AI drawing software page platformization method and system based on hearing data

Abstract

The invention provides an AI drawing software page platformization method and system based on hearing data, and relates to the technical field of hearing assistance. The method comprises the following steps: determining the similarity between the hearing disorder multi-dimensional features of a target hearing patient and the hearing disorder multi-dimensional features of each historical hearing patient; and searching historical recovered hearing data of the historical hearing patients with the similarity higher than a preset similarity threshold after hearing recovery treatment. Inputting the hearing disorder multi-dimensional features of the historical hearing patients with the similarity higher than a preset similarity threshold value and the historical recovery hearing data into a treatment strategy scoring model, determining the quality score of the corresponding hearing treatment strategy, and determining the quality score of the hearing treatment strategy on an AI drawing software page platform. And drawing and displaying a hearing recovery recommendation page including the hearing treatment strategy of the historical hearing patient with the highest quality score and the historical hearing recovery data, so that the target hearing patient knows the most suitable medical strategy and the hearing recovery data which can be reached after treatment.

Description

Technical Field

[0001] The present application relates to the field of hearing assistance technology, and in particular to a method and system for platformizing AI drawing software pages based on hearing data. Background Art

[0002] According to statistics from the World Health Organization, there are 466 million people with disabling hearing loss worldwide, accounting for 5% of the world's total population. Therefore, improving the hearing of hearing loss patients is one of the main medical issues currently faced.

[0003] Hearing loss may be caused by age (such as children and the elderly), or by disease or social environmental noise. Hearing loss mainly includes the following categories: conductive hearing loss, mixed hearing loss, sensorineural hearing loss, etc. Hearing loss can be improved by implanting a cochlear implant or by treatment (such as physical therapy and medication).

[0004] At present, the treatment strategy for improving hearing loss through treatment usually requires hearing patients to go to the doctor on site to confirm, which is subjective, unreliable, wastes medical resources, and also wastes the time cost of hearing patients. Summary of the invention

[0005] The present application provides an AI drawing software page platform method and system based on hearing data, which is used to solve the problem that the treatment strategy in the prior art for improving hearing loss through treatment usually requires hearing patients to go to the doctor on site to confirm, which has a certain degree of subjectivity, low reliability, waste of medical resources, and also wastes the time cost of hearing patients.

[0006] In a first aspect, the present application provides a method for platformizing an AI drawing software page based on hearing data, comprising: On the AI ​​drawing software page platform, the hearing parameter input page is displayed; Receive the target hearing patient's disease information, patient attribute characteristics, and initial hearing data input by the user on the hearing parameter input page, wherein the disease information includes the disease name, disease severity, and symptom description information, the patient attribute characteristics at least include the patient's age, height, weight, and gender, and the initial hearing data includes the lowest sound loudness that the hearing patient can hear for sounds of different frequencies; The hearing patient's symptom information, patient attribute characteristics and initial hearing data are integrated to obtain the hearing impairment portrait data of the target hearing patient, and the hearing impairment portrait data is converted into multi-dimensional features of hearing impairment; Determine the similarity between the multidimensional characteristics of hearing impairment of the target hearing patient and the multidimensional characteristics of hearing impairment of each historical hearing patient in a preset database; Searching for historical restored hearing data of historical hearing patients after hearing restoration treatment whose similarity is higher than a preset similarity threshold, wherein the historical restored hearing data is the lowest sound loudness that the historical hearing patients can hear for sounds of different frequencies at different treatment time points within a preset period after starting treatment; Inputting the multidimensional features of hearing impairment and historical restored hearing data of each historical hearing patient whose similarity is higher than a preset similarity threshold into a pre-trained treatment strategy scoring model to determine the quality score of the corresponding hearing treatment strategy, wherein the treatment strategy scoring model is obtained by inputting a plurality of first training samples into a first neural network for training, and each first training sample includes the historical restored hearing data of the historical hearing patient, the multidimensional features of historical hearing impairment, and the quality score of the corresponding historical hearing treatment strategy; Find the hearing treatment strategies and historical restored hearing data of the historical hearing patients with the highest quality scores; On the AI ​​drawing software page platform, a hearing recovery recommendation page is drawn and displayed, including the hearing treatment strategies and historical hearing recovery data of the historical hearing patients with the highest quality scores.

[0007] In some embodiments, after drawing and displaying a hearing restoration recommendation page including hearing treatment strategies and historical hearing restoration data of historical hearing patients with the highest quality scores on the AI ​​drawing software page platform, the method provided by the present application further includes: Receive the actual restored hearing data of the target hearing patient input by the user on the hearing parameter input page, wherein the actual restored hearing data includes: the lowest sound loudness that the target hearing patient can hear for sounds of different frequencies at different treatment time nodes within a preset period after being treated according to the hearing treatment strategy of the historical hearing patient with the highest quality score; Determine the degree of match between the actual restored hearing data and the historical restored hearing data of the patient with the highest quality score; When the matching degree is lower than the set matching degree threshold, determining the matching degree difference between the matching degree and the matching degree threshold; The hearing impairment portrait data of the target hearing patient, the hearing treatment strategy displayed on the AI ​​drawing software page platform, and the matching difference are input into the pre-trained strategy update module to update the hearing treatment strategy displayed on the AI ​​drawing software page platform, wherein the treatment strategy scoring model is obtained by inputting multiple second training samples into the second neural network for training, and each second training sample includes the hearing impairment portrait data of the historical target hearing patient, the hearing treatment strategy of the historical target hearing patient, the historical matching difference, and the corresponding historical hearing treatment strategy updated by the expert.

[0008] In some implementations, determining the degree of match between the actual restored hearing data and the historical restored hearing data of the historical hearing patient with the highest quality score includes: For each treatment time point, determine the difference between the lowest sound loudness that can be heard at each frequency of the actual restored hearing data and the lowest sound loudness that can be heard at the corresponding frequency in the historical restored hearing data of the patient with the highest quality hearing score; For each treatment time point, determine the mean and covariance of the difference in the lowest sound loudness at each frequency; For each treatment time point, determining the maximum average value of the difference and the minimum average value of the difference among the average values ​​of the multiple differences, and the maximum covariance and the minimum covariance among the covariances of the multiple differences; The degree of match between the actual recovered hearing data and the historical recovered hearing data of the historical hearing patient with the highest quality score is determined based on the average value and covariance of the difference corresponding to each treatment time node, the average value of the largest difference and the average value of the smallest difference among the average values ​​of multiple differences, and the largest covariance and the smallest covariance among the covariances of multiple differences.

[0009] In some implementations, determining the degree of match between the actual restored hearing data and the historical restored hearing data of the historical hearing patient with the highest quality score according to the average value and covariance of the difference corresponding to each treatment time node, the average value of the largest difference and the average value of the smallest difference among the average values ​​of the multiple difference values, and the largest covariance and the smallest covariance among the covariances of the multiple difference values, comprises: Using formula ,determine the matching degree of the actual restored hearing data with the historical restored hearing data of the patient with the highest quality score; in, is the average value of the difference corresponding to each treatment time node, is the covariance of the difference corresponding to each treatment time node, is the average of the maximum differences among the averages of multiple differences, is the average of the minimum differences among the averages of multiple differences, is the largest covariance among multiple covariances, is the smallest covariance among multiple covariances, For the matching degree, is the balancing factor.

[0010] In some embodiments, the historical restored hearing data of the historical hearing patients with the highest quality scores are plotted into multiple line graphs on the AI ​​drawing software page platform, each treatment time node corresponds to a line graph, and the multiple data points in the line graph are the lowest sound loudness that can be heard for sounds of different frequencies.

[0011] In the second aspect, the present application also provides an AI drawing software page platform system based on hearing data, including: A page display unit, used for displaying a hearing parameter input page on an AI drawing software page platform; A data receiving unit is used to receive the disease information, patient attribute characteristics, and initial hearing data of the target hearing patient input by the user on the hearing parameter input page, wherein the disease information includes the disease name, disease severity, and symptom description information, the patient attribute characteristics at least include the patient's age, height, weight, and gender, and the initial hearing data includes the lowest sound loudness that the hearing patient can hear for sounds of different frequencies; A feature conversion unit is used to fuse the symptom information, patient attribute characteristics and initial hearing data of the hearing patient to obtain the hearing impairment portrait data of the target hearing patient, and convert the hearing impairment portrait data into multi-dimensional features of hearing impairment; A similarity determination unit, used to determine the similarity between the multidimensional characteristics of hearing impairment of the target hearing patient and the multidimensional characteristics of hearing impairment of each historical hearing patient in a preset database; A data search unit, used to search for historical restored hearing data of historical hearing patients after hearing restoration treatment whose similarity is higher than a preset similarity threshold, wherein the historical restored hearing data is the lowest sound loudness that the historical hearing patients can hear for sounds of different frequencies at different treatment time nodes within a preset period after starting treatment; A quality scoring unit, used for inputting the multidimensional features of hearing impairment and historical restored hearing data of each historical hearing patient whose similarity is higher than a preset similarity threshold into a pre-trained treatment strategy scoring model to determine the quality score of the corresponding hearing treatment strategy, wherein the treatment strategy scoring model is obtained by inputting a plurality of first training samples into a first neural network for training, and each first training sample includes the historical restored hearing data of the historical hearing patient, the multidimensional features of historical hearing impairment, and the quality score of the corresponding historical hearing treatment strategy; The data search unit is also used to find the hearing treatment strategies and historical restored hearing data of the historical hearing patients with the highest quality scores; The page display unit is also used to draw and display a hearing recovery recommendation page including the hearing treatment strategies and historical hearing recovery data of the historical hearing patients with the highest quality scores on the AI ​​drawing software page platform.

[0012] In some implementations, the data receiving unit is further configured to receive actual restored hearing data of the target hearing patient input by the user on the hearing parameter input page, wherein the actual restored hearing data includes: the lowest sound loudness that the target hearing patient can hear for sounds of different frequencies at different treatment time nodes within a preset period after being treated according to the hearing treatment strategy of the historical hearing patient with the highest quality score; the system further includes: The matching degree determination unit is further used to determine the matching degree between the actual restored hearing data and the historical restored hearing data of the historical hearing patient with the highest quality score; A difference determination unit, used to determine a difference between the matching degree and the matching degree threshold when the matching degree is lower than a set matching degree threshold; A strategy updating unit is used to input the hearing impairment portrait data of the target hearing patient, the hearing treatment strategy displayed on the AI ​​drawing software page platform, and the matching difference into the pre-trained strategy updating module to update the hearing treatment strategy displayed on the AI ​​drawing software page platform, wherein the treatment strategy scoring model is obtained by inputting multiple second training samples into the second neural network for training, and each second training sample includes the hearing impairment portrait data of the historical target hearing patient, the hearing treatment strategy of the historical target hearing patient, the historical matching difference and the corresponding historical hearing treatment strategy updated by the expert.

[0013] In some embodiments, the matching degree determination unit is specifically configured to determine, for each treatment time node, the difference between the lowest sound loudness that can be heard at each frequency of the actual restored hearing data and the lowest sound loudness that can be heard at the corresponding frequency in the historical restored hearing data of the historical hearing patient with the highest quality score; for each treatment time node, determine the average value and covariance of the difference of the lowest sound loudness of each frequency; for each treatment time point, determine the average value of the largest difference and the average value of the smallest difference among the average values ​​of multiple differences, and the maximum covariance and the minimum covariance among the covariances of multiple differences; determine the matching degree between the actual restored hearing data and the historical restored hearing data of the historical hearing patient with the highest quality score according to the average value and covariance of the differences corresponding to each treatment time node, the average value of the largest difference and the average value of the smallest difference among the average values ​​of multiple differences, and the maximum covariance and the minimum covariance among the covariances of multiple differences.

[0014] In some embodiments, the matching degree determination unit is further configured to use the formula ,determine the matching degree of the actual restored hearing data with the historical restored hearing data of the patient with the highest quality score; in, is the average value of the difference corresponding to each treatment time node, is the covariance of the difference corresponding to each treatment time node, is the average of the maximum differences among the averages of multiple differences, is the average of the minimum differences among the averages of multiple differences, is the largest covariance among multiple covariances, is the smallest covariance among multiple covariances, For the matching degree, is the balancing factor.

[0015] In some embodiments, the historical restored hearing data of the historical hearing patients with the highest quality scores are plotted into multiple line graphs on the AI ​​drawing software page platform, each treatment time node corresponds to a line graph, and the multiple data points in the line graph are the lowest sound loudness that can be heard for sounds of different frequencies.

[0016] The present application provides a method and system for platformizing AI drawing software pages based on hearing data, which integrates the symptom information, patient attribute characteristics and initial hearing data of hearing patients to obtain hearing impairment portrait data of target hearing patients, and converts the hearing impairment portrait data into multidimensional features of hearing impairment; determines the similarity between the multidimensional features of hearing impairment of the target hearing patient and the multidimensional features of hearing impairment of each historical hearing patient in a preset database; and searches for historical restored hearing data of historical hearing patients after hearing restoration treatment whose similarity is higher than a preset similarity threshold. It can be understood that when the similarity between the multidimensional features of hearing impairment of the target patient and the multidimensional features of hearing impairment of the historical hearing patient is higher than the preset similarity threshold, if the two adopt the same treatment strategy, the similarity of the improved hearing data is high.

[0017] The multi-dimensional features of hearing impairment and historical restored hearing data of each historical hearing patient whose similarity is higher than a preset similarity threshold are input into the pre-trained treatment strategy scoring model to determine the quality score of the corresponding hearing treatment strategy; since the treatment strategy scoring model is obtained by inputting multiple second training samples into the second neural network for training, each second training sample includes the hearing impairment portrait data of the historical target hearing patient, the hearing treatment strategy of the historical target hearing patient, the historical matching difference, and the corresponding hearing treatment strategy updated by the expert in history, the reliability of the quality score of the corresponding hearing treatment strategy determined is high. The hearing treatment strategy and historical restored hearing data of the historical hearing patient with the highest quality score are found; based on the above, since the two adopt the same treatment strategy, the similarity of the improved hearing data is high, and after the target patient is treated with the hearing treatment strategy of the historical hearing patient with the highest quality score, the quality score of the target patient's restored hearing data is also high. Furthermore, on the AI ​​drawing software page platform, a hearing recovery recommendation page is drawn and displayed, including the hearing treatment strategies and historical hearing recovery data of the historical hearing patients with the highest quality scores, for the target hearing patients to browse, so that the target hearing patients can know the medical strategy that best suits them and the hearing recovery data that can be achieved after treatment. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, a brief introduction will be given below to the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative labor.

[0019] Figure 1 A flowchart of a method for platformizing an AI drawing software page based on hearing data provided in an embodiment of the present application; Figure 2 This is a functional module block diagram of the AI ​​drawing software page platform system based on hearing data provided in an embodiment of the present application. DETAILED DESCRIPTION

[0020] In order to make the purpose, technical solution and advantages of the embodiments of the present application clearer, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments made by ordinary technicians in this field under the enlightenment of the embodiments belong to the scope of protection of the present application.

[0021] The terms "first", "second", "third", "fourth", etc. (if any) in the specification and claims of the present application and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged where appropriate, so that the embodiments of the present application described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0022] See also Figure 1 , the embodiment of the present application provides a method for platformizing an AI drawing software page based on hearing data, including: S101: On the AI ​​drawing software page platform, a hearing parameter input page is displayed.

[0023] S102: receiving the symptom information, patient attribute characteristics, and initial hearing data of the target hearing patient input by the user on the hearing parameter input page.

[0024] Among them, the disease information includes the name of the disease, the severity of the disease, and the symptom description information, the patient attribute characteristics include at least the patient's age, height, weight, and gender, and the initial hearing data includes the minimum sound loudness that the hearing patient can hear for sounds of different frequencies.

[0025] For example, the name of the disease may include central hearing impairment, neurological hearing impairment or cochlear hearing impairment, the severity of the disease may include mild, mild-moderate, moderate, moderate-severe or severe, and the symptom description information may include abnormal cochlear structure or incomplete development, etc.; the initial hearing data may be but is not limited to (125Hz, 60dB HL), (250Hz, 60dB HL),..., (8000Hz, 45dB HL).

[0026] S103: The symptom information, patient attribute characteristics and initial hearing data of the hearing patient are integrated to obtain the hearing impairment portrait data of the target hearing patient, and the hearing impairment portrait data is converted into multi-dimensional features of the hearing impairment.

[0027] S104: Determine the similarity between the multidimensional features of the hearing impairment of the target hearing patient and the multidimensional features of the hearing impairment of each historical hearing patient in a preset database.

[0028] For example, the Euclidean distance between the multidimensional features of hearing impairment of the target hearing patient and the multidimensional features of hearing impairment of each historical hearing patient in a preset database can be calculated to characterize the similarity.

[0029] S105: Searching for historical restored hearing data of historical hearing patients after hearing restoration treatment whose similarity is higher than a preset similarity threshold.

[0030] Among them, the hearing treatment strategies for hearing restoration treatment may include physical therapy strategies, such as the frequency and duration of massaging designated acupoints near the ear (such as Ermen acupoint, Tinggong acupoint, and / or Tinghui acupoint, etc.); drug treatment may include the type of drug (such as nicergoline tablets, methylcobalamin tablets, or ginkgolide injection), dosage and frequency of medication.

[0031] Among them, the historical restored hearing data is the minimum sound loudness that the historical hearing patients can hear for sounds of different frequencies at different treatment time nodes within the preset period after starting treatment.

[0032] For example, the lowest sound loudness that can be heard for sounds of different frequencies at different treatment time points within a preset period after the start of treatment can be the lowest sound loudness that can be heard for sounds of different frequencies at 30 days, 60 days, 90 days, 120 days, 150 days, 180 days, 240 days, 300 days, and 360 days within 1 year after the start of treatment. Sounds of different frequencies can be sounds of 20 Hz, 50 Hz, 100 Hz, 200 Hz, 400 Hz, 600 Hz, ..., and 2000 Hz.

[0033] S106: Input the multi-dimensional characteristics of hearing impairment and historical restored hearing data of each historical hearing patient whose similarity is higher than a preset similarity threshold into a pre-trained treatment strategy scoring model to determine the quality score of the corresponding hearing treatment strategy.

[0034] Among them, the treatment strategy scoring model is obtained by inputting multiple first training samples into the first neural network for training, and each first training sample includes historical hearing recovery data of historical hearing patients, multidimensional characteristics of historical hearing impairment, and quality scores of corresponding historical hearing treatment strategies.

[0035] S107: Find the hearing treatment strategies and historical restored hearing data of the historical hearing patients with the highest quality scores.

[0036] S108: On the AI ​​drawing software page platform, draw and display a hearing recovery recommendation page including the hearing treatment strategies and historical hearing recovery data of the historical hearing patients with the highest quality scores.

[0037] In some embodiments, the historical restored hearing data of the historical hearing patients with the highest quality scores are drawn into multiple line graphs on the AI ​​drawing software page platform, and each treatment time node corresponds to a line graph. The multiple data points in the line graph are the minimum sound loudness that can be heard for sounds of different frequencies. The historical restored hearing data of the historical hearing patients with the highest quality scores can be more intuitively displayed to users.

[0038] In summary, the embodiment of the present application provides an AI drawing software page platform method based on hearing data, which integrates the hearing patient's symptom information, patient attribute characteristics and initial hearing data to obtain the hearing impairment portrait data of the target hearing patient, and converts the hearing impairment portrait data into multidimensional features of hearing impairment; determines the similarity between the multidimensional features of the hearing impairment of the target hearing patient and the multidimensional features of the hearing impairment of each historical hearing patient in a preset database; searches for historical restored hearing data of historical hearing patients after hearing restoration treatment whose similarity is higher than a preset similarity threshold. It can be understood that when the similarity between the multidimensional features of the hearing impairment of the target patient and the multidimensional features of the hearing impairment of the historical hearing patient is higher than the preset similarity threshold, if the two adopt the same treatment strategy, the similarity of the improved hearing data is high.

[0039] The multi-dimensional features of hearing impairment and historical restored hearing data of each historical hearing patient whose similarity is higher than a preset similarity threshold are input into the pre-trained treatment strategy scoring model to determine the quality score of the corresponding hearing treatment strategy; since the treatment strategy scoring model is obtained by inputting multiple second training samples into the second neural network for training, each second training sample includes the hearing impairment portrait data of the historical target hearing patient, the hearing treatment strategy of the historical target hearing patient, the historical matching difference, and the corresponding hearing treatment strategy updated by the expert in history, the reliability of the quality score of the corresponding hearing treatment strategy determined is high. The hearing treatment strategy and historical restored hearing data of the historical hearing patient with the highest quality score are found; based on the above, since the two adopt the same treatment strategy, the similarity of the improved hearing data is high, and after the target patient is treated with the hearing treatment strategy of the historical hearing patient with the highest quality score, the quality score of the target patient's restored hearing data is also high. Furthermore, on the AI ​​drawing software page platform, a hearing recovery recommendation page is drawn and displayed, including the hearing treatment strategies and historical hearing recovery data of the historical hearing patients with the highest quality scores, for the target hearing patients to browse, so that the target hearing patients can know the medical strategy that best suits them and the hearing recovery data that can be achieved after treatment.

[0040] In some implementations, after S108, the method provided in the embodiment of the present application may further include: Step 1: Receive the actual restored hearing data of the target hearing patient input by the user on the hearing parameter input page.

[0041] Among them, the actual restored hearing data includes: the minimum sound loudness that the target hearing patient can hear for sounds of different frequencies at different treatment time nodes within a preset cycle after being treated according to the hearing treatment strategy of the historical hearing patient with the highest quality score.

[0042] Step 2: Determine the degree of match between the actual restored hearing data and the historical restored hearing data of the historical hearing patient with the highest quality score.

[0043] Specifically, step 2 can be implemented as follows: Step A: For each treatment time node, determine the difference between the lowest sound loudness that can be heard at each frequency of the actual restored hearing data and the lowest sound loudness that can be heard at the corresponding frequency in the historical restored hearing data of the historical hearing patient with the highest quality score.

[0044] Step B: For each treatment time node, determine the mean and covariance of the difference in the lowest sound loudness of each frequency.

[0045] Step C: for each treatment time point, determining the maximum average value of the difference and the minimum average value of the difference among the average values ​​of the multiple differences, and the maximum covariance and the minimum covariance among the covariances of the multiple differences.

[0046] Step D: Determine the degree of match between the actual restored hearing data and the historical restored hearing data of the historical hearing patient with the highest quality score based on the average value and covariance of the difference corresponding to each treatment time node, the average value of the largest difference and the average value of the smallest difference among the average values ​​of multiple differences, and the largest covariance and the smallest covariance among the covariances of multiple differences.

[0047] For example, the formula ,determine the matching degree of the actual restored hearing data with the historical restored hearing data of the patients with the highest quality scores.

[0048] in, is the average value of the difference corresponding to each treatment time node, is the covariance of the difference corresponding to each treatment time node, is the average of the maximum differences among the averages of multiple differences, is the average of the minimum differences among the averages of multiple differences, is the largest covariance among multiple covariances, is the smallest covariance among multiple covariances, For the matching degree, is the balancing factor.

[0049] It can be understood that based on the above steps A to D, the matching degree between the actual restored hearing data and the historical restored hearing data of the historical hearing patient with the highest quality score can be determined very accurately with high reliability.

[0050] Step 3: When the matching degree is lower than the set matching degree threshold, determine the matching degree difference between the matching degree and the matching degree threshold.

[0051] Step 4: Input the hearing impairment portrait data of the target hearing patient, the hearing treatment strategy displayed on the AI ​​drawing software page platform, and the matching difference into the pre-trained strategy update module to update the hearing treatment strategy displayed on the AI ​​drawing software page platform.

[0052] Among them, the treatment strategy scoring model is obtained by inputting multiple second training samples into the second neural network for training, and each second training sample includes hearing impairment portrait data of historical target hearing patients, hearing treatment strategies of historical target hearing patients, historical matching degree differences, and corresponding historical hearing treatment strategies updated by experts.

[0053] Steps 1 to 4 can optimize the hearing treatment strategy for the target hearing patient, so that the optimized hearing treatment strategy is more suitable for the treatment of the hearing defect of the target hearing patient, so as to better treat the hearing defect of the target hearing patient in the next cycle.

[0054] See also Figure 2 , the embodiment of the present application also provides an AI drawing software page platform system based on hearing data. It should be noted that the basic principle and technical effects of the AI ​​drawing software page platform system based on hearing data provided in the embodiment of the present application are the same as those in the above embodiment. For the sake of brief description, for the parts not mentioned in the embodiment of the present application, please refer to the corresponding contents in the above embodiment. Figure 2 As shown, the AI ​​drawing software page platform system based on hearing data provided in the embodiment of the present application includes a page display unit, a data receiving unit, a feature conversion unit, a similarity determination unit, a data search unit, and a quality scoring unit; Wherein, the page display unit is used to display the hearing parameter input page on the AI ​​drawing software page platform; A data receiving unit is used to receive the disease information, patient attribute characteristics, and initial hearing data of the target hearing patient input by the user on the hearing parameter input page, wherein the disease information includes the disease name, disease severity, and symptom description information, the patient attribute characteristics at least include the patient's age, height, weight, and gender, and the initial hearing data includes the lowest sound loudness that the hearing patient can hear for sounds of different frequencies; A feature conversion unit is used to fuse the symptom information, patient attribute characteristics and initial hearing data of the hearing patient to obtain the hearing impairment portrait data of the target hearing patient, and convert the hearing impairment portrait data into multi-dimensional features of hearing impairment; A similarity determination unit, used to determine the similarity between the multidimensional characteristics of hearing impairment of the target hearing patient and the multidimensional characteristics of hearing impairment of each historical hearing patient in a preset database; A data search unit, used to search for historical restored hearing data of historical hearing patients after hearing restoration treatment whose similarity is higher than a preset similarity threshold, wherein the historical restored hearing data is the lowest sound loudness that the historical hearing patients can hear for sounds of different frequencies at different treatment time nodes within a preset period after starting treatment; A quality scoring unit, used for inputting the multidimensional features of hearing impairment and historical restored hearing data of each historical hearing patient whose similarity is higher than a preset similarity threshold into a pre-trained treatment strategy scoring model to determine the quality score of the corresponding hearing treatment strategy, wherein the treatment strategy scoring model is obtained by inputting a plurality of first training samples into a first neural network for training, and each first training sample includes the historical restored hearing data of the historical hearing patient, the multidimensional features of historical hearing impairment, and the quality score of the corresponding historical hearing treatment strategy; The data search unit is also used to find the hearing treatment strategies and historical restored hearing data of the historical hearing patients with the highest quality scores; The page display unit is also used to draw and display a hearing recovery recommendation page including the hearing treatment strategies and historical hearing recovery data of the historical hearing patients with the highest quality scores on the AI ​​drawing software page platform.

[0055] In some embodiments, the data receiving unit is further used to receive the actual restored hearing data of the target hearing patient input by the user on the hearing parameter input page, wherein the actual restored hearing data includes: the lowest sound loudness that the target hearing patient can hear for sounds of different frequencies at different treatment time nodes within a preset period after being treated according to the hearing treatment strategy of the historical hearing patient with the highest quality score; the system provided by the embodiment of the present application also includes: The matching degree determination unit is further used to determine the matching degree between the actual restored hearing data and the historical restored hearing data of the historical hearing patient with the highest quality score; A difference determination unit, used to determine a difference between the matching degree and the matching degree threshold when the matching degree is lower than a set matching degree threshold; A strategy updating unit is used to input the hearing impairment portrait data of the target hearing patient, the hearing treatment strategy displayed on the AI ​​drawing software page platform, and the matching difference into the pre-trained strategy updating module to update the hearing treatment strategy displayed on the AI ​​drawing software page platform, wherein the treatment strategy scoring model is obtained by inputting multiple second training samples into the second neural network for training, and each second training sample includes the hearing impairment portrait data of the historical target hearing patient, the hearing treatment strategy of the historical target hearing patient, the historical matching difference and the corresponding historical hearing treatment strategy updated by the expert.

[0056] In some embodiments, the matching degree determination unit is specifically configured to determine, for each treatment time node, the difference between the lowest sound loudness that can be heard at each frequency of the actual restored hearing data and the lowest sound loudness that can be heard at the corresponding frequency in the historical restored hearing data of the historical hearing patient with the highest quality score; for each treatment time node, determine the average value and covariance of the difference of the lowest sound loudness of each frequency; for each treatment time point, determine the average value of the largest difference and the average value of the smallest difference among the average values ​​of multiple differences, and the maximum covariance and the minimum covariance among the covariances of multiple differences; determine the matching degree between the actual restored hearing data and the historical restored hearing data of the historical hearing patient with the highest quality score according to the average value and covariance of the differences corresponding to each treatment time node, the average value of the largest difference and the average value of the smallest difference among the average values ​​of multiple differences, and the maximum covariance and the minimum covariance among the covariances of multiple differences.

[0057] In some embodiments, the matching degree determination unit is further configured to use the formula ,determine the matching degree of the actual restored hearing data with the historical restored hearing data of the patient with the highest quality score; in, is the average value of the difference corresponding to each treatment time node, is the covariance of the difference corresponding to each treatment time node, is the average of the maximum differences among the averages of multiple differences, is the average of the minimum differences among the averages of multiple differences, is the largest covariance among multiple covariances, is the smallest covariance among multiple covariances, For the matching degree, is the balancing factor.

[0058] In some embodiments, the historical restored hearing data of the historical hearing patients with the highest quality scores are plotted into multiple line graphs on the AI ​​drawing software page platform, each treatment time node corresponds to a line graph, and the multiple data points in the line graph are the lowest sound loudness that can be heard for sounds of different frequencies.

[0059] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit it. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or replace some or all of the technical features therein with equivalents. However, these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present application.

Claims

1. A method for platformizing AI drawing software pages based on hearing data, characterized in that: The method comprises: On the AI ​​drawing software page platform, the hearing parameter input page is displayed; Receive the target hearing patient's disease information, patient attribute characteristics, and initial hearing data input by the user on the hearing parameter input page, wherein the disease information includes the disease name, disease severity, and symptom description information, the patient attribute characteristics at least include the patient's age, height, weight, and gender, and the initial hearing data includes the lowest sound loudness that the hearing patient can hear for sounds of different frequencies; The symptom information of the target hearing patient, the patient attribute characteristics and the initial hearing data are integrated to obtain hearing impairment portrait data of the target hearing patient, and the hearing impairment portrait data is converted into multi-dimensional features of hearing impairment; Determine the similarity between the multidimensional characteristics of hearing impairment of the target hearing patient and the multidimensional characteristics of hearing impairment of each historical hearing patient in a preset database; Searching for historical restored hearing data of the historical hearing patients after hearing restoration treatment whose similarity is higher than a preset similarity threshold, wherein the historical restored hearing data is the lowest sound loudness that the historical hearing patients can hear for sounds of different frequencies at different treatment time points within a preset period after starting treatment; Inputting the multidimensional features of hearing impairment and historical restored hearing data of each historical hearing patient whose similarity is higher than a preset similarity threshold into a pre-trained treatment strategy scoring model to determine the quality score of the corresponding hearing treatment strategy, wherein the treatment strategy scoring model is obtained by inputting a plurality of first training samples into a first neural network for training, and each of the first training samples includes the historical restored hearing data of the historical hearing patient, the multidimensional features of historical hearing impairment, and the quality score of the corresponding historical hearing treatment strategy; Find the hearing treatment strategies and historical restored hearing data of the historical hearing patients with the highest quality scores; On the AI ​​drawing software page platform, a hearing recovery recommendation page including the hearing treatment strategies and historical hearing recovery data of the historical hearing patients with the highest quality scores is drawn and displayed.

2. The method according to claim 1, characterized in that After drawing and displaying a hearing restoration recommendation page including hearing treatment strategies and historical hearing restoration data of historical hearing patients with the highest quality scores on the AI ​​drawing software page platform, the method further includes: receiving the actual restored hearing data of the target hearing patient input by the user on the hearing parameter input page, wherein the actual restored hearing data includes: the lowest sound loudness that the target hearing patient can hear for sounds of different frequencies at different treatment time nodes within a preset period after being treated according to the hearing treatment strategy of the historical hearing patient with the highest quality score; Determining a degree of match between the actual restored hearing data and the historical restored hearing data of the historical hearing patient with the highest quality score; When the matching degree is lower than a set matching degree threshold, determining a matching degree difference between the matching degree and the matching degree threshold; The hearing impairment portrait data of the target hearing patient, the hearing treatment strategy displayed on the AI ​​drawing software page platform, and the matching difference are input into the pre-trained strategy update module to update the hearing treatment strategy displayed on the AI ​​drawing software page platform, wherein the treatment strategy scoring model is obtained by inputting multiple second training samples into the second neural network for training, and each of the second training samples includes the hearing impairment portrait data of the historical target hearing patient, the hearing treatment strategy of the historical target hearing patient, the historical matching difference, and the corresponding historical hearing treatment strategy updated by the expert.

3. The method according to claim 2, characterized in that Determining the matching degree between the actual restored hearing data and the historical restored hearing data of the historical hearing patient with the highest quality score includes: For each treatment time node, determining the difference between the lowest sound loudness that can be heard at each frequency of the actual restored hearing data and the lowest sound loudness that can be heard at the corresponding frequency in the historical restored hearing data of the historical hearing patient with the highest quality score; For each treatment time point, determine the mean and covariance of the difference in the lowest sound loudness at each frequency; For each treatment time point, determining the maximum average value of the difference and the minimum average value of the difference among the average values ​​of the multiple differences, and the maximum covariance and the minimum covariance among the covariances of the multiple differences; The degree of match between the actual recovered hearing data and the historical recovered hearing data of the historical hearing patient with the highest quality score is determined based on the average value and covariance of the difference corresponding to each treatment time node, the average value of the largest difference and the average value of the smallest difference among the average values ​​of multiple differences, and the largest covariance and the smallest covariance among the covariances of multiple differences.

4. The method according to claim 3, characterized in that Determining the degree of match between the actual restored hearing data and the historical restored hearing data of the historical hearing patient with the highest quality score according to the average value and covariance of the difference corresponding to each treatment time node, the average value of the largest difference and the average value of the smallest difference among the average values ​​of the multiple differences, and the largest covariance and the smallest covariance among the covariances of the multiple differences, comprises: Using formula , determining a matching degree between the actual restored hearing data and the historical restored hearing data of the historical hearing patient with the highest quality score; in, is the average value of the difference corresponding to each treatment time node, is the covariance of the difference corresponding to each treatment time node, is the average of the maximum differences among the averages of multiple differences, is the average of the minimum differences among the averages of multiple differences, is the largest covariance among multiple covariances, is the smallest covariance among multiple covariances, is the matching degree, is the balancing factor.

5. The method according to any one of claims 1 to 4, characterized in that: The historical restored hearing data of the historical hearing patient with the highest quality score is drawn into multiple line graphs on the AI ​​drawing software page platform, each treatment time node corresponds to a line graph, and the multiple data points in the line graph are the lowest sound loudness that can be heard for sounds of different frequencies.

6. An AI drawing software page platform system based on hearing data, characterized in that: The system comprises: A page display unit, used for displaying a hearing parameter input page on an AI drawing software page platform; a data receiving unit, configured to receive the target hearing patient's symptom information, patient attribute characteristics, and initial hearing data input by the user on the hearing parameter input page, wherein the symptom information includes the symptom name, symptom severity, and symptom description information, the patient attribute characteristics at least include the patient's age, height, weight, and gender, and the initial hearing data includes the lowest sound loudness that the hearing patient can hear for sounds of different frequencies; A feature conversion unit, configured to fuse the symptom information of the hearing patient, the patient attribute characteristics and the initial hearing data to obtain hearing impairment portrait data of the target hearing patient, and convert the hearing impairment portrait data into multi-dimensional features of hearing impairment; A similarity determination unit, used to determine the similarity between the multidimensional characteristics of hearing impairment of the target hearing patient and the multidimensional characteristics of hearing impairment of each historical hearing patient in a preset database; A data search unit, configured to search for historical restored hearing data of the historical hearing patient after hearing restoration treatment whose similarity is higher than a preset similarity threshold, wherein the historical restored hearing data is the lowest sound loudness that the historical hearing patient can hear for sounds of different frequencies at different treatment time nodes within a preset period after starting treatment; A quality scoring unit, used for inputting the multidimensional features of hearing impairment and historical restored hearing data of each historical hearing patient whose similarity is higher than a preset similarity threshold into a pre-trained treatment strategy scoring model to determine the quality score of the corresponding hearing treatment strategy, wherein the treatment strategy scoring model is obtained by inputting a plurality of first training samples into a first neural network for training, and each of the first training samples includes the historical restored hearing data of the historical hearing patient, the multidimensional features of historical hearing impairment, and the quality score of the corresponding historical hearing treatment strategy; The data search unit is further used to search for the hearing treatment strategy and historical hearing recovery data of the historical hearing patient with the highest quality score; The page display unit is also used to draw and display a hearing recovery recommendation page including the hearing treatment strategies and historical hearing recovery data of the historical hearing patients with the highest quality scores on the AI ​​drawing software page platform.

7. The system according to claim 6, characterized in that The data receiving unit is further configured to receive actual restored hearing data of the target hearing patient input by the user on the hearing parameter input page, wherein the actual restored hearing data includes: the lowest sound loudness that the target hearing patient can hear for sounds of different frequencies at different treatment time nodes within a preset period after being treated according to the hearing treatment strategy of the historical hearing patient with the highest quality score; the system further includes: The matching degree determination unit is further used to determine the matching degree between the actual restored hearing data and the historical restored hearing data of the historical hearing patient with the highest quality score; a difference determination unit, configured to determine a difference between the matching degree and the matching degree threshold when the matching degree is lower than a set matching degree threshold; A strategy updating unit is used to input the hearing impairment portrait data of the target hearing patient, the hearing treatment strategy displayed on the AI ​​drawing software page platform, and the matching difference into a pre-trained strategy updating module to update the hearing treatment strategy displayed on the AI ​​drawing software page platform, wherein the treatment strategy scoring model is obtained by inputting multiple second training samples into a second neural network for training, and each of the second training samples includes the hearing impairment portrait data of historical target hearing patients, the hearing treatment strategies of historical target hearing patients, historical matching differences, and the corresponding historical hearing treatment strategies updated by experts.

8. The system according to claim 7, characterized in that The matching degree determination unit is specifically used to determine, for each treatment time node, the difference between the lowest sound loudness that can be heard at each frequency of the actual restored hearing data and the lowest sound loudness that can be heard at the corresponding frequency in the historical restored hearing data of the historical hearing patient with the highest quality score; for each treatment time node, determine the average value and covariance of the difference of the lowest sound loudness of each frequency; for each treatment time point, determine the average value of the largest difference and the average value of the smallest difference among the average values ​​of multiple differences, and the largest covariance and the smallest covariance among the covariances of multiple differences; determine the matching degree of the actual restored hearing data with the historical restored hearing data of the historical hearing patient with the highest quality score according to the average value and covariance of the differences corresponding to each treatment time node, the average value of the largest difference and the average value of the smallest difference among the average values ​​of multiple differences, and the largest covariance and the smallest covariance among the covariances of multiple differences.

9. The system according to claim 8, characterized in that The matching degree determination unit is further configured to use the formula , determining a matching degree between the actual restored hearing data and the historical restored hearing data of the historical hearing patient with the highest quality score; in, is the average value of the difference corresponding to each treatment time node, is the covariance of the difference corresponding to each treatment time node, is the average of the maximum differences among the averages of multiple differences, is the average of the minimum differences among the averages of multiple differences, is the largest covariance among multiple covariances, is the smallest covariance among multiple covariances, is the matching degree, is the balancing factor.

10. The system according to any one of claims 6 to 9, characterized in that: The historical restored hearing data of the historical hearing patient with the highest quality score is drawn into multiple line graphs on the AI ​​drawing software page platform, each treatment time node corresponds to a line graph, and the multiple data points in the line graph are the lowest sound loudness that can be heard for sounds of different frequencies.

Images