Tongue image medical information processing method and system
By analyzing the entropy sequence and frequency consistency of the tongue image and dynamically determining the acquisition parameters, the problem of complex image acquisition scenarios in household medical science is solved, the efficiency and accuracy of tongue image information acquisition is improved, and efficient auxiliary diagnosis is achieved .
Patent Information
- Application Number
- CN202410728080.6
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-06-06
- Publication Date
- 2024-11-22
- Estimated Expiration
- 2044-06-06
AI Technical Summary
In the case of home use, medical image acquisition scenarios are complex, and the prior art is difficult to effectively reduce the difficulty of collecting tongue image information, and the collected information is not accurate enough, which affects the efficiency and accuracy of auxiliary diagnosis.
By setting acquisition parameters of different variable values, the entropy value sequence and frequency consistency of the tongue image are analyzed, the active acquisition parameters and inactive acquisition parameters are dynamically determined, the difficulty of collecting user image information is reduced, and the intelligent analysis model is used to perform tongue image images analyze.
In different home acquisition scenarios, dynamically determine the activity acquisition parameters, reduce information acquisition losses, improve collection efficiency and accuracy, provide analysis results based on platform credibility and big data, and reduce the difficulty of information acquisition.
Smart Images

Figure CN118356163B_ABST
Abstract
Description
Technical Field
[0001] The present invention belongs to the technical field of intelligent health management, and in particular relates to a tongue image medical information processing method and system. Background Art
[0002] With the accelerated pace of modern society and the increase in life pressure, health issues are receiving increasing attention, and the development of intelligent technology has brought new opportunities for health management. At present, social medical resources are in short supply. In the traditional process of doctor-patient consultation and visual examination, doctors need to communicate and interact with patients many times to obtain basic personal information such as age and gender of patients, as well as the description of the most obvious disease manifestations felt by patients, and then give corresponding diagnostic suggestions. However, this method is limited to the time and space when doctors and patients appear together, and it is heavily dependent on the doctor's personal medical judgment ability, the patient's language expression ability and memory ability, lacks objective and detailed comparison of human body signs and images, and is not accurate and efficient.
[0003] With the rapid development of intelligent medical imaging diagnosis, in order to meet the increasingly complex requirements of medical image analysis and processing, artificial intelligence methods have become a research hotspot in the development of medical image processing technology in recent years. Using artificial intelligence methods such as machine learning, deep learning and neural network algorithms to classify and identify images is a common analysis method in industry and academia. And the methods of using artificial intelligence technology to assist medical diagnosis are becoming more and more diverse. For example, in the field of medical imaging, more and more AI-assisted diagnosis systems can automatically classify target subjects through image segmentation, recognition and other technologies.
[0004] In addition, advanced computer technologies represented by artificial intelligence and big data have been widely used in the field of routine health examinations; in particular, machine learning technologies represented by deep learning have received widespread attention in the field of medical imaging. For example, for the diagnosis of oral, dental, eye and other diseases that cannot be identified by the naked eye. For another example, in fundus image detection, deep learning technology can accurately detect a certain feature of fundus images, such as using a large number of samples of macular holes to train deep learning models, and using the trained model to detect macular holes in fundus images. After combining with big data technology, more advice and diagnostic information about physical health can be obtained through in-depth analysis of body parts; for example, the body advice information obtained after scanning the retina is already very accurate. Based on this, it can be seen that image acquisition and intelligent analysis of the user's area of interest to obtain effective health auxiliary information, and related research on obtaining effective information through in-depth analysis of image information and continuous analysis of time duration has become a current hot topic, and the image information that patients collect and provide for medical assistance in non-medical time and space is the basis for improving medical efficiency and diagnostic accuracy with the assistance of artificial intelligence and big data.
[0005] Furthermore, how to make medical information processing for user image analysis home-based, adapt to the cooperation level of different users, and collect enough information to support subsequent intelligent image analysis on the basis of reducing the difficulty of information collection is a technical problem to be solved. Based on the above problems, the present invention aims at the complex collection scenarios in home situations, and facilitates the dynamic determination of activity collection parameters in different home collection scenarios through collection parameter classification, thereby flexibly reducing the difficulty of collecting user image information. Summary of the invention
[0006] In order to solve the above problems in the prior art, the present invention proposes a tongue image medical information processing method and system, the method comprising:
[0007] Step S1: acquiring different tongue images corresponding to the different change values when each acquisition parameter is set to different change values, analyzing the consistency of the acquired different tongue images, and dividing the acquisition parameters into active acquisition parameters and inactive acquisition parameters; wherein: the active acquisition parameters keep changing in the subsequent acquisition process, while the inactive acquisition parameters remain unchanged in the subsequent acquisition process;
[0008] The step S1 specifically includes the following steps:
[0009] Step S11: sequentially obtaining an unprocessed acquisition parameter as a current acquisition parameter;
[0010] Step S12: setting all non-current acquisition parameters to default values, and making the current acquisition parameters change within their parameter ranges, corresponding to the change values of multiple current acquisition parameters;
[0011] Step S13: acquiring different tongue images when the current acquisition parameter has different change values; determining different entropy values corresponding to the different tongue images based on the different tongue images; arranging the entropy values corresponding to the different change values of the current acquisition parameter in order to form an entropy value sequence corresponding to the current acquisition parameter k ; Determine whether all acquisition parameters have been processed, if so, proceed to the next step, otherwise, return to step S11;
[0012] Step S14: Based on the corresponding entropy value sequence of each acquisition parameter k Compare the sequences of adjacent changes in element values to determine the frequency consistency of acquisition parameters; the step S14 specifically includes the following steps:
[0013] Step S141: Obtain the entropy value sequence corresponding to each acquisition parameter k in sequence ; where u is the entropy value sequence The element number in; U is the number of elements in the entropy value sequence;
[0014] Step S142: Calculate adjacent element difference sequence , ;set up ;in: is the element mean of the entropy sequence;
[0015] Step S142: Calculate the unordered distance between any two adjacent element difference sequences respectively ; Among them: the unordered distance means that when comparing two sequences, the position of the elements in the sequence is not considered, but only the order of the element values is considered to compare their element values; specifically: the unordered distance is calculated using the following formula (1): ;in: and is the difference sequence of adjacent elements corresponding to any two acquisition parameters; It is the order of the element values in the adjacent element difference sequence; (1);
[0016] Step S143: Determine the mean of the disordered distance of each acquisition parameter k respectively Specifically: the following formula (2) is used to calculate the mean of the disordered distance of the acquisition parameter k: ; Where: K is the number of acquisition parameters;
[0017] (2);
[0018] Step S15: Based on the corresponding entropy value sequence of each acquisition parameter k The sequence comparison of the element value change in the acquisition parameters is performed to determine the amplitude consistency of the acquisition parameters, and then the acquisition parameters are classified according to the change; the step S15 specifically includes the following steps:
[0019] Step S151: Obtain the entropy value sequence corresponding to each acquisition parameter k in sequence ; where u is the entropy value sequence The element number in; U is the number of elements in the entropy value sequence;
[0020] Step S152: Calculate the mean difference sequence ;in: is the mean of the elements in the entropy sequence;
[0021] Step S153: Normalize the mean difference sequence and set ;
[0022] Step S154: determine whether all entropy value sequences have been processed, if yes, return to step S151, otherwise proceed to the next step;
[0023] Step S155: Calculate the ordered distance between any two mean difference sequences respectively ; Among them: ordered distance refers to the comparison of element values while considering the position of elements in the sequence when comparing two sequences; specifically: the ordered distance is calculated using the following formula (3) ;in: and is the mean difference sequence corresponding to any two acquisition parameters; is the position of the element in the mean difference sequence;
[0024] (3);
[0025] Step S155: Calculate the ordered distance The mean of the ordered distance Specifically, the ordered distance mean is calculated using the following formula (4): ;
[0026] (4);
[0027] Step S155: When the ordered distance between the mean difference sequence corresponding to the acquisition parameter k and other mean difference sequences Greater than the mean of the ordered distance When the proportion exceeds the preset proportion, it is placed in the first category; otherwise, it is placed in the second category; the change classification includes the first category and the second category;
[0028] Step S16: determining the active acquisition parameters and the inactive acquisition parameters according to the amplitude consistency and frequency consistency of the entropy value sequence; specifically: distinguishing the active acquisition parameters and the inactive acquisition parameters from the first category and the second category according to the disordered distance mean;
[0029] Step S2: When the activity acquisition parameters are changed and other acquisition parameters remain unchanged, the tongue image information of the user is acquired; and the tongue image information is input into the intelligent analysis model to obtain the tongue image analysis result.
[0030] Furthermore, the entropy value is a feature value of the tongue image as a whole or locally.
[0031] Furthermore, the overall or local characteristic value of the tongue image includes one or a combination of the tongue image grayscale, brightness mean, gray pixel value at a specific position and / or gradient change value.
[0032] Furthermore, the preset ratio is 50-100%.
[0033] Furthermore, the activity collection parameters are 2.
[0034] A tongue image medical information processing system comprises: an intelligent analysis server and a mobile terminal; the intelligent analysis server is used to implement the above-mentioned tongue image medical information processing method.
[0035] Furthermore, the server is an artificial intelligence analysis server.
[0036] A tongue image medical information processing platform comprises a processor, the processor is coupled to a memory, the memory stores program instructions, and when the program instructions stored in the memory are executed by the processor, the tongue image medical information processing method is implemented.
[0037] A computer-readable storage medium includes a program, which, when running on a computer, enables the computer to execute the tongue image medical information processing method.
[0038] An intelligent analysis server is configured to execute the tongue image medical information processing method.
[0039] The beneficial effects of the present invention include:
[0040] (1) In view of the complex acquisition scenarios in home use, an acquisition parameter classification process is set to obtain different tongue images corresponding to different change values when each acquisition parameter is set to different change values, and the consistency of the different tongue images obtained is analyzed. The active acquisition parameters and the inactive acquisition parameters are determined according to the amplitude consistency and frequency consistency of the entropy value sequence, so as to facilitate the dynamic determination of the active acquisition parameters in different home acquisition scenarios, thereby flexibly reducing the difficulty of acquiring user image information;
[0041] Furthermore, by combining the disordered distance with the ordered distance, the possible frequency changes in the asynchronous state and the amplitude changes in the synchronous state can be found through quantitative calculation, thus reducing the loss of all collected image information.
[0042] (2) Perform multi-analysis mode analysis on tongue images with different change values, and synthesize the analysis results based on the platform credibility and big data accuracy to provide users with effective information;
[0043] (3) Collect the tongue image analysis results and tongue image information at different continuous acquisition times, and analyze to obtain the best combination of change values of activity acquisition parameters that conforms to the user's acquisition habits; this combination of change values is a personalized combination of change values, which can further reduce the difficulty of information acquisition. BRIEF DESCRIPTION OF THE DRAWINGS
[0044] The drawings described herein are used to provide a further understanding of the present invention and constitute a part of the present application, but do not constitute an improper limitation of the present invention. In the drawings:
[0045] Figure 1 Schematic diagram of the tongue image medical information processing method provided by the present invention. DETAILED DESCRIPTION
[0046] The present invention will be described in detail below in conjunction with the accompanying drawings and specific embodiments, wherein the illustrative embodiments and descriptions are only used to explain the present invention but are not intended to limit the present invention.
[0047] The present invention proposes a tongue image medical information processing method and system, as shown in the attached Figure 1 As shown, the method comprises the following steps:
[0048] Step S1: acquiring different tongue images corresponding to the different change values when each acquisition parameter is set to different change values, analyzing the consistency of the acquired different tongue images, and dividing the acquisition parameters into active acquisition parameters and inactive acquisition parameters; wherein: the active acquisition parameters keep changing in the subsequent acquisition process, while the inactive acquisition parameters remain unchanged in the subsequent acquisition process;
[0049] The step S1 specifically includes the following steps:
[0050] Step S11: sequentially obtaining an unprocessed acquisition parameter as a current acquisition parameter;
[0051] Step S12: setting all non-current acquisition parameters to default values, and making the current acquisition parameters change within their parameter ranges, corresponding to the change values of multiple current acquisition parameters;
[0052] Preferably, when setting the activity parameters to default values, the setting method is conventional setting, setting that is convenient for users to adopt, etc., so that the current acquisition parameters change regularly and evenly within the parameter range; for example, according to a preset sequence, the acquisition parameters are changed in sequence from their starting values within the parameter range; for example, the angle of rotation of the tongue is slowly raised from the position facing the main light source;
[0053] Step S13: acquiring different tongue images when the current acquisition parameter has different change values; determining different entropy values corresponding to the different tongue images based on the different tongue images; arranging the entropy values corresponding to the different change values of the current acquisition parameter in order to form an entropy value sequence corresponding to the current acquisition parameter k ; Determine whether all acquisition parameters have been processed, if so, proceed to the next step, otherwise, return to step S11;
[0054] Preferably, the entropy value is a feature value of the tongue image as a whole or locally, such as the grayscale of the tongue image, the brightness mean, the gray pixel value at a specific position, the gradient change value, etc.;
[0055] Step S14: Based on the corresponding entropy value sequence of each acquisition parameter k Compare the sequences of adjacent changes in element values to determine the frequency consistency of acquisition parameters; the step S14 specifically includes the following steps:
[0056] Step S141: Obtain the entropy value sequence corresponding to each acquisition parameter k in sequence ; where u is the entropy value sequence The element number in; U is the number of elements in the entropy value sequence;
[0057] Step S142: Calculate adjacent element difference sequence , ;set up ;in: is the element mean of the entropy sequence;
[0058] Alternative: Setting ;
[0059] Step S142: Calculate the unordered distance between any two adjacent element difference sequences respectively ; Among them: the unordered distance means that when comparing two sequences, the position of the elements in the sequence is not considered, but only the order of the element values is considered to compare their element values; specifically: the unordered distance is calculated using the following formula (1): ;in: and is the difference sequence of adjacent elements corresponding to any two acquisition parameters; It is the order of the element values in the adjacent element difference sequence; for example: Indicated in The element with the second highest value in the element;
[0060] (1);
[0061] Step S143: Determine the mean of the disordered distance of each acquisition parameter k respectively Specifically: the following formula (2) is used to calculate the mean of the disordered distance of the acquisition parameter k: ; Where: K is the number of acquisition parameters;
[0062] (2);
[0063] This is because the tongue image corresponding to the acquisition parameters may change during the acquisition parameter change process, and there are certain rules between these changes, but the changes are not synchronous, that is, inconsistent, and even the change rules are different; the disordered distance can be used to find possible asynchronous similar changes; the acquisition parameters The smaller it is, the better it can represent the changes in the collected information brought about by changes in other collection parameters; and the collection parameters The larger the value, the more likely it is that some general information collection losses may occur if the collection parameters are fixed.
[0064] Step S15: Based on the corresponding entropy value sequence of each acquisition parameter k The sequence comparison of the element value change in the acquisition parameters is performed to determine the amplitude consistency of the acquisition parameters, and then the acquisition parameters are classified according to the change; the step S15 specifically includes the following steps:
[0065] Step S151: Obtain the entropy value sequence corresponding to each acquisition parameter k in sequence ; where u is the entropy value sequence The element number in; U is the number of elements in the entropy value sequence;
[0066] Step S152: Calculate the mean difference sequence ;in: is the mean of the elements in the entropy sequence;
[0067] Alternative: ;
[0068] Step S153: Normalize the mean difference sequence and set ;
[0069] Step S154: determine whether all entropy value sequences have been processed, if yes, return to step S151, otherwise proceed to the next step;
[0070] Step S155: Calculate the ordered distance between any two mean difference sequences respectively ; Among them: ordered distance refers to the comparison of element values while considering the position of elements in the sequence when comparing two sequences; specifically: the ordered distance is calculated using the following formula (3) ;in: and is the mean difference sequence corresponding to any two acquisition parameters; is the position of the element in the mean difference sequence;
[0071] (3);
[0072] Step S155: Calculate the ordered distance The mean of the ordered distance Specifically, the ordered distance mean is calculated using the following formula (4): ;
[0073] (4);
[0074] Step S155: When the ordered distance between the mean difference sequence corresponding to the acquisition parameter k and other mean difference sequences Greater than the mean of the ordered distance When the ratio exceeds the preset ratio, it is placed in the first category; otherwise, it is placed in the second category; wherein: the change classification includes the first category and the second category;
[0075] Preferably: the preset ratio is 50-100%;
[0076] Preferably: the preset ratio is 90-100%;
[0077] Step S16: determining active acquisition parameters and inactive acquisition parameters according to the amplitude consistency and frequency consistency of the entropy value sequence; wherein: the active acquisition parameters are one or more; specifically: determining active acquisition parameters and inactive acquisition parameters from the first category and the second category according to the mean of the disordered distance; of course, as the types of acquisition parameters increase, the number of categories can increase accordingly; the disordered distance and the ordered distance can still cooperate with each other to distinguish the acquisition parameters;
[0078] Preferably: the one with the smallest mean value of disordered distance in the first category and / or the one with the smallest mean value of disordered distance in the second category are used as activity collection parameters;
[0079] Alternatively: using the one with the largest mean value of disordered distance in the first category and / or the one with the smallest mean value of disordered distance in the second category as the activity acquisition parameter;
[0080] Alternatively: the activity collection parameters are 1 or 2;
[0081] Step S2: When the activity acquisition parameter changes and other acquisition parameters remain unchanged, the user's tongue image information is acquired; the tongue image information is input into the intelligent analysis model to obtain the tongue image analysis result;
[0082] Preferably: under the guidance of information, the user is made to cooperate with the changes in the acquisition parameters; for example: the guiding information is voice / video information;
[0083] Preferably: the intelligent analysis model is an artificial intelligence model;
[0084] Preferably: the intelligent analysis model is a neural network model;
[0085] The step S2 specifically includes the following steps:
[0086] Step S21: Collect parameters for each activity in turn , set other acquisition parameters to default values, and make the activity acquisition parameters vary within the range, so as to obtain the same acquisition parameters as each activity Each change in value Corresponding tongue image ;
[0087] Step S22: inputting the tongue image into the intelligent analysis model and / or the tongue image platform to obtain the tongue image analysis result;
[0088] Preferably: before executing this step, an artificial intelligence model for analyzing tongue images is constructed and trained, wherein the input of the artificial intelligence model is the tongue image, and the output result is the tongue image analysis result. ; Of course, the changes in the acquisition parameter ka can also be used as input to the artificial intelligence model; but this may increase the accuracy of the analysis results and the complexity of training;
[0089] Preferably: the tongue image platform is used to analyze the tongue image to obtain the tongue image analysis result; the analysis method here can be a traditional image analysis method or a manual analysis method or other methods;
[0090] Step S3: Based on the values corresponding to different changes Tongue image analysis results , and obtain the tongue image information corresponding to the current acquisition time t; the tongue image analysis result is expressed as ,in: The n classification results corresponding to the tongue image analysis results are as follows: Tongue image analysis results After comprehensive analysis, the tongue image information corresponding to the most tongue image analysis results is obtained as the tongue image information corresponding to the current acquisition time t ;
[0091] Further: the step S3 is specifically: using the following formula (5) for comprehensive analysis; wherein: is the accuracy of the tongue image analysis results obtained when the activity acquisition parameter ka is used in the historical data; The tongue image analysis results are obtained The credibility of the intelligent analysis model and / or tongue imaging platform;
[0092] (5);
[0093] Preferred: Involving The accuracy rate can be determined by manual review, and finally adopted by statistics, and determined to be tongue image information, etc.;
[0094] Preferably, the step S3 further comprises: obtaining the tongue image information As the result of the tongue image medical information processing, the tongue image information at this moment is fed back to the user. It is real-time feedback information;
[0095] The method further comprises step S4;
[0096] Step S4: collecting tongue image analysis results and tongue image information at different continuous acquisition times t, analyzing them to obtain the best combination of change values of active acquisition parameters; in the subsequent tongue image acquisition process, setting the active acquisition parameters to the combination of change values, and setting the inactive acquisition parameters to the default values to acquire tongue images, inputting the tongue images obtained under the guidance of the acquisition parameters into the intelligent analysis model and / or the tongue image platform to obtain tongue image analysis results; and using the tongue image analysis results as the final tongue image information; at this time, there is no need to integrate the tongue image analysis results;
[0097] The analysis is performed to obtain the best combination of change values of the activity acquisition parameters, specifically: statistics of different change values Corresponding tongue image analysis results And its final tongue image information Keep the same number of times, and then count the collection parameters for each activity For example, the change value corresponding to the one with the highest number of consistent times is called the optimal change value. , the best change values of various types of activity collection parameters are combined to obtain the best change value combination of the activity collection parameters; that is, if the activity collection parameters are several tuples, then the change value combination is also several tuples, where each tuple corresponds to the best setting change value of the activity collection parameter;
[0098] Based on the same inventive concept, the present invention also provides a tongue image medical information processing system, the system comprising: an intelligent analysis server and a mobile terminal; the intelligent analysis server is used to implement the above-mentioned tongue image medical information processing method; the mobile terminal is used to set acquisition parameters and obtain tongue images in an acquisition environment constructed by the acquisition parameters; and send the acquired tongue images to the intelligent analysis server;
[0099] Preferably: the number of the intelligent analysis server is one or more;
[0100] Preferably: the mobile terminal is a mobile phone, a mobile camera, or a small camera;
[0101] A computer program (also referred to as a program, software, software application, script, or code) can be written in any form of a programming language, including an assembled or interpreted language, a declarative or procedural language, and it can be deployed in any form, including as a stand-alone program or as a module, component, subroutine, object, or other unit suitable for use in a computing environment. A computer program may, but need not, correspond to a file in a file system. A program can be stored in a portion of a file that holds other programs or data (e.g., one or more scripts stored in a markup language document), in a single file dedicated to the program, or in multiple collaborative files (e.g., files storing one or more modules, subroutines, or code portions). A computer program can be deployed to execute on one computer or on multiple computers located at one site or distributed across multiple sites and interconnected by a communication network.
[0102] It will be appreciated by those skilled in the art that embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Furthermore, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.
[0103] The present invention is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of the processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.
[0104] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.
[0105] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.
[0106] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the above embodiments, ordinary technicians in the relevant field should understand that the specific implementation methods of the present invention can still be modified or replaced by equivalents. Any modification or equivalent replacement that does not depart from the spirit and scope of the present invention should be covered within the scope of protection of the claims of the present invention.
Claims
1. A method for processing medical information of tongue images, characterized in that: The method comprises: Step S1: acquiring different tongue images corresponding to the different change values when each acquisition parameter is set to different change values, analyzing the consistency of the acquired different tongue images, and dividing the acquisition parameters into active acquisition parameters and inactive acquisition parameters; wherein: the active acquisition parameters keep changing in the subsequent acquisition process, while the inactive acquisition parameters remain unchanged in the subsequent acquisition process; The step S1 specifically includes the following steps: Step S11: sequentially obtaining an unprocessed acquisition parameter as a current acquisition parameter; Step S12: setting all non-current acquisition parameters to default values, and making the current acquisition parameters change within their parameter ranges, corresponding to the change values of multiple current acquisition parameters; Step S13: acquiring different tongue images when the current acquisition parameter has different change values; determining different entropy values corresponding to the different tongue images based on the different tongue images; arranging the entropy values corresponding to the different change values of the current acquisition parameter in order to form an entropy value sequence corresponding to the current acquisition parameter k ; Determine whether all acquisition parameters have been processed, if so, proceed to the next step, otherwise, return to step S11; Step S14: Based on the corresponding entropy value sequence of each acquisition parameter k Compare the sequences of adjacent changes in element values to determine the frequency consistency of acquisition parameters; the step S14 specifically includes the following steps: Step S141: Obtain the entropy value sequence corresponding to each acquisition parameter k in sequence ; where u is the entropy value sequence The element number in; U is the number of elements in the entropy value sequence; Step S142: Calculate adjacent element difference sequence , ;set up ;in: is the element mean of the entropy sequence; Step S142: Calculate the unordered distance between any two adjacent element difference sequences respectively ; Among them: the unordered distance means that when comparing two sequences, the position of the elements in the sequence is not considered, but only the order of the element values is considered to compare their element values; specifically: the unordered distance is calculated using the following formula (1): ;in: and is the difference sequence of adjacent elements corresponding to any two acquisition parameters; It is the order of the element values in the adjacent element difference sequence; (1); Step S143: Determine the mean of the disordered distance of each acquisition parameter k respectively Specifically: the following formula (2) is used to calculate the mean of the disordered distance of the acquisition parameter k: ; Where: K is the number of acquisition parameters; (2); Step S15: Based on the corresponding entropy value sequence of each acquisition parameter k The sequence comparison of the element value change in the acquisition parameters is performed to determine the amplitude consistency of the acquisition parameters, and then the acquisition parameters are classified according to the change; the step S15 specifically includes the following steps: Step S151: Obtain the entropy value sequence corresponding to each acquisition parameter k in sequence ; where u is the entropy value sequence The element number in; U is the number of elements in the entropy value sequence; Step S152: Calculate the mean difference sequence ;in: is the mean of the elements in the entropy sequence; Step S153: Normalize the mean difference sequence and set ; Step S154: determine whether all entropy value sequences have been processed, if yes, return to step S151, otherwise proceed to the next step; Step S155: Calculate the ordered distance between any two mean difference sequences respectively ; Among them: ordered distance refers to the comparison of element values while considering the position of elements in the sequence when comparing two sequences; specifically: the ordered distance is calculated using the following formula (3) ;in: and is the mean difference sequence corresponding to any two acquisition parameters; is the position of the element in the mean difference sequence; (3); Step S155: Calculate the ordered distance The mean of the ordered distance Specifically, the ordered distance mean is calculated using the following formula (4): ; (4); Step S155: When the ordered distance between the mean difference sequence corresponding to the acquisition parameter k and other mean difference sequences Greater than the mean of the ordered distance When the proportion exceeds the preset proportion, it is placed in the first category; otherwise, it is placed in the second category; the change classification includes the first category and the second category; Step S16: determining the active acquisition parameters and the inactive acquisition parameters according to the amplitude consistency and frequency consistency of the entropy value sequence; specifically: distinguishing the active acquisition parameters and the inactive acquisition parameters from the first category and the second category according to the disordered distance mean; Step S2: When the activity acquisition parameters are changed and other acquisition parameters remain unchanged, the tongue image information of the user is acquired; and the tongue image information is input into the intelligent analysis model to obtain the tongue image analysis result.
2. The tongue image medical information processing method according to claim 1, characterized in that: The entropy value is a feature value of the tongue image as a whole or locally.
3. The tongue image medical information processing method according to claim 2, characterized in that: The overall or local characteristic value of the tongue image includes one or a combination of the tongue image grayscale, brightness mean, gray pixel value at a specific position and / or gradient change value.
4. The tongue image medical information processing method according to claim 3, characterized in that: The preset ratio is 50~100%.
5. The tongue image medical information processing method according to claim 4, characterized in that: There are two activity collection parameters.
6. A tongue image medical information processing system, characterized in that: include: Intelligent analysis server and mobile terminal; the intelligent analysis server is used to implement the tongue image medical information processing method described in any one of claims 1-5 above.
7. The tongue image medical information processing system according to claim 6, characterized in that: The server is an artificial intelligence analysis server.
8. A tongue image medical information processing platform, characterized in that: It includes a processor, which is coupled to a memory, and the memory stores program instructions. When the program instructions stored in the memory are executed by the processor, the tongue image medical information processing method described in any one of claims 1 to 5 is implemented.
9. A computer-readable storage medium, characterized in that: It includes a program which, when executed on a computer, enables the computer to execute the tongue image medical information processing method as described in any one of claims 1 to 5.
10. An intelligent analysis server, characterized in that: The intelligent analysis server is configured to execute the tongue image medical information processing method as described in any one of claims 1-5.
Citation Information
Patent Citations
Analyzing events
CN102257487A
Medical device and method for generating modulated high frequency electrical stimulation pulses
CN113457010A
Tongue three-dimensional modeling method and system based on multi-angle image fusion
CN115578523A
Alzheimer risk monitoring training device and method
CN116672564A
Control method based on image acquisition device and control method and device of holder
CN116783568A
