Pet disease medical intelligent diagnosis system

Abstract

The invention discloses a pet disease medical intelligent diagnosis system, and relates to the technical field of data analysis, the pet disease medical intelligent diagnosis system comprises a processing module, a judgment module and a verification module, first data features are obtained, first analysis is carried out to obtain the skin disease type of a current pet, severity levels are divided, the disease type is verified and updated, and the final disease type is output. The image features and biochemical data are combined, the diagnosis accuracy is improved, historical data are used for analysis, the mode and features of the disease can be better recognized, the diagnosis efficiency is improved through automatic image and data processing, the disease types are verified according to the expert diagnosis result, the model is updated, and the diagnosis efficiency is improved. The system can continuously learn and optimize, the accuracy and reliability of diagnosis are improved, medical resources can be reasonably allocated through the intelligent diagnosis system, veterinarians can more concentrate on complex and emergency cases, and the quality and efficiency of overall medical services are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of data analysis, and in particular to an intelligent medical diagnosis system for pet diseases. Background Art

[0002] In recent years, AI-driven tools and machine learning algorithms have been able to analyze massive data sets, discover patterns and identify abnormalities that may be missed by the human eye. They have integrated clinical medical knowledge bases, multimodal medical data, and tens of millions of image samples to build an intelligent diagnosis and treatment system that covers consultation, diagnosis, medication, monitoring and verification. Through deep learning algorithms and big data analysis, AI technology can quickly and accurately analyze pet medical images, significantly improving diagnostic efficiency. AI systems can also provide verification analysis to help veterinarians predict disease outbreaks and track the progression of chronic diseases, enabling veterinarians to be more proactive in treatment and improve patient management.

[0003] At present, a pet disease probability diagnosis method and system are disclosed in a Chinese invention patent with publication number CN113724855A. The method calculates the weights of each part of the pet disease probability calculation by the pet's type, symptom characteristics, physiological characteristics, past experience and auxiliary examination results, and then calculates the probability of the pet suffering from the disease according to the pet disease diagnosis rule standard. However, the related art does not centrally organize and summarize the disease diagnosis rules based on various historical diagnosis data, and determine the most likely disease type through the summary results. It lacks the ability to coordinate and utilize data, and is not conducive to the accuracy of diagnosis. The disease type is not confirmed by the expert system, lacks the credibility of the disease diagnosis, and the database is not updated according to the verification results, which is not conducive to the continuous optimization of the system. Summary of the Invention

[0004] The technical problem solved by the present invention is that the relevant technology does not centrally organize and summarize the diagnostic rules of diseases based on various historical diagnostic data, and determine the most likely type of disease through the summary results. It lacks the ability to coordinate and utilize data, which is not conducive to the accuracy of diagnosis. The type of disease is not confirmed through the expert system, and the credibility of disease diagnosis is lacking. The database is not updated according to the verification results, which is not conducive to the continuous optimization of the system.

[0005] To solve the above technical problems, the present invention provides the following technical solutions: a pet disease medical intelligent diagnosis system, comprising a processing module, a judgment module and a verification module; The processing module obtains a first image feature set based on the pet's skin surface image and obtains a first data feature based on the pet's biochemical data; The judgment module performs a first analysis based on historical data regarding the first image feature set, and obtains the type of skin disease of the current pet based on the first analysis result, and obtains the severity level of the type of skin disease of the current pet based on the first data feature; The verification module verifies and updates the disease type according to the expert diagnosis results and outputs the final disease type.

[0006] As a preferred embodiment of the intelligent medical diagnosis system for pet diseases described in the present invention, the present invention combines image features and biochemical data, and the system can more comprehensively evaluate the health status of pets, thereby improving the accuracy of diagnosis. By using historical data for analysis, it can better identify the patterns and characteristics of diseases, further improving the accuracy of diagnosis. Automated image and data processing reduces the time and workload of manual operations, improves the efficiency of diagnosis, and can quickly generate preliminary diagnosis results to provide a reference for veterinarians, saving diagnosis time. According to the type and severity of skin diseases, the system can provide personalized treatment recommendations for pets, improve treatment effects, verify the type of disease and update the model based on the expert diagnosis results, so that the system can continuously learn and optimize, improve the accuracy and reliability of diagnosis, and through the intelligent diagnosis system, it can reasonably allocate medical resources, so that veterinarians can focus more on complex and urgent cases, and improve the quality and efficiency of overall medical services.

[0007] As a preferred embodiment of the intelligent medical diagnosis system for pet diseases of the present invention, wherein: pet biochemical data is obtained according to medical testing equipment, and the pet biochemical data includes blood routine data, liver function test data, kidney function test data, myocardial enzyme spectrum heart function test data, thyroid function test data, blood pressure data, blood sugar data and respiratory rate data; According to the pet standard biochemical data range, the first data is compared with the pet standard biochemical data range to obtain a first comparison result, and a first operation is performed according to the first comparison result. The first operation includes setting the first data feature of the current biochemical data to normal and jumping to the next biochemical data, and calculating the first data feature of the current biochemical data and jumping to the next biochemical data.

[0008] As a preferred embodiment of the intelligent medical diagnosis system for pet diseases of the present invention, the first comparison result includes: the first data being distributed within the pet standard biochemical data range, and the first data being not distributed within the pet standard biochemical data range; the first data being not distributed within the pet standard biochemical data range being represented by the first data being greater than an upper limit of the pet standard biochemical data range or the first data being less than a lower limit of the pet standard biochemical data range; When the first comparison result is that the first data is distributed within the range of pet standard biochemical data, the first operation is set to set the first data feature of the current biochemical data to normal and jump to the next biochemical data; When the first comparison result is that the first data is not distributed within the range of the pet's standard biochemical data, the first operation is set to calculate the first data feature of the current biochemical data and jump to the next biochemical data.

[0009] As a preferred embodiment of the intelligent medical diagnosis system for pet diseases of the present invention, the logic for calculating the first data feature of the current biochemical data includes: When the current biochemical data is less than the lower limit of the pet's standard biochemical data range, calculating a first difference between the lower limit of the pet's standard biochemical data range and the current biochemical data, and calculating a first ratio of the first difference to the lower limit of the pet's standard biochemical data range, and setting the first ratio as a first data feature of the current biochemical data; When the current biochemical data is greater than the upper limit of the pet's standard biochemical data range, the second difference between the current biochemical data and the upper limit of the pet's standard biochemical data range is calculated, and the second ratio of the second difference to the upper limit of the pet's standard biochemical data range is calculated, and the second ratio is set as the first data feature of the current biochemical data.

[0010] As a preferred embodiment of the intelligent medical diagnosis system for pet diseases of the present invention, the historical data on the first image feature set is represented by historical medical data on the type of skin disease corresponding to each second feature value in the first image feature set; The historical data includes the pet type, past medical history, diet status in the first time period before the visit, deworming status in the first time period before the visit, and stress status in the first time period before the visit, wherein the diet status in the first time period before the visit, deworming status in the first time period before the visit, and stress status in the first time period before the visit are orally described by the pet owner and classified by the physician; The dietary situation in the first period before the consultation included eating allergic foods, eating items that should not be eaten, sudden food changes, and normal diet; The deworming status in the first period before the visit included normal internal deworming and normal external deworming, no internal deworming and normal external deworming, normal internal deworming and no external deworming, and no internal deworming and no external deworming; Stressful situations during the first period before the visit included changes in living environment, the presence of outsiders or animals, being beaten or frightened by the owner, being frightened by external noises, and the absence of stressful situations; The severity levels include a first level, a second level, and a third level, and the severity of the disease represented by the first level, the second level, and the third level gradually increases.

[0011] As a preferred scheme of the pet disease medical intelligent diagnosis system, wherein: the logic of performing the first analysis according to the historical data of the first image feature set comprises: For any skin disease category in the first image feature set, the number of occurrences of each sub-category in the historical data of the skin disease category is counted, the first sum value of the number of occurrences of each sub-category is calculated, the third ratio value of the number of occurrences of each sub-category and the first sum value is calculated respectively, the third ratio value is set as the weight of the corresponding sub-category, and the probability expression of the skin disease category is constructed according to the weight of the sub-category. The probability expression of each skin disease category is obtained by traversing the skin disease categories in each first image feature set. The current data of the current pet is obtained, the current data of the current pet is brought into the probability expression of each skin disease category, the first probability corresponding to each skin disease category is obtained respectively, the first probability is sorted in descending order, and the skin disease category corresponding to the first probability with the largest value is set as the skin disease category of the current pet.

[0012] As a preferred scheme of the pet disease medical intelligent diagnosis system, wherein: the logic of obtaining the severity level of the skin disease category of the current pet according to the first data feature comprises: The first feature of any biochemical data is obtained. When the first data feature is normal, the state score of the corresponding biochemical data is set as the first score value. When the first data feature is the first ratio value or the second ratio value, the second value is set as the score threshold value, the first ratio value or the second ratio value is compared with the score threshold value respectively, when the first ratio value or the second ratio value is less than or equal to the second value, the state score of the corresponding biochemical data is set as the second score value, and when the first ratio value or the second ratio value is greater than the second value, the state score of the corresponding biochemical data is set as the third score value. Wherein, the first score value, the second score value and the third score value are in descending order, and the larger the score value is, the better the state is. The first average value of the state score of each biochemical data is calculated, the third value and the fourth value are set as the level boundary value, wherein the third value is greater than the fourth value, and the third value and the fourth value are both distributed in the interval with the first score value as the upper limit and the third score value as the lower limit, and the severity level is divided according to the first average value.

[0013] As a preferred scheme of the pet disease medical intelligent diagnosis system, wherein: when the first average value is greater than or equal to the third value, the severity level is set as the first level, when the first average value is less than the third value and greater than or equal to the fourth value, the severity level is set as the second level, and when the first average value is less than the fourth value, the severity level is set as the fourth level.

[0014] As a preferred embodiment of the intelligent medical diagnosis system for pet diseases of the present invention, the expert system is called up, and the current data and skin surface image of the current pet are input into the expert system to obtain a diagnosis result, wherein the diagnosis result represents the type of disease of the pet diagnosed by the expert; The diagnosis result is compared with the current disease type. When the diagnosis result is the same as the current disease type, the current disease type is set as the final disease type and output to the pet owner's feedback terminal. When the diagnosis result is different from the current disease type, the diagnosis result is set as the final disease type and output to the pet owner's feedback terminal. At the same time, the historical data is updated. The update logic includes: The current data corresponding to the final disease type is added to the historical data and saved.

[0015] The beneficial effects of the present invention are as follows: by combining image features and biochemical data, the system can more comprehensively assess the health status of pets, thereby improving the accuracy of diagnosis; by using historical data for analysis, it can better identify the patterns and characteristics of diseases, further improving the accuracy of diagnosis; automated image and data processing reduces the time and workload of manual operations, improves the efficiency of diagnosis, and can quickly generate preliminary diagnosis results to provide a reference for veterinarians, saving diagnosis time; based on the type and severity of skin diseases, the system can provide pets with personalized treatment recommendations and improve treatment effects; based on the expert diagnosis results, the disease type is verified and the model is updated, so that the system can continuously learn and optimize, improving the accuracy and reliability of diagnosis; through the intelligent diagnostic system, medical resources can be reasonably allocated, so that veterinarians can focus more on complex and urgent cases, and improve the quality and efficiency of overall medical services. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 This is a basic flow chart of a pet disease medical intelligent diagnosis system provided by one embodiment of the present invention. DETAILED DESCRIPTION

[0017] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are described in detail below in conjunction with the drawings. It is obvious that the described embodiments are only part of the embodiments of the present invention, but not all of the embodiments.

[0018] Example, see Figure 1 , as one embodiment of the present invention, provides a pet disease medical intelligent diagnosis system, including a processing module, a judgment module and a verification module; The processing module obtains a first image feature set based on the pet's skin surface image and obtains a first data feature based on the pet's biochemical data; The judgment module performs a first analysis based on historical data regarding the first image feature set, and obtains the type of skin disease of the current pet based on the first analysis result, and obtains the severity level of the type of skin disease of the current pet based on the first data feature; The verification module verifies and updates the disease type based on the expert diagnosis results and outputs the final disease type.

[0019] By combining image features and biochemical data, the present invention enables the system to more comprehensively assess the health status of pets, thereby improving the accuracy of diagnosis. By using historical data for analysis, it can better identify the patterns and characteristics of diseases, further improving the accuracy of diagnosis. Automated image and data processing reduces the time and workload of manual operations, improves the efficiency of diagnosis, and can quickly generate preliminary diagnosis results to provide a reference for veterinarians, saving diagnosis time. Based on the type and severity of skin diseases, the system can provide pets with personalized treatment recommendations to improve treatment effects. Based on the expert diagnosis results, the disease type is verified and the model is updated, enabling the system to continuously learn and optimize, improving the accuracy and reliability of diagnosis. Through the intelligent diagnostic system, medical resources can be reasonably allocated, allowing veterinarians to focus more on complex and urgent cases, improving the quality and efficiency of overall medical services.

[0020] The processing module obtains a first image feature based on the pet's skin surface image and obtains a first data feature based on the pet's biochemical data, wherein the pet's skin surface includes an ear canal skin surface and a body skin surface; The processing module obtains a pet skin surface image captured by a medical imaging device, extracts a first feature value of the pet skin surface image based on machine vision, retrieves a pet skin disease database, and extracts a second feature value of an image of any skin disease type in the pet skin disease database, wherein the first feature value and the second feature value are expressed as shape feature values; Calculate the first similarity between the first feature quantity and the second feature quantity by using the cosine similarity formula, set the first value as the first similarity threshold, compare the first similarity with the first value, and when the first similarity is greater than or equal to the first value, retain the corresponding second feature quantity; when the first similarity is less than the first value, delete the corresponding second feature quantity, jump to the next second feature quantity, and repeat the process of comparing the first similarity with the first value for the next second feature quantity; The respective retained second feature quantities are set as the first image feature set of the current pet.

[0021] In specific implementation, through extraction of shape feature quantity and calculation of cosine similarity, the system can more accurately identify and match image features of skin diseases, thereby improving the accuracy of diagnosis. Automatic image processing and feature extraction reduce the time and workload of manual operation, improve the efficiency of diagnosis, and according to the type and severity level of skin diseases, the system can provide personalized treatment suggestions for pets, improve the treatment effect, and continuously update and optimize the feature set according to new diagnosis results, improve the accuracy and reliability of diagnosis. The intelligent diagnosis system for pet diseases can improve the accuracy and efficiency of diagnosis by combining image processing, data analysis and expert verification, provide personalized treatment suggestions for pets, and continuously learn and optimize to provide a powerful auxiliary tool for veterinarians and pet owners.

[0022] Obtaining pet biochemical data from a medical testing device, the pet biochemical data including blood routine data, liver function test data, kidney function test data, myocardial enzyme spectrum heart function test data, thyroid function test data, blood pressure data, blood glucose data, and respiratory rate data; Comparing the first data with the pet standard biochemical data range to obtain a first comparison result, and performing a first operation according to the first comparison result, the first operation including setting the first data feature of the current biochemical data to normal and jumping to the next biochemical data, and calculating the first data feature of the current biochemical data and jumping to the next biochemical data.

[0023] The first comparison result includes that the first data is distributed within the pet standard biochemical data range, and that the first data is not distributed within the pet standard biochemical data range, which means that the first data is greater than the upper limit of the pet standard biochemical data range or the first data is less than the lower limit of the pet standard biochemical data range; When the first comparison result is that the first data is distributed within the pet standard biochemical data range, the first operation is set to set the first data feature of the current biochemical data to normal and jump to the next biochemical data; When the first comparison result is that the first data is not distributed within the pet standard biochemical data range, the first operation is set to calculate the first data feature of the current biochemical data and jump to the next biochemical data.

[0024] In specific implementation, by comparing biochemical data with the standard range, the system can more accurately identify abnormal biochemical indicators, thereby improving the accuracy of diagnosis. It can continuously update and optimize the feature set based on new diagnostic results, improve the accuracy and reliability of diagnosis, and perform remote diagnosis through the Internet, so that pet owners can obtain preliminary diagnostic results at home, reducing the inconvenience of pet medical treatment. For example, among the currently tested blood routine data, liver function test data, kidney function test data, myocardial enzyme spectrum cardiac function test data, thyroid function test data, blood pressure data, blood sugar data and respiratory rate data, one or more categories have abnormal values. By retrieving and analyzing the skin disease type data of one or more categories with abnormal values ​​from the skin disease medical records in the hospital system, since many skin diseases are mainly confined to the skin and its appendages, such as hair follicles, sebaceous glands, etc., the range of lesions is relatively limited, and the impact on the function of organs throughout the body is relatively small, so the changes in biochemical indicators are not obvious. For example, canine atopic dermatitis is mainly manifested by symptoms such as itchy skin and rashes. The results of blood routine and blood biochemical index tests are usually within the normal range. It is not a systemic disease and will not cause systemic metabolic disorders or organ dysfunction, so the biochemical indicators do not change much. Fungal skin diseases mainly affect the local skin. Even if the condition is severe, there are generally no obvious abnormalities in biochemical indicators. The liver has a strong compensatory capacity. Even if it suffers a certain degree of damage, it can still maintain normal metabolic function, so that the relevant biochemical indicators remain within the normal range. For example, in some mild skin infections or inflammations, the liver can compensate by increasing anabolism and maintain the stability of indicators such as serum albumin. Many skin diseases are caused by non-infectious factors, such as allergies, genetics, endocrine disorders, etc. These factors mainly cause diseases by affecting the immune response or physiological functions of the skin, and have little effect on systemic biochemical metabolism. For example, pemphigus foliaceus is an autoimmune skin disease. Among its biochemical indicators, only alkaline phosphatase is elevated, and other indicators are normal. Even for skin diseases caused by infectious factors, such as mite infection, the infection range is usually limited to the local skin, and the systemic reaction caused by the infection is relatively mild, which has limited impact on biochemical indicators. There may be multiple categories of skin diseases to be retrieved. Then, the types of skin diseases are roughly screened according to the biochemical indicators to reduce the number of calculations of the probability of skin diseases in the subsequent probability model, thereby greatly improving the rapid responsiveness of the system. At the same time, the hospital's medical records are retrieved for comparison, which improves the scientific nature of the initial screening of skin diseases.

[0025] The logic for calculating the first data feature of the current biochemical data includes: When the current biochemical data is less than the lower limit of the pet standard biochemical data range, a first difference value between the lower limit of the pet standard biochemical data range and the current biochemical data is calculated, a first ratio value between the first difference value and the lower limit of the pet standard biochemical data range is calculated, and the first ratio value is set as a first data feature of the current biochemical data; When the current biochemical data is greater than the upper limit of the pet standard biochemical data range, a second difference value between the current biochemical data and the upper limit of the pet standard biochemical data range is calculated, a second ratio value between the second difference value and the upper limit of the pet standard biochemical data range is calculated, and the second ratio value is set as the first data feature of the current biochemical data.

[0026] In specific implementation, by calculating the difference value and the ratio value, the system can quantify the abnormality degree of the biochemical data, convert the abnormality into a specific numerical feature, facilitate subsequent analysis and diagnosis, and the optimized data feature can more accurately reflect the abnormality of the biochemical data, which helps the system to more accurately identify and judge the severity of the disease, thereby improving the accuracy of diagnosis. The abnormal data and the standard range are calculated by ratio, so that the abnormality degrees of different indicators are compared on the same scale, which facilitates the system to perform unified analysis and processing.

[0027] The historical data about the first image feature set is represented as historical treatment data of the skin disease category corresponding to each second feature quantity in the first image feature set; The historical data includes pet category, past medical history, diet in the first time period before treatment, deworming in the first time period before treatment, and stress in the first time period before treatment, wherein the diet in the first time period before treatment, the deworming in the first time period before treatment, and the stress in the first time period before treatment are described by the pet owner and classified by the doctor; The diet in the first time period before treatment includes eating allergenic food, eating items that should not be used, sudden food change, and normal diet; The deworming in the first time period before treatment includes normal internal deworming and normal external deworming, no internal deworming and normal external deworming, normal internal deworming and no external deworming, and no internal deworming and no external deworming; The stress in the first time period before treatment includes changing living environment, presence of external personnel or animals, being beaten or frightened by the owner, being frightened by external sound, and absence of stress; The severity levels include a first level, a second level, and a third level, and the severity of the disease represented by the first level, the second level, and the third level gradually increases.

[0028] In specific implementation, by comprehensively considering various historical data, the system can more comprehensively evaluate the health status of the pet, thereby improving the accuracy of diagnosis, the information of previous medical history and stress in the historical data helps to identify potential disease inducements, further improves the accuracy of diagnosis, the division of severity grade helps the veterinarian to develop a more reasonable treatment plan according to the severity of the disease, by recording and analyzing the diet, deworming and stress of the pet in detail, the system can help the veterinarian to better communicate with the pet owner, and enhance the trust and cooperation between doctors and patients.

[0029] The logic of performing the first analysis according to the historical data about the first image feature set comprises: Selecting a skin disease category in any first image feature set, counting the number of occurrences of each sub-category in the historical data of the skin disease category, calculating the first sum value of the number of occurrences of each sub-category, calculating the third ratio value of the number of occurrences of each sub-category and the first sum value respectively, setting the third ratio value as the weight of the corresponding sub-category, and constructing a probability expression of the skin disease category according to the weight of the sub-category; Traversing the skin disease categories in each first image feature set to obtain the probability expression of each skin disease category; Obtaining the current data of the current pet, bringing the current data of the current pet into the probability expression of each skin disease category to obtain the first probability corresponding to each skin disease category respectively, sorting the first probabilities in descending order, and setting the skin disease category corresponding to the first probability with the largest value as the skin disease category of the current pet.

[0030] The logic of obtaining the severity grade of the skin disease category of the current pet according to the first data feature comprises: Obtaining a first feature of any biochemical data; When the first data feature is normal, setting the state score of the corresponding biochemical data as a first score value; When the first data feature is the first ratio value or the second ratio value, setting a second value as a score threshold, comparing the first ratio value or the second ratio value with the score threshold respectively, when the first ratio value or the second ratio value is less than or equal to the second value, setting the state score of the corresponding biochemical data as a second score value, when the first ratio value or the second ratio value is greater than the second value, setting the state score of the corresponding biochemical data as a third score value; Wherein, the first score value, the second score value and the third score value are in descending order, and the larger the score value is, the better the state is; Calculating a first average value of the state scores of each biochemical data, setting the third value and the fourth value as grade boundary values, wherein the third value is greater than the fourth value, and the third value and the fourth value are both distributed in an interval with the first score value as the upper limit and the third score value as the lower limit, and dividing the severity grade according to the first average value.

[0031] When the first average value is greater than or equal to the third value, the severity level is set to the first level, when the first average value is less than the third value and greater than or equal to the fourth value, the severity level is set to the second level, and when the first average value is less than the fourth value, the severity level is set to the fourth level.

[0032] In specific implementation, by counting the occurrence times of each sub-category in historical data and calculating the weight, the system can more accurately evaluate the influence of each sub-category on the skin disease type, thereby improving the accuracy of diagnosis. The construction of the probability expression and the input of the current data enable the system to calculate the probability of the skin disease type according to the specific conditions of the current pet, further improving the accuracy of diagnosis.

[0033] The expert system is called, and the current data and the skin surface image of the current pet are input into the expert system to obtain a diagnosis result, which represents the disease type of the pet diagnosed by the expert; The diagnosis result is compared with the current disease type, when the diagnosis result is the same as the current disease type, the current disease type is set as the final disease type and output to the feedback end of the pet owner, when the diagnosis result is different from the current disease type, the diagnosis result is set as the final disease type and output to the feedback end of the pet owner, and the historical data is updated, the updating logic including: The current data corresponding to the final disease type is added to the historical data and saved.

[0034] In specific implementation, by calling the expert system, the system can utilize the knowledge and experience of experts for diagnosis, further improving the accuracy of diagnosis. The diagnosis result of the expert system as the final disease type ensures the reliability and authority of the diagnosis result. When the diagnosis result is different from the current disease type, the system updates the historical data, adds the new diagnosis result and related data to the historical data. This updating mechanism enables the system to continuously learn and optimize, improving the accuracy and reliability of future diagnosis.

[0035] The application can more comprehensively evaluate the health status of pets by combining image features and biochemical data, thereby improving the accuracy of diagnosis, analyzing historical data to better identify patterns and characteristics of diseases, further improving the accuracy of diagnosis, automated image and data processing reducing the time and workload of manual operation, improving the efficiency of diagnosis, and quickly generating preliminary diagnosis results to provide reference for veterinarians, saving diagnosis time, providing personalized treatment recommendations for pets according to the types and severity levels of skin diseases, improving treatment effect, verifying the disease types according to expert diagnosis results and updating the model to enable the system to continuously learn and optimize, improving the accuracy and reliability of diagnosis, and through the intelligent diagnosis system, medical resources can be reasonably allocated to enable veterinarians to focus more on complex and urgent cases, improving the quality and efficiency of overall medical services.

[0036] Those skilled in the art will appreciate that embodiments of the application can be provided as methods, systems or computer program products. Accordingly, the application can be embodied in the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the application can be embodied in the form of a computer program product embodied on one or more computer-usable storage media having computer-usable program code embodied thereon. The storage media can be any type of volatile or non-volatile storage device, or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read only memory (EEPROM), erasable programmable read only memory (EPROM), programmable read only memory (PROM), read only memory (ROM), magnetic storage, flash memory, magnetic disk or optical disk. These computer program instructions can also be stored in a computer-readable memory that can direct a computer or other programmable data processing device to work in a specific way, so that the instructions stored in the computer-readable memory produce a product including instruction means, which realize the processes Figure 1 The functions specified in one or more processes and / or blocks Figure 1 The functions specified in one or more processes and / or blocks

[0037] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention can be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, and all of these should be included in the scope of the claims of the present invention.

Claims

1. Intelligent medical diagnosis system for pet diseases, characterized by: It includes processing module, judgment module and verification module; The processing module obtains a first image feature set based on the pet's skin surface image and obtains a first data feature based on the pet's biochemical data; The judgment module performs a first analysis based on historical data regarding the first image feature set, and obtains the type of skin disease of the current pet based on the first analysis result, and obtains the severity level of the type of skin disease of the current pet based on the first data feature; The verification module verifies and updates the disease type according to the expert diagnosis results and outputs the final disease type.

2. The intelligent medical diagnosis system for pet diseases according to claim 1, characterized in that: The processing module obtains a first image feature based on a pet skin surface image and obtains a first data feature based on the pet biochemical data, wherein the pet skin surface includes an ear canal skin surface and a body skin surface; The processing module obtains a pet skin surface image captured by a medical imaging device, extracts a first feature value of the pet skin surface image based on machine vision, retrieves a pet skin disease database, and extracts a second feature value of an image of any skin disease type in the pet skin disease database, wherein the first feature value and the second feature value are represented as shape feature values; Calculate the first similarity between the first feature quantity and the second feature quantity by using the cosine similarity formula, set the first value as the first similarity threshold, compare the first similarity with the first value, and when the first similarity is greater than or equal to the first value, retain the corresponding second feature quantity; when the first similarity is less than the first value, delete the corresponding second feature quantity, jump to the next second feature quantity, and repeat the process of comparing the first similarity with the first value for the next second feature quantity; The respective retained second feature quantities are set as the first image feature set of the current pet.

3. The intelligent medical diagnosis system for pet diseases according to claim 1, characterized in that: Obtaining pet biochemical data using medical testing equipment, including blood routine data, liver function test data, kidney function test data, myocardial enzyme spectrum heart function test data, thyroid function test data, blood pressure data, blood sugar data, and respiratory rate data; According to the pet standard biochemical data range, the first data is compared with the pet standard biochemical data range to obtain a first comparison result, and a first operation is performed according to the first comparison result. The first operation includes setting the first data feature of the current biochemical data to normal and jumping to the next biochemical data, and calculating the first data feature of the current biochemical data and jumping to the next biochemical data.

4. The intelligent medical diagnosis system for pet diseases according to claim 3, characterized in that: The first comparison result includes the first data being distributed within the pet's standard biochemical data range, and the first data being not distributed within the pet's standard biochemical data range, wherein the first data being not distributed within the pet's standard biochemical data range is represented by the first data being greater than an upper limit of the pet's standard biochemical data range or the first data being less than a lower limit of the pet's standard biochemical data range; When the first comparison result is that the first data is distributed within the range of pet standard biochemical data, the first operation is set to set the first data feature of the current biochemical data to normal and jump to the next biochemical data; When the first comparison result is that the first data is not distributed within the range of the pet's standard biochemical data, the first operation is set to calculate the first data feature of the current biochemical data and jump to the next biochemical data.

5. The intelligent medical diagnosis system for pet diseases according to claim 4, characterized in that: The logic for calculating the first data feature of the current biochemical data includes: When the current biochemical data is less than the lower limit of the pet's standard biochemical data range, calculating a first difference between the lower limit of the pet's standard biochemical data range and the current biochemical data, and calculating a first ratio of the first difference to the lower limit of the pet's standard biochemical data range, and setting the first ratio as a first data feature of the current biochemical data; When the current biochemical data is greater than the upper limit of the pet's standard biochemical data range, the second difference between the current biochemical data and the upper limit of the pet's standard biochemical data range is calculated, and the second ratio of the second difference to the upper limit of the pet's standard biochemical data range is calculated, and the second ratio is set as the first data feature of the current biochemical data.

6. The intelligent medical diagnosis system for pet diseases according to claim 1, wherein: The historical data about the first image feature set is represented by historical medical data of skin disease types corresponding to each second feature value in the first image feature set; The historical data includes the pet type, past medical history, diet status in the first time period before the visit, deworming status in the first time period before the visit, and stress status in the first time period before the visit, wherein the diet status in the first time period before the visit, deworming status in the first time period before the visit, and stress status in the first time period before the visit are orally described by the pet owner and classified by the physician; The dietary situation in the first period before the consultation included eating allergic foods, eating items that should not be eaten, sudden food changes, and normal diet; The deworming status in the first period before the visit included normal internal deworming and normal external deworming, no internal deworming and normal external deworming, normal internal deworming and no external deworming, and no internal deworming and no external deworming; Stressful situations during the first period before the visit included changes in living environment, the presence of outsiders or animals, being beaten or frightened by the owner, being frightened by external noises, and the absence of stressful situations; The severity levels include a first level, a second level, and a third level, and the severity of the disease represented by the first level, the second level, and the third level gradually increases.

7. The intelligent medical diagnosis system for pet diseases according to claim 1, characterized in that: The logic for performing a first analysis based on historical data regarding the first set of image features includes: Selecting a skin disease category in any first image feature set, counting the number of occurrences of each subcategory in historical data of the skin disease category, calculating a first sum of the number of occurrences of each subcategory, respectively calculating a third ratio of the number of occurrences of each subcategory to the first sum, setting the third ratio as a weight of the corresponding subcategory, and constructing a probability expression for the skin disease category based on the subcategory weight; Traversing the types of skin diseases in each first image feature set to obtain a probability expression for each type of skin disease; Obtain the current data of the current pet, substitute the current data of the current pet into the probability expression of each skin disease type, obtain the first probability corresponding to each skin disease type, sort the first probabilities in descending order, and set the skin disease type corresponding to the first probability with the largest value as the skin disease type of the current pet.

8. The intelligent medical diagnosis system for pet diseases according to claim 5, characterized in that: The logic for obtaining the severity level of the current pet's skin disease type based on the first data feature includes: Obtaining a first feature of any type of biochemical data; When the first data feature is normal, setting the status score of the corresponding biochemical data to a first score; When the first data feature is a first ratio or a second ratio, the second value is set as a score threshold, and the first ratio or the second ratio is respectively compared with the score threshold; when the first ratio or the second ratio is less than or equal to the second value, the state score of the corresponding biochemical data is set to the second score; when the first ratio or the second ratio is greater than the second value, the state score of the corresponding biochemical data is set to a third score; Among them, the first score, the second score and the third score are in descending order, and the larger the score, the better the status; Calculate the first average value of the status score of each biochemical data, set the third score and the fourth score as the level boundary values, wherein the third value is greater than the fourth value, and the third value and the fourth value are both distributed in the interval with the first score as the upper limit and the third score as the lower limit, and divide the severity level according to the first average value.

9. The intelligent medical diagnosis system for pet diseases according to claim 8, characterized in that: When the first average value is greater than or equal to the third value, the severity level is set to the first level; when the first average value is less than the third value and greater than or equal to the fourth value, the severity level is set to the second level; when the first average value is less than the fourth value, the severity level is set to the fourth level.

10. The intelligent medical diagnosis system for pet diseases according to claim 1, characterized in that: Invoking an expert system, inputting the current data and skin surface image of the current pet into the expert system to obtain a diagnosis result, wherein the diagnosis result represents the type of disease of the pet diagnosed by the expert; The diagnosis result is compared with the current disease type. When the diagnosis result is the same as the current disease type, the current disease type is set as the final disease type and output to the pet owner's feedback terminal. When the diagnosis result is different from the current disease type, the diagnosis result is set as the final disease type and output to the pet owner's feedback terminal. At the same time, the historical data is updated. The update logic includes: The current data corresponding to the final disease type is added to the historical data and saved.

Images