

[0054] In operation 501, the electrophysiological response measurement system and the stimulation system generate a stimulation to a neural region (e.g., the auditory nerve) of a recipient, receive one or more measurements of neural activity induced in the neural region in response to the stimulation, and provide values indicating the one or more measurements to the input layer of the ANN. For example, operation 501 may generate one or more stimulations at one or more stimulation contacts (e.g., one or more electrodes) implanted in the cochlea of a recipient in a cochlear implant system. For example, operation 501 may generate one or more objective measurements of neural activity induced within the neural region at one or more stimulation contacts (e.g., one or more electrodes) in a cochlear implant system. Operation 501 may include operations 401-405 as disclosed herein with respect to Figure 4 The level of the initial stimulation provided in operation 501 may be selected, for example, to be at the midpoint of a search range of stimulation levels that ranges from an expected minimum threshold level (T level) to an expected maximum comfort level (C level) of one or more stimulation channels or electrodes in the stimulation system.

[0055] In operation 502, the ANN determines whether each measurement received in operation 501 includes a neural response to the stimulation. In operation 502, the ANN outputs a value indicating whether the measurement includes a neural response of the neural region to the stimulation or does not include such a neural response (e.g., merely indicates noise). Operation 502 may include operation 406 as disclosed herein with respect to Figure 4 In operation 503, if it is determined that the measurement analyzed by the ANN in operation 502 does not include a neural response, the electrophysiological response measurement system selects an increased stimulation level to be provided to the neural region of the recipient. The electrophysiological response measurement system may select an increased stimulation level in operation 503, for example, at the midpoint between the stimulation level previously provided in operation 501 and the maximum stimulation level of the search range of stimulation levels.

[0056] In operation 504, if it is determined that the measurement analyzed by the ANN in operation 502 includes a neural response, the electrophysiological response measurement system selects a decreased stimulation level to be provided to the neural region of the recipient. The electrophysiological response measurement system may select a decreased stimulation level in operation 504, for example, at the midpoint between the stimulation level previously provided in operation 501 and the minimum stimulation level of the search range of stimulation levels.

[0057] In operation 505, if the stimulation level selected in the previous iteration of operation 503 or 504 is at or within the desired stimulation level, then Figure 5For example, if it is determined that the stimulation level selected in the previous iteration of operation 503 or 504 is equal to or close to the lowest stimulation level (e.g., the threshold of the neural response) that induces the neural response of the auditory nerve, the process may be terminated in operation 505. Figure 5 process.

[0058] If it is not determined in operation 505 that the stimulation level selected in the previous iteration of operation 503 or 504 is within the desired stimulation level or within the desired stimulation level range, then Figure 5 The process returns to operation 501 for continuation. Operation 501 is then repeated. In the second and subsequent iterations of operation 501, the electrophysiological response measurement system and the stimulation system generate stimulation to the neural region of the recipient at the increased or decreased stimulation level selected in the previous iteration of operation 503 or 504. A measurement of neural activity evoked in response to the stimulation is then received, and a value indicative of the measurement is provided to the input layer of the ANN. Operations 502-505 are then repeated after each iteration of operation 501 until Figure 5 The process is terminated as described above.

[0059] For example, a stimulation contact (e.g., for each electrode) in the cochlea of ​​an implant recipient in a cochlear implant system may be performed. Figure 5 As a specific example, which is not intended to be limiting, one may perform Figure 5 Operations 501-505 are performed to detect each electrode (e.g. Figure 1A-1B After determining the neural response for one electrode, operations 501-505 may be performed for another electrode in the electrode array of the cochlear implant system. The electrodes in the electrode array may be stimulated in any desired order during iterations of operation 501. For example, the neural response determined for each electrode in the cochlear implant system in operations 501-505 may be used to generate a dynamic range of stimulation for each electrode, including a minimum threshold level (T level) and / or a maximum comfort level (C level).

[0060] Figure 6 A flow chart showing an example of operations that may be performed to determine whether measurements performed after applying stimulation to a neural region of a recipient include a neural response by providing pixels of a trajectory of the measurements to an artificial neural network (ANN). Figure 3 Operational training Figure 6 The ANN used in the operation. Initially, execute Figure 4 Operations 401-403 of the present invention are to provide stimulation to a neural region (e.g., an auditory nerve) of a recipient, generate a measurement of neural activity induced in the neural region in response to the stimulation, and provide the measurement to an electrophysiological response measurement system.

[0061] In operation 601, the electrophysiological response measurement system generates a measured trace. The electrophysiological response measurement system generates an image of a trace formed by pixels. As a specific non-limiting example, the electrophysiological response measurement system can generate an N×N image of a trace formed by N 2 pixels, where N is any positive integer greater than 0. The electrophysiological response measurement system provides the pixels from the image of the trace to the ANN. The ANN includes an input layer having input nodes (e.g., N 2 input nodes), as shown, for example, in Figure 2 . The ANN is executed by a computing system.

[0062] In operation 602, the input nodes in the input layer of the ANN receive the pixels from the image of the measured trace. Each input node in the input nodes of the ANN receives a different / distinct pixel from the pixels of the image of the trace. For example, each of the N 2 input nodes in the ANN can receive a different pixel from the N 2 pixels of the image of the trace. In operation 603, the ANN generates an output at a single output node of the ANN indicating whether the measurement includes a neural response (e.g., output 1) or does not include a neural response (e.g., output 0). The ANN can include one or more hidden layers between the input layer and the output layer, as disclosed herein, for example, with respect to Figure 2 .

[0063] Figure 7 depicts a flowchart illustrating an example of operations that can be performed to determine whether a measurement performed after applying a stimulus to a neural region of a recipient includes a neural response by providing frequency components of a measured sample to an artificial neural network (ANN). The ANN used in the operations of Figure 3 can be trained according to the operations of Figure 7 . Initially, operations 401 - 403 of Figure 4 are performed to provide a stimulus to a neural region of the recipient (e.g., the auditory nerve), generate a measurement of neural activity evoked in the neural region in response to the stimulus, and provide the measurement to the electrophysiological response measurement system.

[0064] In operation 701, a signal indicative of a measurement is sampled (e.g., using a sampler in an electrophysiological response measurement system) to generate a sampled signal. For example, the signal indicative of a measurement sampled in operation 701 can be a waveform of a trace of a finite-duration signal from one period of a signal assumed to represent a repeating periodic signal. In operation 702, a discrete Fourier transform (DFT) is performed on the sampled signal (e.g., using a fast Fourier transform) to extract the frequency components of the sampled signal. Each frequency component in the frequency components of the sampled signal represents a frequency of the sampled signal. For example, operation 702 can be performed by software in an electrophysiological response measurement system. Then, the frequency components of the sampled signal (or a subset of the frequency components of the sampled signal) are provided to the input layer of the ANN. The ANN is executed by a computing system.

[0065] In operation 703, the input nodes in the input layer of the ANN receive the frequency components of the sampled signal (or a subset of the frequency components). Each input node in the input nodes of the ANN receives a different / distinct frequency component from the frequency components of the sampled signal. Thus, N frequency components of the sampled signal are received by N input nodes of the ANN. In operation 704, the ANN generates an output at a single output node of the ANN indicative of whether the measurement includes a neural response (e.g., generating output 1) or does not include a neural response (e.g., generating output 0). The ANN can include one or more hidden layers between the input layer and the output layer, as disclosed herein, for example, with respect to Figure 2 disclosed.

[0066] Figure 8 FIG. shows a flowchart of an example of operations that can be performed to determine whether a measurement performed after applying a stimulus to a neural region of a recipient includes a neural response by providing a sample of a signal indicative of the measurement to an artificial neural network (ANN). The ANN used in the operations of Figure 3 can be trained according to the operations of Figure 8 At first, operations 401 - 403 of Figure 4 are performed to provide a stimulus to a neural region of a recipient (e.g., the auditory nerve), generate a measurement of neural activity evoked in the neural region in response to the stimulus, and provide the measurement to an electrophysiological response measurement system.

[0067] In operation 801, a signal indicative of a measurement is sampled (e.g., using a sampler in an electrophysiological response measurement system) to generate a sample of the signal. For example, the signal indicative of a measurement sampled in operation 801 can be a waveform of a trace of a finite-duration signal from one period of a signal assumed to represent a repeating periodic signal. Then, the sample of the signal (or a subset of the samples) is provided to the input layer of the ANN. The ANN is executed by a computing system.

[0068] In operation 802, input nodes in the input layer of the ANN receive samples (or subsets of samples) of the signal. Each of the input nodes in the input layer of the ANN receives a different / unique sample of the samples of the signal. Thus, N samples of the signal are received by the N input nodes of the ANN. In operation 803, the ANN generates, at a single output node of the ANN, an output indicating whether the measurement includes a neural response (e.g., generating an output of 1) or does not include a neural response (e.g., generating an output of 0). The ANN may include one or more hidden layers between the input layer and the output layer, as disclosed herein, for example, with respect to Figure 2 disclosed.

[0069] Figure 9 FIG. depicts an example of a suitable computing system 900 that can perform any of the operations or functions disclosed herein. For example, computing system 900 can be used to implement any ANN disclosed herein. As disclosed herein, computing system 900 can use an ANN to generate an indication of whether a measurement of neural activity evoked in a neural region in response to a stimulus includes a neural response or does not include a neural response. For example, computing system 900 can be part of an electrophysiological response measurement system. Computing systems, environments, or configurations suitable for use with the examples disclosed herein include, but are not limited to, personal computers, server computers, handheld devices, laptop devices, multiprocessor systems, microprocessor-based systems, programmable consumer electronics (e.g., smartphones), network computers, minicomputers, mainframe computers, tablet computers, distributed computing environments including any of the above systems or devices, and the like. Computing system 900 can be a single virtual or physical device operating in a networked environment via a communication link to one or more remote devices. The remote device can be an auditory prosthesis (e.g., Figure 1A-1B the auditory prosthesis), a personal computer, a server, a router, a network personal computer, a peer device, or other common network nodes.

[0070] Computing system 900 includes at least one processing unit 902 and a memory 904. The processing unit 902 includes one or more hardware or software processors (e.g., a central processing unit) that can obtain and execute instructions. The processing unit 902 can communicate with and control the execution of other components of computing system 900. The memory 904 is one or more software-based or hardware-based computer-readable storage media operable to store information accessible by the processing unit 902.

[0071] In addition to storing other data, the memory 904 may also store instructions that can be executed by the processing unit 902 to implement an application program (software) or to perform any of the functions or operations disclosed herein. The memory 904 can be a volatile memory (e.g., random access memory or RAM), a non-volatile memory (e.g., read-only memory or ROM), or a combination thereof. The memory 904 may also include one or more removable or non-removable storage devices. The memory 904 can include transient memory and / or non-transitory computer-readable storage media. Compared with a medium that only transmits propagating electrical signals such as a wire, a non-transitory computer-readable storage medium is a tangible computer-readable storage medium that stores data for later access. In an example, the memory 904 may include non-transitory computer-readable storage media such as RAM, ROM, EEPROM (electrically erasable programmable read-only memory), flash memory, optical disc storage devices, magnetic storage devices, solid-state storage devices, or any other memory medium that can be used to store information for later access. In an example, the memory 904 encompasses a modulated data signal (e.g., a signal whose one or more characteristics are set or changed in a manner that encodes information in the signal), such as a carrier wave or other transmission mechanism, and includes any information delivery medium. By way of example and not limitation, the memory 904 may include a wired medium (e.g., a wired network or a direct wired connection), as well as a wireless medium (e.g., acoustic, radio frequency, infrared, and other wireless media), or a combination thereof.

[0072] In the illustrated example, the computing system 900 further includes a network adapter 906, one or more input devices 908, and one or more output devices 910. The system 900 may include other components such as a system bus, component interfaces, a graphics system, a power source (e.g., a battery), and other components.

[0073] The network adapter 906 is a component of the computing system 900 that provides network access to the network 912. The network adapter 906 can provide wired or wireless network access and can support one or more of a variety of communication technologies and protocols, such as Ethernet, cellular, Bluetooth, near field communication, and RF (radio frequency), and so on. The network adapter 906 may include one or more antennas and associated components configured for wireless communication according to one or more wireless communication technologies and protocols.

[0074] One or more input devices 908 are devices through which the computing system 900 receives input from a user. One or more input devices 908 may include physically actuatable user interface elements (e.g., buttons, switches, or dials), touchscreens, keyboards, mice, pens, and voice input devices, as well as other input devices.

[0075] One or more output devices 910 are devices through which computing system 900 can provide output to a user. Output devices 910 can include a display, a speaker, and a printer, among other output devices.

[0076] Unless otherwise expressly indicated, any embodiment or any feature disclosed herein can be combined with any one or more other embodiments and / or other features disclosed herein. Any embodiment or any feature disclosed herein can be expressly excluded from use with any one or more other embodiments and / or other features disclosed herein, unless otherwise expressly indicated. Note that any method detailed herein also corresponds to a disclosure of a device and / or system configured to perform one or more or all of the method acts associated with the device and / or system as detailed herein. Note also that any disclosure of a device and / or system detailed herein corresponds to a method of making and / or using the device and / or system, including a method of using the device in accordance with the functions detailed herein.

[0077] The foregoing description of the exemplary embodiments of the present invention has been presented for purposes of illustration. The foregoing description is not intended to be exhaustive or to limit the invention to the examples disclosed herein. In some instances, features of the invention can be used without the corresponding use of other features described. Many modifications, substitutions, and variations are possible in light of the above teachings without departing from the scope of the invention.

Claims

1. A computing system, comprising: at least one processing unit implementing an artificial neural network, wherein the artificial neural network generates, at a single output node, an output indicating whether a measurement performed after stimulating a neural region of an individual includes a neural response.

2. The computing system according to claim 1, wherein input nodes of the artificial neural network receive pixels of an image of a trajectory of the measurement.

3. The computing system according to claim 1, wherein input nodes of the artificial neural network receive frequency components of a sampled signal generated by sampling a signal indicating the measurement.

4. The computing system according to claim 1, wherein input nodes of the artificial neural network receive samples of a signal indicating the measurement.

5. The computing system according to any one of claims 1-4, wherein the computing system trains the artificial neural network using training data comprising measurements performed after stimulating the neural region and labels or classifications indicating whether the measurements include neural responses.

6. The computing system according to any one of claims 1-5, wherein an electrophysiological response measurement system generates a trajectory indicating the measurement and provides values indicating the trajectory to an input layer of the artificial neural network.

7. The computing system according to any one of claims 1-6, wherein the electrophysiological response measurement system performs a search algorithm by generating stimuli at different levels to be applied to the neural region and analyzes measurements from the individual to determine a stimulus level at which the neural response is generated from the neural region.

8. The computing system according to any one of claims 1-7, wherein a stimulation system uses stimulation contacts to provide stimuli to the neural region, wherein a set of values indicating an objective measurement after each of the stimuli is provided to input nodes of the artificial neural network, and wherein the single output node generates an additional output indicating whether each of the objective measurements includes a neural response.

9. The computing system according to any one of claims 1-8, wherein the single output node is the only output node of the artificial neural network that generates the output indicating whether the measurement includes a neural response.

10. A method, comprising: receiving, at an artificial neural network in a computing system, values indicating a measurement performed after providing a stimulus to a neural region of an individual; and generating, at a single output node of the artificial neural network, an output indicating whether the measurement includes a neural response.

11. The method according to claim 10, further comprising: using an electrophysiological response measurement system to generate the values indicating the measurement.

12. The method according to any one of claims 10-11, wherein receiving the values indicating the measurement further comprising: receiving, at input nodes of the artificial neural network, pixels of an image of a trajectory of the measurement.

13. The method according to any one of claims 10-11, wherein receiving the values indicating the measurement further comprising: Receive, at an input node of the artificial neural network, frequency components of a sampled signal generated by sampling a signal indicative of the measurement.

14. The method according to any one of claims 10 - 11, wherein receiving the value indicative of the measurement further comprises: Receive, at an input node of the artificial neural network, samples of a signal indicative of the measurement.

15. The method according to any one of claims 10 - 14, further comprises: Provide the stimulation to a first stimulation contact in a stimulation system; Receive the measurement at a second stimulation contact in the stimulation system; and Provide the measurement to the computing system.

16. The method according to any one of claims 10 - 15, further comprises: If the output indicates that the measurement does not include a neural response, increase the stimulation level provided to the neural region; and If the output indicates that the measurement includes a neural response, decrease the stimulation level provided to the neural region.

17. The method according to any one of claims 10 - 16, wherein generating the output indicative of whether the measurement includes a neural response further comprises: Generate, at a single output node of the artificial neural network, the output indicative of whether the measurement includes a neural response.

18. A non - transitory computer - readable storage medium, the non - transitory computer - readable storage medium comprising computer - readable instructions stored thereon for causing a computing system to: Receive, at an input node of an artificial neural network in the computing system, pixels of an image of a trace from a measurement performed after stimulating an individual's auditory nerve; and Generate, using the artificial neural network, an indication of whether the measurement includes a neural response based on the pixels.

19. The non - transitory computer - readable storage medium according to claim 18, wherein the computer - readable instructions further cause the computing system to: Receive different pixels among the pixels at each of the input nodes of the artificial neural network.

20. The non - transitory computer - readable storage medium according to any one of claims 18 - 19, wherein the computer - readable instructions further cause the computing system to: Generate, at a single output node of the artificial neural network, the indication of whether the measurement includes a neural response or does not include a neural response.

21. The non - transitory computer - readable storage medium according to any one of claims 18 - 20, wherein the computer - readable instructions further cause the computing system to: Provide a varying stimulation level to the individual's auditory nerve to determine, using an electrophysiological response measurement system, the stimulation level that generates the neural response from the auditory nerve.

22. A method, comprises: Sample a signal indicative of a measurement performed after stimulating an individual's auditory nerve to generate a sampled signal; Perform a Fourier transform on the sampled signal to extract frequency components of the sampled signal; Receive, at an input node of an artificial neural network in a computing system, the frequency components of the sampled signal; and Generate an output using the artificial neural network indicating whether the measurement includes a neural response.

23. The method according to claim 22, wherein receiving the frequency components of the sampled signal at the input nodes of the artificial neural network further comprises: Receiving different frequency components of the frequency components at each of the input nodes of the artificial neural network.

24. The method according to any one of claims 22-23, wherein generating the output indicating whether the measurement includes a neural response further comprises: Generating the output indicating whether the measurement includes a neural response or does not include a neural response at a single output node of the artificial neural network.

25. The method according to any one of claims 22-24, further comprises: Generating the stimulation of the auditory nerve using a first stimulation contact; Generating the measurement using a second stimulation contact; Increasing the level of additional stimulation to be provided to the auditory nerve if the output indicates that the measurement does not include a neural response; Decreasing the level of the additional stimulation to be provided to the auditory nerve if the output indicates that the measurement includes a neural response; and Generating the additional stimulation of the auditory nerve using the first stimulation contact.

Images