Urine image AI automatic identification and cloud health monitoring platform and system

Abstract

The invention discloses a urine image AI automatic identification and cloud health monitoring platform and system, and relates to the technical field of health monitoring, and the method comprises the steps: obtaining current urine key substance data and other health data of a user, and uploading the data to a cloud database; analyzing according to the current urine key substance data and historical urine key substance data of the user; constructing a disease occurrence probability model; and judging the risk level of the disease according to the disease occurrence probability. The urine key substance data is extracted through the urine image AI automatic identification technology, the health condition is analyzed from multiple dimensions in combination with other health data of the user, the remarkable abnormal fluctuation of the urine key substance is identified by using a time sequence analysis method in combination with historical urine data instead of depending on single urinalysis data, and the accuracy of the urine key substance identification is improved. Therefore, the long-term trend of urine indexes can be monitored, the limitation that traditional urine detection can only provide instant detection data is avoided, and the accuracy of health monitoring is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of health monitoring, in particular to an AI automatic recognition of urine images, a cloud health monitoring platform and a system. Background Art

[0002] Urine is one of the important excretions of human metabolism, and its composition and properties can reflect the physiological state and health status of the human body. In the field of medical testing, urine analysis is widely used for disease screening, health monitoring and disease diagnosis. Traditional urine detection methods mainly rely on laboratory chemical analysis methods, such as dipstick testing, colorimetric analysis and spectroscopic analysis, etc. Although these methods can provide relatively accurate test results, they often require professional testing equipment and experimental personnel support, the testing process is relatively complex, and it is difficult to achieve real-time and continuous health monitoring.

[0003] In recent years, with the development of computer vision, artificial intelligence (AI) and cloud computing technologies, the automatic recognition and health monitoring technology based on urine images has gradually become a research hotspot. Through high-resolution image acquisition technology combined with computer vision technology, key feature information such as the color characteristics, turbidity, urine sugar, and urine protein of urine can be obtained under non-invasive conditions. Combining AI algorithms to analyze the changing trends of urine components can thus achieve remote assessment of the user's health status. In addition, the development of cloud computing technology has made it possible for remote storage, intelligent analysis and personalized health management of health data, improving the intelligence and convenience of urine health monitoring. However, most of the existing urine detections only provide instant test data and cannot perform dynamic change analysis by combining the user's historical urine data; and most of the existing systems only perform health assessments based on the test results, and fail to effectively combine other dimensions of the user's health information to construct a disease occurrence probability model, thus making it difficult to provide accurate disease risk predictions. Summary of the Invention

[0004] In view of the problems in the prior art that most urine detections only provide instant test data and cannot perform dynamic change analysis by combining the user's historical urine data when detecting urine components; and most of them only perform health assessments based on the test results, and fail to effectively combine other dimensions of the user's health information to construct a disease occurrence probability model, thus making it difficult to provide accurate disease risk predictions and other problems, the present invention is proposed.

[0005] Therefore, the problem to be solved by the present invention is how to provide a method that can construct a disease occurrence probability model and provide more accurate disease risk predictions.

[0006] To solve the above technical problems, the present invention provides the following technical solutions:

[0007] In a first aspect, an embodiment of the present invention provides a urine image AI automatic recognition and cloud health monitoring platform, which includes collecting user urine image information and performing preprocessing to obtain current urine key substance data, and at the same time collecting other health data of the user, uploading the current urine key substance data and other health data to the cloud database and sorting them according to the timestamp; analyzing the current urine key substance data and the user's historical urine key substance data, and based on the time series, identifying whether the current urine key substance data is a significant abnormal fluctuation; constructing a disease occurrence probability model based on the user's current urine key substance data and other health data; calculating the probability of a disease caused by the current significant abnormal fluctuation according to the disease occurrence probability model, and judging the risk level of the disease according to the probability of the disease occurrence.

[0008] As a preferred solution of the urine image AI automatic recognition and cloud health monitoring platform of the present invention, wherein: the urine key substance data includes color characteristics, turbidity, urine sugar, and urine protein, and the other health data includes weight, blood pressure, and BMI.

[0009] As a preferred solution of the urine image AI automatic recognition and cloud health monitoring platform of the present invention, wherein: analyzing the current urine key substance data, the calculation formula of the abnormal factor of the current urine key data is:

[0010]

[0011] wherein, X u is the abnormal degree of the current urine key substance, X is the value of the current detected urine key substance; u is the normal mean value of this urine index, statistically based on a large amount of healthy user data; σ is the standard deviation of this urine index, used to measure the data fluctuation range;

[0012] The specific situation of the abnormal factor of the current urine data is as follows:

[0013] wherein, if X u is positive, it means that this index is higher than the normal value, the larger the value, the more serious the deviation degree, and at this time if X u is positive and higher than the first threshold, it is identified as a significant abnormal fluctuation, if X u is positive and lower than the first threshold, it is identified as a non-significant abnormal fluctuation, and the output is "healthy";

[0014] If X u is negative, it means that this index is lower than the normal value, the smaller the value, the more serious the deviation degree, and at this time if X u is negative and less than the second threshold, it is identified as a significant abnormal fluctuation, X u is negative and greater than the second threshold, it is identified as a non-significant abnormal fluctuation, and the output is "healthy".

[0015] As a preferred solution of the urine image AI automatic recognition and cloud health monitoring platform described in the present invention, where: when the abnormal factor of the current key urine data is recognized as a significant abnormal fluctuation, the key urine substance data is analyzed, and the exponential weighted moving average method is used to calculate the change trend X of the urine index over a period of time in the past h , and the calculation formula is:

[0016] X h =αX t +(1 - α)X t-1

[0017] where, X t is the current urine test value; X t-1 is the urine value of the previous test; α is the smoothing coefficient, and the value range is 0 < α < 1, which is used to control the importance of recent data. The larger α is, the greater the influence of the most recent urine data;

[0018] The specific situation of the change trend of the urine index over a period of time in the past is as follows:

[0019] If the change trend X of the urine index over a period of time in the past h is greater than or equal to the first preset value, it indicates that the urine index has improved;

[0020] If the change trend X of the urine index over a period of time in the past h is less than or equal to the second preset value, it indicates that the urine index has deteriorated;

[0021] If the change trend X of the urine index over a period of time in the past h is between the first preset value and the second preset value, it indicates that the urine index tends to be stable.

[0022] As a preferred solution of the urine image AI automatic recognition and cloud health monitoring platform described in the present invention, where: the calculation formula for the disease risk of other health data is:

[0023]

[0024] where, X o represents the disease risk of other health data; H is the current health data; H norm is the standard mean value of this health index;

[0025] The specific situation of the disease risk of other health data is as follows:

[0026] If X o is greater than the first risk value and less than the second risk value, it indicates that this health data is abnormal;

[0027] If X oIf it is less than the first risk value and greater than the second risk value, it indicates that the health data is within the normal range.

[0028] As a preferred solution of the urine image AI automatic recognition and cloud health monitoring platform described in the present invention, wherein: integrating the current urine data anomaly factor, the change trend of urine indicators over a past period of time, and other health data influencing factors, a disease occurrence probability model is constructed using Logistic, and the calculation formula is:

[0029]

[0030] Among them, P(D) is the probability of the occurrence of disease D; A f is the age correction factor; w1 is the weight of the urine key substance anomaly factor; w2 is the weight of the change trend of urine indicators over a past period of time; w3 is the weight of other health data influencing factors; b is the bias term, used to control the baseline risk of the entire model;

[0031] The specific situation of the disease occurrence probability model is as follows:

[0032] If the probability P(D) of the occurrence of disease D is less than the first disease risk threshold, it is indicated as low risk;

[0033] If the probability P(D) of the occurrence of disease D is greater than or equal to the first disease risk threshold and less than or equal to the second disease risk threshold, it is indicated as medium risk;

[0034] If the probability P(D) of the occurrence of disease D is greater than the second disease risk threshold, it is indicated as high risk.

[0035] As a preferred solution of the urine image AI automatic recognition and cloud health monitoring platform described in the present invention, wherein: the calculation formula of the age correction factor is:

[0036]

[0037] Among them, A is the current user's age; A norm is the reference age; k is the age influence weight.

[0038] In a second aspect, to further solve the security problems existing in the identification of online public opinion data, the embodiments of the present invention provide an online public opinion data identification system based on a neural network, which includes: a preprocessing module for obtaining the current urine key substance data and other health data of the user, and uploading the current urine key substance data and other health data to the cloud database; an anomaly detection module for analyzing the current urine key substance data and the historical urine key substance data of the user to identify significant abnormal fluctuations in the current urine key substance data; a model construction module for constructing a disease occurrence probability model based on the current urine key substance data and other health data of the user; and a risk prediction module for judging the risk level of disease occurrence according to the probability of disease occurrence.

[0039] In a third aspect, the embodiments of the present invention provide a computer device, including a memory and a processor, where the memory stores a computer program, and: when the computer program is executed by the processor, any step of the urine image AI automatic recognition and cloud health monitoring platform as described in the first aspect of the present invention is implemented.

[0040] In a fourth aspect, the embodiments of the present invention provide a computer-readable storage medium, on which a computer program is stored, and: when the computer program is executed by the processor, any step of the urine image AI automatic recognition and cloud health monitoring platform as described in the first aspect of the present invention is implemented.

[0041] The beneficial effects of the present invention are as follows:

[0042] 1. The present invention extracts urine key substance data through urine image AI automatic recognition technology, combines other health data of the user, analyzes the health status from multiple dimensions, and also combines historical urine data, uses time series analysis methods to identify significant abnormal fluctuations in urine key substances, rather than relying solely on single urine test data, so as to be able to monitor the long-term trend of urine indicators, avoid the limitation that traditional urine tests can only provide immediate test data, and improve the accuracy of health monitoring;

[0043] 2. Different from the existing urine test methods that directly give abnormal indicators, the present invention uses models such as Logistic regression, combines urine data, health data, age correction factors, etc. to comprehensively calculate the probability of disease occurrence, and divides the risk level, providing a more scientific basis for health assessment for users;

[0044] 3. The present invention uses the exponentially weighted moving average method to calculate the change trend of urine indicators, can more accurately identify the trend of improvement or deterioration of the health status, reduce the misjudgment caused by single abnormal data, and improve the reliability of disease prediction;

[0045] 4. The present invention uploads urine data and health data to the cloud database and sorts them according to the timestamp, enabling users to track their own health conditions over a long period of time and even share the data with medical institutions, providing technical support for personalized health management and telemedicine. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] To more clearly illustrate the technical solutions of the embodiments of the present invention, the following will briefly introduce the drawings required for the description of the embodiments. Obviously, the following drawings are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can be obtained based on these drawings. Among them:

[0047] Figure 1 It is a flowchart for implementing the present invention in Embodiment 1. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0048] To make the above objects, features, and advantages of the present invention more obvious and understandable, the following will provide a detailed description of the specific embodiments of the present invention with reference to the drawings in the specification.

[0049] In the following description, many specific details are set forth to fully understand the present invention. However, the present invention can also be implemented in other ways different from those described herein. Those skilled in the art can make similar generalizations without departing from the connotation of the present invention. Therefore, the present invention is not limited by the specific embodiments disclosed below.

[0050] Secondly, the so-called "one embodiment" or "embodiment" herein refers to a specific feature, structure, or characteristic that can be included in at least one implementation manner of the present invention. The "in one embodiment" that appears in different places in this specification does not necessarily refer to the same embodiment, nor is it a separate or selectively exclusive embodiment from other embodiments.

[0051] Embodiment 1

[0052] Refer to Figure 1 , which is the first embodiment of the present invention. This embodiment provides a urine image AI automatic recognition and cloud health monitoring platform, including the following steps:

[0053] S1: Collect the urine image information of the user through the urine test device and perform preprocessing to obtain the current urine key substance data. At the same time, collect other health data of the user, upload the current urine key substance data and other health data to the cloud database and sort them according to the timestamp.

[0054] Among them, the key urine substance data include color characteristics, turbidity, urine sugar, and urine protein, and the other health data include weight, blood pressure, and BMI; the preprocessing includes image quality enhancement (including color standardization and denoising), color feature extraction, turbidity analysis, and key substance identification.

[0055] S2: Analyze based on the current key urine substance data and the user's historical key urine substance data. Based on the time series, identify whether the current key urine substance data shows significant abnormal fluctuations.

[0056] Specifically, the calculation formula for the abnormal factor of the current key urine data is:

[0057]

[0058] Among them, X u is the degree of abnormality of the current key urine substance. X is the value of the currently detected key urine substance; u is the normal mean of this urine index, statistically based on a large amount of healthy user data; σ is the standard deviation of this urine index, used to measure the data fluctuation range;

[0059] Specifically, substitute the current key urine data including color characteristics, turbidity, urine sugar, urine protein, etc. into the calculation formula of the abnormal factor to identify whether there are significant abnormal fluctuations;

[0060] The specific situation of the abnormal factor of the current urine data is as follows:

[0061] If X u is positive, it means that this index is higher than the normal value. The larger the value, the more serious the deviation. And at this time, if X u is positive and higher than the first threshold, it is identified as a significant abnormal fluctuation. If X u is positive and lower than the first threshold, it is identified as a non-significant abnormal fluctuation, and the output is "healthy";

[0062] If X u is negative, it means that this index is lower than the normal value. The smaller the value, the more serious the deviation. And at this time, if X u is negative and less than the second threshold, it is identified as a significant abnormal fluctuation. If X u is negative and greater than the second threshold, it is identified as a non-significant abnormal fluctuation, and the output is "healthy".

[0063] S3: Build a disease occurrence probability model based on the user's current key urine substance data and other health data.

[0064] S31. When the abnormal factor of the current key urine data is identified as a significant abnormal fluctuation, analyze the abnormal key urine substance data, and use the exponentially weighted moving average method to calculate the change trend X of the urine index over a past period of timeh , the calculation formula is:

[0065] X h = αX t + (1 - α)X t-1

[0066] Wherein, X t is the current urine test value; X t-1 is the urine value of the previous test; α is the smoothing coefficient, and the value range is 0 < α < 1, which is used to control the importance of recent data. The larger α is, the greater the influence of the most recent urine data;

[0067] The specific situation of the change trend of urine indicators in the past period of time is as follows:

[0068] If the change trend X h of urine indicators in the past period of time is greater than or equal to the first preset value, it indicates that the urine indicators have improved;

[0069] If the change trend X h of urine indicators in the past period of time is less than or equal to the second preset value, it indicates that the urine indicators have deteriorated

[0070] If the change trend X h of urine indicators in the past period of time is between the first preset value and the second preset value, it indicates that the urine indicators tend to be stable;

[0071] S32, the calculation formula for the disease risk of other health data is:

[0072]

[0073] Wherein, X o represents the disease risk of other health data; H is the current health data; H norm is the standard mean value of this health indicator;

[0074] Specifically, for the key urine substance data with significant abnormalities, match the corresponding other health data. For example, if the urine sugar in the current key urine substance data shows significant abnormalities, it may be related to diabetes, and diabetes usually causes weight loss. Therefore, when the urine sugar in the current key urine substance data shows significant abnormalities, the corresponding other health data is weight.

[0075] The specific situation of the disease risk of other health data is as follows:

[0076] If X o is greater than the first risk value and less than the second risk value, it indicates that this health data is abnormal;

[0077] If X oIf it is less than the first risk value and greater than the second risk value, it indicates that the health data is within the normal range;

[0078] S32, the calculation formula for the age correction factor is:

[0079]

[0080] where A is the current user's age; A norm is the reference age, which can be set according to historical data and conforms to the law that the disease risk increases with age; k is the age influence weight, usually taking 0.1 - 0.5, indicating the adjustment ratio of the disease probability for every 10% increase in age;

[0081] S33, considering the current urine data abnormality factor, the change trend of urine indicators over a past period of time, and other health data influence factors, use Logistic to construct a disease occurrence probability model. The calculation formula is:

[0082]

[0083] where P(D) is the probability of the occurrence of disease D; A f is the age correction factor; w1 is the weight of the urine key substance abnormality factor; w2 is the weight of the change trend of urine indicators over a past period of time; w3 is the weight of other health data influence factors; b is the bias term, used to control the baseline risk of the entire model.

[0084] S4: Estimate the probability of the current significant abnormal fluctuation triggering a disease according to the disease occurrence probability model, and judge the risk level of the occurrence of the disease based on the probability of the occurrence of the disease.

[0085] The specific situation of the disease occurrence probability model is as follows:

[0086] If the probability P(D) of the occurrence of disease D is less than the first disease risk threshold, it indicates a low risk and continuous health monitoring is required;

[0087] If the probability P(D) of the occurrence of disease D is greater than or equal to the first disease risk threshold and less than or equal to the second disease risk threshold, it indicates a medium risk and attention is required. It is recommended to adjust the diet or lifestyle;

[0088] If the probability P(D) of the occurrence of disease D is greater than the second disease risk threshold, it indicates a high risk and further examinations are required.

[0089] This embodiment also provides an opinion data recognition system based on a neural network, including: a preprocessing module, configured to obtain the current urine key substance data and other health data of a user, and upload the current urine key substance data and other health data to a cloud database; an anomaly detection module, configured to analyze the current urine key substance data and the historical urine key substance data of the user to identify significant abnormal fluctuations in the current urine key substance data; a model construction module, configured to construct a disease occurrence probability model based on the current urine key substance data and other health data of the user; and a risk prediction module, configured to determine the risk level of disease occurrence according to the probability of disease occurrence.

[0090] This embodiment also provides a computer device applicable to the case of an AI automatic recognition of urine images and a cloud health monitoring platform, including: a memory and a processor; the memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions to implement the AI automatic recognition of urine images and the cloud health monitoring platform as proposed in the above embodiment.

[0091] This computer device can be a terminal, and this computer device includes a processor, a memory, a communication interface, a display screen, and an input device connected through a system bus. Among them, the processor of this computer device is used to provide computing and control capabilities. The memory of this computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The communication interface of this computer device is used to communicate with an external terminal in a wired or wireless manner, and the wireless manner can be implemented through WIFI, a carrier network, NFC (Near Field Communication), or other technologies. The display screen of this computer device can be a liquid crystal display screen or an electronic ink display screen, and the input device of this computer device can be a touch layer covering the display screen, or a button, a trackball, or a touchpad provided on the housing of the computer device, or an external keyboard, a touchpad, or a mouse, etc.

[0092] This embodiment also provides a storage medium, on which a computer program is stored. When the program is executed by a processor, it implements the urine image AI automatic recognition and cloud health monitoring platform proposed in the above embodiment; the storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (Static Random Access Memory, abbreviated as SRAM), electrically erasable programmable read-only memory (Electrically Erasable Programmable Read-Only Memory, abbreviated as EEPROM), erasable programmable read-only memory (Erasable Programmable Read-Only Memory, abbreviated as EPROM), programmable read-only memory (Programmable Red-Only Memory, abbreviated as PROM), read-only memory (Read-Only Memory, abbreviated as ROM), magnetic memory, flash memory, magnetic disk or optical disc.

[0093] In summary, the present invention extracts urine key substance data through urine image AI automatic recognition technology, combines other health data of users, analyzes the health status from multiple dimensions, and also combines historical urine data. Using time series analysis method to identify significant abnormal fluctuations of urine key substances, rather than relying solely on single urine test data, so as to be able to monitor the long-term trend of urine indicators, avoid the limitation that traditional urine tests can only provide instant test data, and improve the accuracy of health monitoring.

[0094] Embodiment 2

[0095] Referring to Tables 1 to 2, this is the second embodiment of the present invention. The difference between this embodiment and the first embodiment is that in order to verify its beneficial effects, the operation data and related descriptions of the present invention in the actual environment are provided.

[0096] This embodiment collects user urine image information and other health data of the user and performs corresponding processing, and then analyzes based on the current urine key substance data and the user's historical urine key substance data to identify whether the current urine key substance data is a significant abnormal fluctuation. It is assumed that the first threshold is 0.5 and the second threshold is -0.5. The recognition result distribution table is shown in Table 1.

[0097] Table 1 Identifying Whether the Current Urine Key Substance Data is a Significant Abnormal Fluctuation

[0098]

[0099]

[0100] As can be seen from the above table, the present invention analyzes the current urine key substance data in combination with the historical urine key substance data to identify abnormal data points that do not conform to the expected fluctuation law.

[0101] As shown in Table 2, it is the disease D occurrence probability table. According to the disease occurrence probability model, the probability of the current significant abnormal fluctuation triggering disease D is calculated, and the risk level of the occurrence of the disease is judged based on the probability of the occurrence of the disease, thereby intuitively explaining the risk of the occurrence of the disease caused by the current significant abnormal fluctuation. Assume that the first disease risk threshold is 0.3 and the second disease risk threshold is 0.6.

[0102] Table 2 Disease D occurrence probability table

[0103] Experimental subject Probability of disease D occurrence Judgment result 1 0.75 High risk 2 0.23 Low risk 3 0.58 Medium risk ... ... ... n 0.87 High risk

[0104] As can be seen from the above table, the present invention calculates the probability of the current significant abnormal fluctuation triggering a disease through the disease occurrence probability model, and judges the risk level of the occurrence of the disease based on the probability of the occurrence of the disease, and then formulates corresponding treatment measures to improve the accuracy of disease prediction.

[0105] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit them. Although the present invention has been described in detail with reference to the preferred embodiments, those of ordinary skill in the art should understand that the technical solutions of the present invention can be modified or equivalently replaced without departing from the spirit and scope of the technical solutions of the present invention, and they should all be covered within the scope of the claims of the present invention.

Claims

1. AI automatic urine image recognition and cloud health monitoring platform, characterized by: include: S1, collect the user's urine image information and pre-process it to obtain the current urine key substance data, and at the same time collect the user's other health data, upload the current urine key substance data and other health data to the cloud database and sort them according to timestamps; S2, analyzing the current urine key substance data and the user's historical urine key substance data, and identifying whether the current urine key substance data is a significant abnormal fluctuation based on the time series; S3, builds a disease occurrence probability model based on the user's current urine key substance data and other health data; S4, calculate the probability of the current significant abnormal fluctuation causing the disease based on the disease occurrence probability model, and determine the risk level of the disease based on the probability of the disease occurrence.

2. The AI ​​automatic urine image recognition and cloud health monitoring platform according to claim 1, characterized in that: The key urine substance data include color characteristics, turbidity, urine sugar, and urine protein, and the other health data include weight, blood pressure, and BMI.

3. The AI ​​automatic urine image recognition and cloud health monitoring platform according to claim 2, characterized in that: The current urine key substance data is analyzed, and the calculation formula for the abnormal factor of the current urine key data is: Among them, X u is the abnormal degree of the current key urine substance, X is the value of the key urine substance currently detected; u is the normal mean value of the urine index; σ is the standard deviation of the urine index; The specific conditions of the abnormal factors of the current urine data are as follows: If X u is positive, indicating that the indicator is higher than the normal value, and at this time if X u If X is positive and higher than the first threshold, it is identified as a significant abnormal fluctuation. u If it is positive and lower than the first threshold, it is identified as a non-significant abnormal fluctuation and the output is "healthy"; If X u If it is negative, it means that the indicator is lower than the normal value, and if X u If it is negative and less than the second threshold, it is identified as a significant abnormal fluctuation, X u If it is negative and greater than the second threshold, it is identified as a non-significant abnormal fluctuation and the output is "healthy".

4. The AI ​​automatic urine image recognition and cloud health monitoring platform according to claim 3, characterized in that: When the abnormal factors of the current urine key data are identified as significant abnormal fluctuations, the urine key substance data are analyzed and the exponentially weighted moving average method is used to calculate the change trend of urine indicators over the past period of time X h , the calculation formula is: X h =αX t +(1-a)X t-1 Among them, X t is the current urine test value; X t-1 is the urine value of the last test; α is the smoothing coefficient, with a value range of 0<α<1, which is used to control the importance of recent data. The larger α is, the greater the impact of the recent urine data; the specific situation of the change trend of urine indicators in the past period of time is as follows: If the change trend of urine index in the past period of time is X h If it is greater than or equal to the first preset value, it indicates that the urine index has improved; If the change trend of urine index in the past period of time is X h If it is less than or equal to the second preset value, it indicates that the urine index has deteriorated; If the change trend of urine index in the past period of time is X h If the urine index is between the first preset value and the second preset value, it indicates that the urine index tends to be stable.

5. The AI ​​automatic urine image recognition and cloud health monitoring platform according to claim 4, characterized in that: The formula for calculating disease risk of other health data is: Among them, X o Represents other health data disease risk; H is current health data; H norm is the standard mean value of the health indicator; The details of other health data disease risks are as follows: If X o If it is greater than the first risk value and less than the second risk value, it indicates that the health data is abnormal; If X o If it is less than the first risk value and greater than the second risk value, it indicates that the health data is within the normal range.

6. The AI ​​automatic urine image recognition and cloud health monitoring platform according to claim 5, characterized in that: Based on the abnormal factors of current urine data, the changing trends of urine indicators in the past period of time, and other factors affecting health data, a disease occurrence probability model is constructed using Logistic, and the calculation formula is: Where P(D) is the probability of disease D occurring; A f is the age correction factor; w1 is the weight of the abnormal factor of key urine substances; w2 is the weight of the change trend of urine indicators in the past period of time; w3 is the weight of other health data influencing factors; b is the bias term, which is used to control the baseline risk of the entire model; The details of the disease occurrence probability model are as follows: If the probability P(D) of disease D occurring is less than the first disease risk threshold, it indicates low risk; If the probability P(D) of disease D occurring is greater than or equal to the first disease risk threshold and less than or equal to the second disease risk threshold, it indicates a medium risk; If the probability P(D) of disease D occurring is greater than the second disease risk threshold, it indicates a high risk.

7. The AI ​​automatic urine image recognition and cloud health monitoring platform according to claim 6, characterized in that: The calculation formula of the age correction factor is: Among them, A is the current user's age; A norm is the reference age; k is the age impact weight.

8. A urine image AI automatic recognition and cloud health monitoring system, based on the urine image AI automatic recognition and cloud health monitoring platform according to any one of claims 1 to 7, characterized in that: include, A preprocessing module is used to obtain the user's current urine key substance data and other health data, and upload the current urine key substance data and other health data to the cloud database; The anomaly detection module analyzes the current urine key substance data and the user's historical urine key substance data to identify significant abnormal fluctuations in the current urine key substance data; Model building module, which builds a disease occurrence probability model based on the user's current urine key substance data and other health data; The risk prediction module determines the risk level of the disease based on the probability of the disease occurring.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the AI ​​automatic recognition of urine images and the cloud health monitoring platform as described in any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of AI automatic recognition of urine images and cloud health monitoring platform described in any one of claims 1 to 7 are implemented.

Images