CN112653980B - Interactive self-checking and matching method for intelligent hearing aid - Google Patents
Interactive self-checking and matching method for intelligent hearing aid Download PDFInfo
- Publication number
- CN112653980B CN112653980B CN202110035197.2A CN202110035197A CN112653980B CN 112653980 B CN112653980 B CN 112653980B CN 202110035197 A CN202110035197 A CN 202110035197A CN 112653980 B CN112653980 B CN 112653980B
- Authority
- CN
- China
- Prior art keywords
- voice
- parameter adjustment
- patient
- hearing
- evaluation
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04R—LOUDSPEAKERS, MICROPHONES, GRAMOPHONE PICK-UPS OR LIKE ACOUSTIC ELECTROMECHANICAL TRANSDUCERS; DEAF-AID SETS; PUBLIC ADDRESS SYSTEMS
- H04R25/00—Deaf-aid sets, i.e. electro-acoustic or electro-mechanical hearing aids; Electric tinnitus maskers providing an auditory perception
- H04R25/30—Monitoring or testing of hearing aids, e.g. functioning, settings, battery power
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04R—LOUDSPEAKERS, MICROPHONES, GRAMOPHONE PICK-UPS OR LIKE ACOUSTIC ELECTROMECHANICAL TRANSDUCERS; DEAF-AID SETS; PUBLIC ADDRESS SYSTEMS
- H04R25/00—Deaf-aid sets, i.e. electro-acoustic or electro-mechanical hearing aids; Electric tinnitus maskers providing an auditory perception
- H04R25/50—Customised settings for obtaining desired overall acoustical characteristics
- H04R25/505—Customised settings for obtaining desired overall acoustical characteristics using digital signal processing
- H04R25/507—Customised settings for obtaining desired overall acoustical characteristics using digital signal processing implemented by neural network or fuzzy logic
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04R—LOUDSPEAKERS, MICROPHONES, GRAMOPHONE PICK-UPS OR LIKE ACOUSTIC ELECTROMECHANICAL TRANSDUCERS; DEAF-AID SETS; PUBLIC ADDRESS SYSTEMS
- H04R2225/00—Details of deaf aids covered by H04R25/00, not provided for in any of its subgroups
- H04R2225/77—Design aspects, e.g. CAD, of hearing aid tips, moulds or housings
Abstract
The invention discloses an interactive self-fitting method for an intelligent hearing aid, which comprises the following steps: step 1, a hearing-impaired patient listens test voice sent by a hearing aid to perform five-level evaluation, and a hearing aid sound quality evaluation index is formed; step 2, forming a feature sequence according to 11-dimensional audiogram and 3-dimensional personal information of a patient through voice evaluation, then calculating the similarity of the feature sequence and the feature sequence in the fitting historical database, and designing a value function for parameter adjustment and calculating a parameter adjustment strategy; a slight value, so that the value of each parameter adjustment is maximized; step 4, constructing a neural network to obtain the weight of the neural network; and 5, inputting the test voice frequency spectrum, voice evaluation, the individual information of the patient and 3 continuous groups of historical parameter adjustment strategies into a neural network, and regenerating the test voice to evaluate the hearing-impaired patient. The invention comprehensively considers the subjective and objective voice evaluation method, combines the deep learning network, integrates the cognition and personal factors into the hearing aid design, and has good application prospect.
Description
Technical Field
The invention relates to an interactive self-fitting technology of an intelligent hearing aid, and belongs to the technical field of hearing aid fitting.
Background
Due to the deterioration of cognitive abilities of hearing-impaired patients, the traditional method has low efficiency, and the design concept of fitting by the patients gradually becomes a hotspot of research. Currently, the simplest self-fitting method is to let the user select from a few pre-programmed algorithm configurations, which depends mainly on the mode optimization strategy, i.e. the degree of matching preset with the user population.
Relevant studies show that more sophisticated interactive self-fitting methods can better reflect the personalized differences of patients. With the development of smart phones and headsets, the concept of self-fitting hearing aids has received increasing attention. The effectiveness of early hearing aids configured by presets depends on how well the presets match the gain requirements of the user population. Furthermore, the method by which the user can directly control the broadband, low frequency and high frequency gains, although faster, is currently focused only on linear (non-compressive) gains and is not evaluated in real experiments. In recent years, it has become a trend to replace the role of hearing professionals with artificial intelligence algorithms, such as genetic algorithms to optimize hearing aid algorithm parameters. However, the convergence rate of the genetic algorithm is slow, the stability is poor, and the practicability of the algorithm is influenced. Relatively speaking, the interactive comparison self-fitting algorithm based on the Gaussian process and the active learning strategy is more efficient, the hearing aid user is allowed to interact with each other through a human-computer interaction in daily life, and the intelligent algorithm is used for replacing a hearing specialist to adjust the hearing aid by self, so that the cognitive factors are introduced in a phase-changing manner. However, this research work does not take full advantage of historical experience and lacks an effective update mechanism. In 2017, the U.S. has first introduced hearing aids into over-the-counter products through the over-the-counter hearing aid law, and adults with mild to moderate hearing loss do not need to be tested and instructed by professional hearing persons to select and prepare the hearing aids. This marks the critical phase of self-fitting hearing aid research, and in order to comply with the development of the hearing aid industry, china must develop a fitting-free hearing aid core technology with proprietary intellectual property rights.
Through the description, how to establish the interactive self-fitting method for the intelligent hearing aid is the key for improving the fitting performance and efficiency of the hearing aid, and has important research significance for improving the speech comprehension and algorithm universality of hearing-impaired patients.
Disclosure of Invention
The purpose of the invention is as follows:
the invention aims to overcome the problem that the traditional hearing aid fitting method depends heavily on audiologists, and aims to establish an intelligent hearing aid interactive self-fitting method, fully utilize comprehensive information of patients and improve fitting efficiency. The interactive fitting method not only considers the comprehensive information of the patient and the historical fitting information, but also realizes the parameter adjustment optimization by simulating the audiologist based on the intelligent algorithm design model. The method is ingenious and novel, and has good application prospect.
The technical scheme is as follows:
in order to achieve the purpose, the invention adopts the technical scheme that:
an interactive self-fitting method for a smart hearing aid, comprising the steps of:
step 1, hearing-impaired patients listen to test voice emitted by hearing aids, five-level evaluation is carried out on the four aspects of the sound intensity, the clarity degree, the sound depth degree and the sound sharp degree of the test voice, and the evaluation index of the sound quality of the hearing aids is formed, wherein the calculation method comprises the following steps of
Wherein r represents the speech assessment, piThe evaluation values represent all aspects, i represents the serial numbers of the four evaluations, and N represents the number of evaluation indexes;
step 2, evaluating p according to 11-dimensional audiogram, 3-dimensional personal information and voice of the patientiForming a characteristic sequence x, wherein 3-dimensional personal information comprises age, gender and hearing aid wearing time, and then calculating the similarity s of the characteristic sequence x and the characteristic sequence y in the fitting historical databasek，
Wherein k represents the number of characteristic sequences in the database, M represents the number of index sequences, and xjRepresenting the jth feature, y, in the feature sequence of the current patientjRepresenting the ith feature in the feature sequence y of the patient in the database; j represents the serial number of 17 indexes, and skSorting from high to low, selecting 1 group of characteristic sequences with the highest numerical value, and obtaining 3 continuous groups of historical parameter adjustment strategies corresponding to the group of characteristics in a database;
step 3, designing a value function of parameter adjustment, and calculating the value of a parameter adjustment strategy to maximize the value of each parameter adjustment;
step 4, constructing a neural network, designing a loss function, and training in a mode of minimizing the loss function so as to obtain a weight of the neural network;
step 5, testing the voice frequency spectrum, and evaluating the voice piAnd inputting the individual information of the patient and 3 groups of continuous historical parameter adjustment strategies into a neural network, and regenerating test voice to evaluate the hearing-impaired patient until the patient is satisfied.
Further, the cost function of the parameter adjustment in step 3 is
Q*(h,d)＝E[r+ξQ*(h',d')|h,d] (3)
Wherein h represents the frequency spectrum of the voice signal, d represents the parameter adjustment, and r represents the voice evaluation; h 'and d' represent state information and parameter adjustment values under the current optimal strategy; ξ is the adjustment weight and E represents the mathematical expectation.
Further, step 4 specifically includes: constructing a neural network with weight value theta, and designing a loss function as
Li(θi)＝E[(yi-Q(h,d；θi))2] (4)
Wherein the content of the first and second substances,
is the objective of the ith iteration when optimizing the loss function Li(θi) Then, the experience of fitting the model is stored in each time step, and the experience e of each time step is storedt＝(ht,dt,rt,ht+1) Stored in the data set D ═ e1,…,eNIn (e)t,ht,rtRespectively representing the fitting experience representation, signal spectrum, parameter adjustment and voice evaluation at the time t; h ist+1Representing the signal spectrum at the next moment, randomly extracting a stored fitting experience e from a stored sample pool each time in an internal loop of the algorithm, selecting a greedy strategy d according to a set probability beta by a fitting model,
has the advantages that:
1) the evaluation of the hearing-impaired patient is subjected to targeted grading, and comprehensive information and matching historical information of the patient are fully utilized, so that a preliminary strategy for parameter adjustment is quickly acquired according to the evaluation of the patient;
2) and a value function for parameter adjustment is constructed, and voice characteristics, patient evaluation and personal information are fused, so that the parameter adjustment is more reasonable.
3) And designing a neural network training cost function to aim at the expectation maximization of parameter adjustment, thereby replacing the selection of an optimized parameter adjustment strategy by an audiologist.
Drawings
Fig. 1 is a model block diagram of an interactive self-fitting method for a smart hearing aid according to the present invention;
FIG. 2 is a schematic illustration of a speech recognition rate comparison of an embodiment of the present invention.
Detailed Description
The invention will be further described with reference to the accompanying drawings. The invention discloses an interactive self-fitting method of an intelligent hearing aid, aiming at changing the fitting mode of the traditional hearing aid based on hearing experts, and leading a hearing-impaired patient to listen to test voice at first and feed back the voice according to an evaluation list; then, the analysis module acquires a parameter updating strategy related to the patient from a knowledge base according to the user feedback and the individual information; the model calculates the characteristic parameters of the current evaluation voice, the personalized information of the patient, the evaluation information of the user and the historical parameter updating strategy, and the information is used as the input of the designed neural network, so that the voice to be evaluated is regenerated and is provided for the hearing-impaired patient to select iteratively. The parameter tuning strategy depends on the model tuning method and needs to be trained based on user interaction, patient personal information and historical fitting data. The invention comprehensively considers the subjective and objective voice evaluation method, combines the deep learning network, integrates the cognition and personal factors into the hearing aid design, and has good application prospect
As shown in fig. 1, the interactive self-fitting method for a smart hearing aid according to the present invention comprises the following steps:
step 1, hearing-impaired patients listen to test voice and own voice emitted by hearing aids, five-level evaluation is carried out on the four aspects of the sound intensity, the clarity degree, the sound depth degree and the sound sharp degree of the test voice and the own voice, and a hearing aid sound quality evaluation index is formed
Wherein r represents the speech assessment, piThe evaluation values representing the respective aspects, i represents the numbers of the four evaluations, and N represents the number of evaluation indexes.
Step 2, according to the 11-dimensional audiogram of the patient, namely the hearing threshold of the patient at 11 frequency points (unit Hz), (11 frequency points are respectively 125Hz, 250Hz, 500Hz, 750Hz, 1KHz, 1.5KHz, 2KHz, 3KHz, 4KHz, 6KHz and 8KHz), 3-dimensional personalized information (including age, gender and hearing aid wearing time), and p is evaluated by voiceiForming a characteristic sequence x, and then calculating the similarity s of the characteristic sequence x and the characteristic sequence y in the matching historical databasek，
Wherein k represents the number of characteristic sequences in the database, M represents the number of index sequences, and xjRepresenting the jth feature, y, in the feature sequence of the current patientjRepresenting the ith feature in the feature sequence y of the patient in the database; j represents the serial number of 17 indexes, 17 indexes are 11-dimensional audiogram, 3-dimensional personal information and 4 evaluations, and s is obtainedkAnd sorting from high to low, selecting 1 group of feature sequences with the highest numerical value, and obtaining 3 continuous groups of historical parameter adjustment strategies corresponding to the group of features in the database.
And 3, designing a value function of parameter adjustment, and calculating the value of a parameter adjustment strategy to maximize the value of each parameter adjustment. The value function of the parameter adjustment is
Q*(h,d)＝E[r+ξQ*(h',d')|h,d] (11)
Wherein h represents the frequency spectrum of the voice signal, d represents the parameter adjustment, and r represents the voice evaluation; h 'and d' represent state information and parameter adjustment values under the current optimal strategy; ξ is the adjustment weight and E represents the mathematical expectation.
Step 4, constructing a neural network with weight theta, wherein the neural network consists of 4 layers of fully-connected layers, each layer comprises 512 neurons, an activation function is a ReLU function, the input of the layer 1 comprises 183 dimensions, and the input comprises 11-dimensional audiogram, 3-dimensional personal information, 5-dimensional hearing evaluation and 65-dimensional voice frequency spectrum (128-point FFT); the 3 groups of gain adjustment parameters have 99 dimensions, and comprise parameters of 50dB, 70dB and 85dB SPL of three sound pressure levels, the gain adjustment parameter under each sound pressure level is 11 dimensions, and the output of the output layer is 33 dimensions of the gain adjustment parameters under the three sound pressure levels.
And designing a loss function, and training in a mode of minimizing the loss function so as to obtain the weight of the neural network.
Designed loss function of
Li(θi)＝E[(yi-Q(h,d；θi))2] (12)
Wherein the content of the first and second substances,
is the objective of the ith iteration when optimizing the loss function Li(θi) Then, the experience of fitting the model is stored in each time step, and the experience e of each time step is storedt＝(ht,dt,rt,ht+1) Stored in the data set D ═ e1,…,eNIn (e)t,ht,rtRespectively representing the fitting experience representation, signal spectrum, parameter adjustment and voice evaluation at the time t; h ist+1Representing the signal spectrum at the next moment, randomly extracting a stored fitting experience e from a stored sample pool each time in an internal loop of the algorithm, selecting a greedy strategy d according to a probability beta by a fitting model,
selecting and executing a parameterNumber adjustment, i.e., probability of 1-beta selects a greedy strategy, and probability of beta selects a random strategy.
Step 5, testing the voice frequency spectrum, and evaluating the voice piInputting the individual information of the patient and 3 groups of continuous historical parameter adjusting strategies into a neural network, and regenerating test voice to evaluate the hearing-impaired patient until the patient is satisfied.
Fig. 2 is a comparison of the fitting effect of six patients, and the comparison algorithm includes the conventional fitting method and the method proposed by the present invention. As can be seen from the figure, in the aspect of voice test, the fitting effect of the proposed method is good, the average recognition rate reaches 75.4%, and is increased by 11.7% compared with the traditional algorithm. Of these, the improvement in patient S1 was most pronounced.
The foregoing is only a preferred embodiment of the present invention, and it should be noted that, for those skilled in the art, various modifications and decorations can be made without departing from the principle of the present invention, and these modifications and decorations should also be regarded as the protection scope of the present invention.
Claims (1)
1. An interactive self-fitting method for a smart hearing aid, comprising the steps of:
step 1, hearing-impaired patients listen to test voice emitted by hearing aids, five-level evaluation is carried out on the four aspects of the sound intensity, the clarity degree, the sound depth degree and the sound sharp degree of the test voice, and the evaluation index of the sound quality of the hearing aids is formed, wherein the calculation method comprises the following steps of
Wherein r represents the speech assessment, piThe evaluation values represent all aspects, i represents the serial numbers of the four evaluations, and N represents the number of evaluation indexes;
step 2, evaluating p according to 11-dimensional audiogram, 3-dimensional personal information and voice of the patientiForming a characteristic sequence x, wherein the 3-dimensional personal information comprises age, gender and hearing aid wearing time, and then calculating the characteristic sequence in the fitting history databaseSimilarity s of column yk，
Wherein k represents the number of characteristic sequences in the database, M represents the number of index sequences, and xjRepresenting the jth feature, y, in the feature sequence of the current patientjRepresenting the ith feature in the feature sequence y of the patient in the database; j represents the serial number of 17 indexes, and skSorting from high to low, selecting 1 group of characteristic sequences with the highest numerical value, and obtaining 3 continuous groups of historical parameter adjustment strategies corresponding to the group of characteristics in a database;
step 3, designing a value function of parameter adjustment, and calculating the value of a parameter adjustment strategy to maximize the value of each parameter adjustment;
step 4, constructing a neural network, designing a loss function, and training in a mode of minimizing the loss function so as to obtain a weight of the neural network;
step 5, testing the voice frequency spectrum, and evaluating the voice piInputting the individual information of the patient and the continuous 3 groups of historical parameter adjustment strategies into a neural network, and regenerating test voice to evaluate the hearing-impaired patient until the patient is satisfied;
wherein the cost function is
Q*(h,d)＝E[r+ξQ*(h',d')|h,d] (3)
Wherein h represents the frequency spectrum of the voice signal, d represents the parameter adjustment, and r represents the voice evaluation; h 'and d' represent state information and parameter adjustment values under the current optimal strategy; xi is the adjustment weight, E represents the mathematical expectation;
the loss function is
Li(θi)＝E[(yi-Q(h,d；θi))2] (4)
Wherein the content of the first and second substances,
is the target of the ith iteration.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110035197.2A CN112653980B (en) | 2021-01-12 | 2021-01-12 | Interactive self-checking and matching method for intelligent hearing aid |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110035197.2A CN112653980B (en) | 2021-01-12 | 2021-01-12 | Interactive self-checking and matching method for intelligent hearing aid |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112653980A CN112653980A (en) | 2021-04-13 |
| CN112653980B true CN112653980B (en) | 2022-02-18 |
Family
ID=75368036
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110035197.2A Active CN112653980B (en) | 2021-01-12 | 2021-01-12 | Interactive self-checking and matching method for intelligent hearing aid |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112653980B (en) |
Families Citing this family (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113411733B (en) * | 2021-06-18 | 2023-04-07 | 南京工程学院 | Parameter self-adjusting method for non-fitting hearing aid |
| CN113825080A (en) * | 2021-09-11 | 2021-12-21 | 武汉左点科技有限公司 | Self-checking and matching method and device for hearing aid |
| CN114339564B (en) * | 2021-12-23 | 2023-06-16 | 清华大学深圳国际研究生院 | Neural network-based self-adaptation method for self-adaptive hearing aid of user |
| CN114938487B (en) * | 2022-05-13 | 2023-05-30 | 东南大学 | Hearing aid self-checking method based on sound field scene discrimination |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2018125663A (en) * | 2017-01-31 | 2018-08-09 | リオン株式会社 | Hearing aid fitting data provision method, hearing aid fitting data provision system, and hearing aid fitting data provision program |
| CN109151692A (en) * | 2018-07-13 | 2019-01-04 | 南京工程学院 | Hearing aid based on deep learning network tests method of completing the square certainly |
| CN111417062A (en) * | 2020-04-27 | 2020-07-14 | 陈一波 | Prescription for testing and matching hearing aid |
| CN111818436A (en) * | 2020-07-14 | 2020-10-23 | 无锡清耳话声科技有限公司 | Real ear analysis test system based on machine learning |
Family Cites Families (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US8135138B2 (en) * | 2007-08-29 | 2012-03-13 | University Of California, Berkeley | Hearing aid fitting procedure and processing based on subjective space representation |
| CN109147808B (en) * | 2018-07-13 | 2022-10-21 | 南京工程学院 | Speech enhancement hearing aid method |
| EP3641344B1 (en) * | 2018-10-16 | 2023-12-06 | Sivantos Pte. Ltd. | A method for operating a hearing instrument and a hearing system comprising a hearing instrument |
-
2021
- 2021-01-12 CN CN202110035197.2A patent/CN112653980B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2018125663A (en) * | 2017-01-31 | 2018-08-09 | リオン株式会社 | Hearing aid fitting data provision method, hearing aid fitting data provision system, and hearing aid fitting data provision program |
| CN109151692A (en) * | 2018-07-13 | 2019-01-04 | 南京工程学院 | Hearing aid based on deep learning network tests method of completing the square certainly |
| CN111417062A (en) * | 2020-04-27 | 2020-07-14 | 陈一波 | Prescription for testing and matching hearing aid |
| CN111818436A (en) * | 2020-07-14 | 2020-10-23 | 无锡清耳话声科技有限公司 | Real ear analysis test system based on machine learning |
Non-Patent Citations (1)
| Title |
|---|
| 《助听器双重反馈抑制算法中的参数效应研究》;凃磊等;《中国听力语言康复科学杂志》;20200517;第18卷(第5期);第383-386页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112653980A (en) | 2021-04-13 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112653980B (en) | Interactive self-checking and matching method for intelligent hearing aid | |
| EP1608431A1 (en) | Cochlear implant system with map optimization using a genetic algorithm | |
| US9049529B2 (en) | Hearing aids and methods and apparatus for audio fitting thereof | |
| CN109151692B (en) | Hearing aid self-checking and matching method based on deep learning network | |
| CN111064327B (en) | Driving motor optimization method and system based on sound quality | |
| Takagi et al. | Interactive evolutionary computation-based hearing aid fitting | |
| Hu et al. | Environment-specific noise suppression for improved speech intelligibility by cochlear implant users | |
| US9666181B2 (en) | Systems and methods for tuning automatic speech recognition systems | |
| CN107103901B (en) | Artificial cochlea sound scene recognition system and method | |
| CN106303874B (en) | A kind of adaptive confirmed method of completing the square of digital deaf-aid | |
| US8559662B2 (en) | Genetic algorithms with subjective input for hearing assistance devices | |
| EP2811904B1 (en) | Evaluation of sound quality and speech intelligibility from neurograms | |
| CN116390013A (en) | Hearing aid self-verification method based on neural network deep learning | |
| CN108711419A (en) | A kind of the ambient sound cognitive method and system of artificial cochlea | |
| CN114339564B (en) | Neural network-based self-adaptation method for self-adaptive hearing aid of user | |
| CN114938487B (en) | Hearing aid self-checking method based on sound field scene discrimination | |
| CN116564477B (en) | Cognitive dysfunction rehabilitation training system | |
| CN112562698A (en) | Power equipment defect diagnosis method based on fusion of sound source information and thermal imaging characteristics | |
| CN108922616A (en) | A kind of hearing aid is quickly from testing method of completing the square | |
| JP3865600B2 (en) | Adaptive characteristic hearing aid and optimum hearing aid processing characteristic determination device | |
| CN116467672A (en) | Depression recognition system based on brain electricity-voice bimodal decision fusion | |
| Harczos et al. | Comparative evaluation of cochlear implant coding strate-gies via a model of the human auditory speech processing | |
| CN106941653A (en) | Artificial cochlea's intelligent predicting system and method | |
| CN109657734A (en) | It is a kind of can dynamic change machine learning model construction method | |
| Chen et al. | A novel model-based hearing compensation design using a gradient-free optimization method |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |