Method and device for osteoarthritis risk warning based on dietary caffeine intake, equipment and storage medium

Abstract

The present application relates to a method, device, equipment and storage medium for early warning of osteoarthritis risk based on dietary caffeine intake, the method comprising: obtaining an original paired data set, analyzing a case knee joint original field set and a control knee joint original field set, generating a case knee joint record and a control knee joint record corresponding to each matching group, the case knee joint record and the control knee joint record comprising: matching group number, subject number, knee joint side, classification label, dietary caffeine intake and model covariates; training a pre-constructed risk association model based on the case knee joint record and the control knee joint record to obtain a trained risk association model; inputting the to-be-evaluated feature data containing the dietary caffeine intake into the trained risk association model to obtain a comprehensive risk association index representing the correlation between the to-be-evaluated feature data and the knee joint degeneration progression. The present application realizes early warning of the to-be-evaluated feature data of an individual.

Description

Technical Field

[0001] This invention relates to the field of smart medical technology, and in particular to a method, device, equipment, and storage medium for early warning of osteoarthritis risk based on dietary caffeine intake. Background Technology

[0002] Osteoarthritis (OA), especially knee osteoarthritis (KOA), is a common chronic disease that leads to lower limb dysfunction in middle-aged and elderly people. Radiographic diagnosis of knee osteoarthritis (ROA) is usually based on the Kellgren-Lawrence (KL) classification.

[0003] However, conventional osteoarthritis research results are usually presented as overall regression coefficients or odds ratios, without providing individual-specific warnings.

[0004] Therefore, existing technologies have shortcomings and need to be improved and developed. Summary of the Invention

[0005] This application provides a method, device, equipment, and storage medium for early warning of osteoarthritis risk based on dietary caffeine intake, in order to address the technical problem that osteoarthritis research results are usually presented as overall regression coefficients or odds ratios, without providing individual-specific early warnings.

[0006] To achieve the above objectives, this application adopts the following technical solution: A method for predicting osteoarthritis risk based on dietary caffeine intake, the method comprising: Obtain the original paired dataset, which includes multiple matching groups. Each matching group includes: a matching group number, a set of original fields for case knee joints, and a set of original fields for control knee joints. The original field set of the case knee joint and the original field set of the control knee joint are parsed to generate case knee joint records and control knee joint records corresponding to each matching group. The case knee joint records and control knee joint records include: matching group number, subject number, knee side, classification label, dietary caffeine intake, and model covariates. The pre-constructed risk association model is trained based on the knee joint records of the cases and the control knee joint records to obtain the trained risk association model. The data of features to be evaluated, including dietary caffeine intake, are input into a trained risk association model to obtain a comprehensive risk association index that characterizes the correlation between the data of features to be evaluated and the progression of knee joint degeneration.

[0007] In this embodiment of the application, the steps for constructing the risk association model include: Using dietary caffeine intake as the core exposure variable in the risk association model and model covariates as covariates in the risk association model, a risk association model is constructed based on a generalized estimation equation using a binomial distribution and a logit link function.

[0008] In the embodiments of this application, during the training of the risk association model, the subject number is used as the cluster identifier, and robust covariance estimation is used to process the correlation between the two knee records of the same subject. The classification label is a binary code, wherein the classification label in the case knee joint record is configured as the first code value, and the classification label in the control knee joint record is configured as the second code value.

[0009] In this embodiment of the application, the feature data to be evaluated, including dietary caffeine intake, is input into a trained risk association model to obtain a comprehensive risk association index characterizing the correlation between the feature data to be evaluated and the progression of knee joint degeneration, including: Receive feature data to be evaluated, including dietary caffeine intake; The numerical input range and category code in the feature data to be evaluated are verified; If the verification passes, a comprehensive risk association index representing the correlation between the feature data to be evaluated and the progression of knee joint degeneration is obtained based on the trained risk association model.

[0010] In this embodiment of the application, after inputting the feature data to be evaluated, including dietary caffeine intake, into a trained risk association model to obtain a comprehensive risk association index characterizing the correlation between the feature data to be evaluated and the progression of knee joint degeneration, the method further includes: Several levels of boundary thresholds are preset, and graded early warning prompts are output based on the numerical relationship between comprehensive risk correlation indicators and boundary thresholds.

[0011] In this embodiment, several levels of boundary thresholds are pre-set, and graded early warning prompts are output based on the numerical relationship between the comprehensive risk correlation index and the boundary thresholds, including: Two levels of thresholds are preset, and three types of graded early warning prompts are output based on the numerical relationship between the comprehensive risk correlation index and the threshold. When the comprehensive risk correlation index is less than or equal to the first threshold of the two-level threshold, a risk warning that is lower than the research reference level will be output. When the comprehensive risk correlation index is between the two-level boundary threshold, a risk warning close to the research reference level is output. When the comprehensive risk correlation index is greater than or equal to the second boundary threshold in the two-level boundary threshold, a risk warning that is higher than the research reference level will be output. The two-level boundary thresholds can be adaptively adjusted.

[0012] In this embodiment of the application, the model covariates include: age, sex, body mass index, race, depression score, physical activity score, comorbidity score, radiographic status of the contralateral knee joint, history of knee injury, and history of knee surgery. The data to be evaluated also include: age, sex, body mass index, race, depression score, physical activity score, comorbidity score, history of knee injury, and history of knee surgery.

[0013] This application also provides a device for early warning of osteoarthritis risk based on dietary caffeine intake, the device comprising: The acquisition module is used to acquire the original paired dataset, which includes multiple matching groups. Each matching group includes: a matching group number, a set of original fields for case knee joints, and a set of original fields for control knee joints. The generation module is used to parse the original field set of the case knee joint and the original field set of the control knee joint, and generate the case knee joint record and control knee joint record corresponding to each matching group. The case knee joint record and control knee joint record include: matching group number, subject number, knee side, classification label, dietary caffeine intake and model covariates. The training module is used to train a pre-built risk association model based on the case knee joint records and control knee joint records to obtain a trained risk association model. The inference module is used to input the feature data to be evaluated, including dietary caffeine intake, into the trained risk association model to obtain a comprehensive risk association index that characterizes the correlation between the feature data to be evaluated and the progression of knee joint degeneration.

[0014] This application also provides an apparatus comprising: a memory, a processor, and an osteoarthritis risk warning program based on dietary caffeine intake stored in the memory and executable on the processor, wherein the osteoarthritis risk warning program based on dietary caffeine intake, when executed by the processor, implements the steps of the osteoarthritis risk warning method based on dietary caffeine intake as described above.

[0015] This application also provides a computer-readable storage medium storing a computer program that can be executed to implement the steps of the osteoarthritis risk warning method based on dietary caffeine intake as described above.

[0016] The beneficial effects of this invention are as follows: The method of this embodiment obtains an original paired dataset, which includes multiple matching groups. Each matching group includes: a matching group number, a set of original fields for the case knee joint, and a set of original fields for the control knee joint. The method parses the original fields for the case knee joint and the control knee joint to generate a case knee joint record and a control knee joint record corresponding to each matching group. These records include: a matching group number, a subject number, knee side, a classification label, dietary caffeine intake, and model covariates. A pre-constructed risk association model is trained based on these records to obtain a trained risk association model. The unassessed feature data, including dietary caffeine intake, is input into the trained risk association model to obtain a comprehensive risk association index characterizing the correlation between the unassessed feature data and the progression of knee joint degeneration. This application trains a risk association model using case knee joint records and control knee joint records, thereby obtaining a comprehensive risk association index characterizing the correlation between the unassessed feature data and the progression of knee joint degeneration, achieving early warning for individual unassessed feature data. Attached Figure Description

[0017] Figure 1 This is a flowchart of a preferred embodiment of the osteoarthritis risk warning method based on dietary caffeine intake in this invention.

[0018] Figure 2 This is a functional principle block diagram of a preferred embodiment of the osteoarthritis risk warning device based on dietary caffeine intake in this invention.

[0019] Figure 3 This is a functional principle block diagram of a preferred embodiment of the device in this invention. Detailed Implementation

[0020] To make the objectives, technical solutions, and advantages of this invention clearer and more explicit, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.

[0021] Current clinical assessments mainly rely on X-rays, magnetic resonance imaging, and symptom scales, while related risk studies often use factors such as age, gender, BMI (Body Mass Index), history of knee injury, and imaging indicators.

[0022] The existing technology has the following drawbacks: First, the original case-control data is stored in a paired wide table format, with the variables for the case knee and the control knee set in the same row, making it inconvenient to directly perform knee joint-level analysis; there are also problems such as differences in field naming, missing values, and the same subject contributing records for both knees. Second, existing osteoarthritis assessments mostly focus on age, BMI, injury history, and imaging indicators, lacking a calculation process that incorporates baseline dietary caffeine intake and multiple covariates into the same association model and outputs the factor contribution. Third, conventional research results are usually presented as overall regression coefficients or odds ratios, which are not convenient for directly forming early warning information for individuals; traditional diagnosis is usually made after structural damage to the joint occurs, lacking early warning of the pre-disease window period.

[0023] This application first extracts dietary caffeine intake (i.e., daily dietary caffeine intake) and related covariates from case-control data, converting paired wide tables into single-row standardized data, and validating matching relationships, field completeness, and numerical validity. Then, using caffeine intake as the core variable, covariates such as age, gender, and BMI are incorporated into a generalized estimation equation model to obtain the association parameters between each variable and the onset of new-onset radiographic knee osteoarthritis. In the application phase, the system acquires the user's personal information, calculates the overall risk association index (OR) relative to the research reference group, and outputs warning prompts indicating levels below, close to, or above the research reference level. This invention realizes a complete technical process from research data processing and risk association model construction to website warning application.

[0024] The following description, with reference to the accompanying drawings, outlines a method, apparatus, device, and storage medium for predicting osteoarthritis risk based on dietary caffeine intake, according to embodiments of this application. To address the issue mentioned in the background that osteoarthritis research results are typically presented as overall regression coefficients or odds ratios, without directly providing individual-specific early warning information, this application provides a method for early warning of osteoarthritis risk based on dietary caffeine intake. This method involves obtaining an original paired dataset, which includes multiple matching groups. Each matching group includes: a matching group number, a set of original field sets for case knee joints, and a set of original field sets for control knee joints. The original field sets for case and control knee joints are parsed to generate corresponding case and control knee joint records for each matching group. These records include: a matching group number, a subject number, knee side, a classification label, dietary caffeine intake, and model covariates. A pre-constructed risk association model is trained based on these records to obtain a trained risk association model. The feature data to be evaluated, including dietary caffeine intake, is input into the trained risk association model to obtain a comprehensive risk association index characterizing the correlation between the feature data to be evaluated and the progression of knee joint degeneration. This application trains a risk association model using case knee joint records and control knee joint records, thereby obtaining a comprehensive risk association index that characterizes the correlation between the feature data to be assessed and the progression of knee joint degeneration, thus enabling early warning of the individual's feature data to be assessed.

[0025] Please see Figure 1 The osteoarthritis risk warning method based on dietary caffeine intake according to embodiments of the present invention includes the following steps: Step S100: Obtain the original paired dataset, which includes multiple matching groups. Each matching group includes: a matching group number, a set of original fields for the case knee joint, and a set of original fields for the control knee joint.

[0026] For example, dietary caffeine is a quantifiable dietary exposure factor. The Osteoarthritis Initiative (OAI) project’s OAI Core MRI New Osteoarthritis Cohort Study (POMA Incidence Study) provides case-control data that includes baseline dietary information, individual characteristics, and knee imaging follow-up results, which can be used as the original paired dataset in this application.

[0027] This application can obtain the original paired dataset based on the OAI-POMA cohort, or it can obtain the original paired dataset based on other sub-cohorts in the OAI database or other public databases containing diet and osteoarthritis follow-up data.

[0028] like Figure 1 As shown, the osteoarthritis risk warning method based on dietary caffeine intake further includes the following steps: Step S200: Parse the original field set of the case knee joint and the original field set of the control knee joint to generate the case knee joint record and control knee joint record corresponding to each matching group. The case knee joint record and control knee joint record include: matching group number, subject number, knee side, classification label, dietary caffeine intake and model covariates.

[0029] Specifically, based on the base variable name and the case / control group field identifiers, the system extracts the subject ID, knee side, dietary caffeine intake (i.e., daily dietary caffeine intake), and related covariates. For each matched group, one case knee joint record and one control knee joint record are generated, ensuring that each row represents only one knee joint. Each record includes at least the matched group ID, subject ID, knee side, case or control label, dietary caffeine intake, and model covariates.

[0030] This application further verifies whether each matching group contains exactly one case knee and one control knee, and records or excludes fields with missing values, values ​​that cannot be converted, and abnormal category codes.

[0031] In this embodiment of the application, the model covariates include: age, sex, body mass index, race, depression score, physical activity score, comorbidity score, radiographic status of the contralateral knee joint, history of knee injury, and history of knee surgery.

[0032] Specifically, this application uses age, sex, body mass index, race, depression score, physical activity score, comorbidity score, contralateral knee imaging status, history of knee injury, and history of knee surgery as model covariates. That is, the dataset used in this application includes attributes such as dietary caffeine intake, age, sex, body mass index (BMI), race, depression score (CES-D score), physical activity score (PASE score), comorbidity score, contralateral knee imaging status, and history of knee injury and surgery.

[0033] This application automatically extracts, cleans, and validates caffeine and covariate data from case and control knees, converting paired wide tables into standardized data with one row per knee, while retaining the matching number and subject number. The matching number ensures the correspondence between cases and controls, and the subject number identifies potential contributions from the same subject to both knee records. Subjects are divided into a case group (KLG grade progressing from 0-1 to ≥2 during follow-up) and a control group (KLG grade remaining <2 during follow-up), i.e., a binary label.

[0034] like Figure 1 As shown, the osteoarthritis risk warning method based on dietary caffeine intake further includes the following steps: Step S300: Train the pre-constructed risk association model based on the case knee joint records and control knee joint records to obtain the trained risk association model.

[0035] In this embodiment of the application, the steps for constructing the risk association model include: using dietary caffeine intake as the core exposure variable of the risk association model, using model covariates as covariates of the risk association model, and constructing the risk association model based on the generalized estimation equation of binomial distribution and logit connection function.

[0036] For example, the risk association model is as follows: ; in, Let J represent the outcome variable for the j-th knee joint of the i-th subject, and its characteristic variables are: to ; This indicates dietary caffeine intake (e.g., daily dietary caffeine intake). to The data included age, sex, BMI, race, CES-D depression score, PASE physical activity score, comorbidity score, history of knee injury, and history of knee surgery. For constant terms, to The regression coefficients are for data on dietary caffeine intake, age, sex, BMI, race, CES-D depression score, PASE physical activity score, comorbidity score, history of knee injury, and history of knee surgery.

[0037] This application uses a generalized estimation equation based on a binomial distribution and a logit link function to construct a risk association model. After the model is fitted, the system obtains the regression coefficients corresponding to caffeine and each covariate, and obtains the odds ratio by performing an exponential transformation on the regression coefficients. At the same time, it generates 95% confidence intervals and significance parameters.

[0038] This application saves the trained model parameters and provides them to the risk warning module to facilitate the calculation of the user's overall risk correlation index (OR) relative to the research reference object.

[0039] In one embodiment of this application, during the training of the risk association model, subject IDs are used as cluster identifiers, and robust covariance estimation is employed to handle the correlation between bilateral knee records of the same subject. The classification labels are binary codes, wherein the classification labels in the case knee joint records are configured as first coded values, and the classification labels in the control knee joint records are configured as second coded values.

[0040] For example, this application uses the occurrence of new radiographic knee osteoarthritis as a binary outcome variable, wherein the outcome value for the case group is set to 1 and the outcome value for the control group is set to 0; the average daily dietary caffeine intake is used as the core exposure variable, and age, sex, BMI, race, CES-D score, PASE score, comorbidity score, history of knee injury, and history of knee surgery are used as covariates.

[0041] Since the same subject may contribute records for both the left and right knees, the two knee joint records are not entirely independent. This invention uses subject IDs as the clustering unit, allowing the model to consider the correlation between knee joint records within the same subject when estimating parameter uncertainty, thereby improving the robustness of the model results. Specifically, robust covariance estimation is used to handle the correlation between the records of both knees of the same subject.

[0042] This application uses baseline dietary caffeine as the core variable, incorporates multiple covariates into the generalized estimation equation (GEE) model, and processes the correlation of bilateral knee records of the same subject according to the subject number.

[0043] In addition to using the GEE model, this application may also employ machine learning algorithms such as logistic regression and random forest.

[0044] like Figure 1 As shown, the osteoarthritis risk warning method based on dietary caffeine intake further includes the following steps: Step S400: Input the feature data to be evaluated, including dietary caffeine intake, into the trained risk association model to obtain a comprehensive risk association index that characterizes the correlation between the feature data to be evaluated and the progression of knee joint degeneration.

[0045] In this embodiment of the application, the feature data to be evaluated also includes: age, gender, body mass index, race, depression score, physical activity score, comorbidity score, history of knee injury, and history of knee surgery.

[0046] Specifically, the model input data includes: users' dietary caffeine intake, age, gender, body mass index, race, depression score, physical activity score, comorbidity score, history of knee injury, and history of knee surgery. The model results indicate the statistical association between caffeine intake and new-onset ROA.

[0047] The caffeine intake data in this application can be entered into the model using different measurement scales. For example, caffeine intake can be entered into the model in the form of an increase of 1 mg / day or an increase of one sample standard deviation. That is, one approach is to enter the model in the raw unit of mg / day to estimate the change in the dominance of new ROA when "daily dietary caffeine intake increases by 1 mg / day". For example, if daily dietary caffeine intake increases from 10 mg / day to 11 mg / day, this corresponds to the model effect of a 1 mg / day increase in caffeine. Another approach is to enter the model in the form of "an increase of one sample standard deviation". The sample standard deviation is used to represent the dispersion of caffeine intake in the study sample; in this embodiment, the sample standard deviation of caffeine intake is approximately 19.38 mg / day. Therefore, each increase of one sample standard deviation corresponds to an increase of approximately 19.38 mg / day in daily dietary caffeine intake. For example, if daily dietary caffeine intake increases from 10 mg / day to approximately 29.38 mg / day, this corresponds to the model effect of an increase of one sample standard deviation in caffeine.

[0048] In this embodiment of the application, step S400 specifically includes: Step S410: Receive the feature data to be evaluated, including dietary caffeine intake; Step S420: Verify the numerical input range and category code in the feature data to be evaluated; Step S430: If the verification passes, a comprehensive risk association index is obtained based on the trained risk association model, which characterizes the correlation between the feature data to be evaluated and the progression of knee joint degeneration.

[0049] For example, the risk warning system of this application can be deployed as a website, server interface, hospital intranet program, or mobile terminal application, and can also be integrated into existing electronic medical record systems, physical examination report interpretation systems, or independent mobile health management apps. After the user inputs their daily dietary caffeine intake and personal information such as age, gender, and BMI, the system first verifies the input range and category coding, and then calls the pre-saved risk association model parameters.

[0050] Range validation in this application refers to determining whether the continuous or ordinal variables input by the user fall within the range supported by the research sample. For example, the allowed range for age is 45 to 78 years, for BMI it is 18.5 to 43.7 kg / m², for CES-D score it is 0 to 41, for PASE physical activity score it is 0 to 454, for comorbidity score it is 0 to 6, and for daily dietary caffeine intake it is 0 to 239.2 mg / day. Categorical coding validation refers to determining whether the categorical variables input by the user belong to the coded categories defined in the model. For example, gender is coded as 1 or 2, race as 1, 2, or 3, and a history of knee injury and knee surgery is represented by 0 or 1, respectively. If the user input value exceeds the above range, or the input categorical code does not belong to the allowed code of the corresponding variable, the system stops the calculation and prompts the user to modify the input; if all inputs pass the validation, the system then calls the pre-saved risk association model parameters for subsequent calculations. The above input ranges are used to limit the application of the model within the range supported by the research sample and do not represent the medically normal range or recommended intake range.

[0051] This application also establishes a reference object composed of typical values ​​from the research sample. For continuous variables, the system calculates the contribution of the variable based on the difference between the regression coefficient of the model and the user value and the reference value; for categorical variables, the system calculates the contribution based on the difference between the corresponding parameters of the user category and the reference category. The system combines all contributions and performs an exponential transformation to obtain a comprehensive risk association index of the user relative to the research reference object.

[0052] Typical values ​​in a research sample are a set of representative reference values ​​selected from the current research sample to construct a research reference object; they serve as a mathematical benchmark for comparison. The comprehensive risk correlation index of the reference object is defined as 1, and other users are compared to it.

[0053] For example, information about the study participants included: age 60 years, BMI 28.2, CES-D depression score 4, PASE physical activity score 159, comorbidity score 0, and caffeine intake 1.99 mg / day. These continuous or ordinal variables were primarily taken from the median or near-median typical values ​​of the study sample.

[0054] In this application, continuous variables include age, BMI, CES-D depression score, PASE physical activity score, comorbidity score, and daily dietary caffeine intake, which are input into the model in numerical form. Categorical variables include gender, race, history of knee injury, and history of knee surgery. Gender is coded with 1 or 2, race with 1, 2, or 3, and history of knee injury and knee surgery with 0 or 1, where 0 indicates none and 1 indicates yes. For continuous variables, the system verifies whether their values ​​fall within the range supported by the research sample; for categorical variables, the system verifies whether their codes belong to the categories allowed by the model.

[0055] The reference values ​​and reference categories in this application are pre-defined. The system will first set a research reference object. This reference object is not calculated temporarily, but is determined in advance based on the typical values ​​of the research sample before the model is applied. Each variable entering the individual assessment corresponds to a reference value or reference category. Continuous variables correspond to reference values, and categorical variables correspond to reference categories.

[0056] In this embodiment of the application, after step S400, the method further includes: step S500, pre-setting several levels of boundary thresholds, and outputting graded early warning prompts based on the numerical relationship between the comprehensive risk correlation index and the boundary thresholds.

[0057] This application is capable of providing tiered early warning prompts, enabling early warnings for an individual's unassessed characteristic data.

[0058] In one embodiment of this application, step S500 specifically includes: pre-setting two-level boundary thresholds, and outputting three types of graded early warning prompts based on the numerical relationship between the comprehensive risk correlation index and the boundary thresholds; when the comprehensive risk correlation index is less than or equal to the first boundary threshold in the two-level boundary thresholds, outputting a risk prompt below the research reference level; when the comprehensive risk correlation index is between the two-level boundary thresholds, outputting a risk prompt close to the research reference level; when the comprehensive risk correlation index is greater than or equal to the second boundary threshold in the two-level boundary thresholds, outputting a risk prompt higher than the research reference level; wherein, the two-level boundary thresholds can be adaptively adjusted.

[0059] For example, the system outputs a three-level early warning based on the comprehensive risk correlation index: below the research reference level, close to the research reference level, or above the research reference level. For instance, when the comprehensive risk correlation index is less than or equal to 0.8, the output is "below the research reference level"; when it is greater than or equal to 1.2, the output is "above the research reference level"; and for other values, the output is "close to the research reference level". The threshold values ​​are used to display the relative level and can be adjusted based on subsequent verification results.

[0060] This application calculates a comprehensive risk correlation index relative to the research reference object based on the user's personal information and preset model parameters, and outputs a three-level early warning prompt that is lower than, close to or higher than the research reference level, thereby realizing early warning for the individual's characteristic data to be evaluated.

[0061] In one embodiment, such as Figure 2 As shown, based on the above-mentioned osteoarthritis risk warning method based on dietary caffeine intake, the present invention also provides an osteoarthritis risk warning device based on dietary caffeine intake, comprising: The acquisition module 100 is used to acquire the original paired dataset, which includes multiple matching groups. Each matching group includes: a matching group number, a set of original fields for case knee joints, and a set of original fields for control knee joints. The generation module 200 is used to parse the original field set of the case knee joint and the original field set of the control knee joint, and generate the case knee joint record and control knee joint record corresponding to each matching group. The case knee joint record and control knee joint record include: matching group number, subject number, knee side, classification label, dietary caffeine intake and model covariates. The training module 300 is used to train a pre-constructed risk association model based on the case knee joint records and control knee joint records to obtain a trained risk association model. The inference module 400 is used to input the feature data to be evaluated, including dietary caffeine intake, into the trained risk association model to obtain a comprehensive risk association index that characterizes the correlation between the feature data to be evaluated and the progression of knee joint degeneration.

[0062] It should be noted that the foregoing explanation of the osteoarthritis risk warning method embodiment based on dietary caffeine intake also applies to the osteoarthritis risk warning device based on dietary caffeine intake in this embodiment, and will not be repeated here.

[0063] This invention discloses an osteoarthritis risk warning device based on dietary caffeine intake. It acquires an original paired dataset comprising multiple matching groups, each including a matching group number, a set of original field sets for case knee joints, and a set of original field sets for control knee joints. The original field sets for case and control knee joints are parsed to generate corresponding case and control knee joint records for each matching group. These records include the matching group number, subject number, knee side, classification label, dietary caffeine intake, and model covariates. A pre-constructed risk association model is trained based on these records to obtain a trained risk association model. The unassessed feature data, including dietary caffeine intake, is input into the trained risk association model to obtain a comprehensive risk association index characterizing the correlation between the unassessed feature data and the progression of knee joint degeneration. This application trains a risk association model using case and control knee joint records, thereby obtaining a comprehensive risk association index characterizing the correlation between the unassessed feature data and the progression of knee joint degeneration, enabling early warning based on an individual's unassessed feature data.

[0064] Figure 3 A schematic diagram of the structure of a device provided in an embodiment of this application. The device may include: The memory 501, the processor 502, and the computer program stored on the memory 501 and capable of running on the processor 502.

[0065] When the processor 502 executes the program, it implements the osteoarthritis risk warning method based on dietary caffeine intake provided in the above embodiments.

[0066] Furthermore, the equipment also includes: Communication interface 503 is used for communication between memory 501 and processor 502.

[0067] The memory 501 is used to store computer programs that can run on the processor 502.

[0068] The memory 501 may include high-speed RAM memory, and may also include non-volatile memory, such as at least one disk storage device.

[0069] If the memory 501, processor 502, and communication interface 503 are implemented independently, they can be interconnected via a bus to communicate with each other. The bus can be an Industry Standard Architecture (ISA) bus, a Peripheral Component Interconnect (PCI) bus, or an Extended Industry Standard Architecture (EISA) bus, etc. Buses can be categorized as address buses, data buses, control buses, etc. For ease of representation, only one line is used in the diagram, but this does not imply that there is only one bus or one type of bus.

[0070] Optionally, in a specific implementation, if the memory 501, processor 502, and communication interface 503 are integrated on a single chip, then the memory 501, processor 502, and communication interface 503 can communicate with each other through an internal interface.

[0071] Processor 502 may be a central processing unit (CPU), an application specific integrated circuit (ASIC), or one or more integrated circuits configured to implement the embodiments of this application.

[0072] This embodiment also provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the osteoarthritis risk warning method based on dietary caffeine intake as described above.

[0073] In the description of this specification, the references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of this application. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples. Moreover, without contradiction, those skilled in the art can combine and integrate the different embodiments or examples described in this specification, as well as the features of different embodiments or examples.

[0074] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one of that feature. In the description of this application, "N" means at least two, such as two, three, etc., unless otherwise explicitly specified.

[0075] Any process or method description in the flowchart or otherwise herein can be understood as representing a module, segment, or portion of code comprising one or N executable instructions for implementing custom logic functions or processes, and the scope of the preferred embodiments of this application includes additional implementations in which functions may be performed not in the order shown or discussed, including substantially simultaneously or in reverse order depending on the functions involved, as should be understood by those skilled in the art to which embodiments of this application pertain.

[0076] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as a ordered list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by, or in conjunction with, an instruction execution system, apparatus, or device (such as a computer-based system, a processor-included system, or other system that can read and execute instructions from or in conjunction with such an instruction execution system, apparatus, or device). For the purposes of this specification, "computer-readable medium" can be any means that can contain, store, communicate, propagate, or transmit programs for use by or in conjunction with an instruction execution system, apparatus, or device. More specific examples (a non-exhaustive list) of computer-readable media include: an electrical connection having one or more wires (electronic device), a portable computer disk drive (magnetic device), random access memory (RAM), read-only memory (ROM), erasable and editable read-only memory (EPROM or flash memory), fiber optic devices, and portable optical disc read-only memory (CDROM). In addition, computer-readable media can even be paper or other suitable media on which programs can be printed, because programs can be obtained electronically by optically scanning paper or other media, then editing, interpreting or otherwise processing them as necessary, and then storing them in computer memory.

[0077] It should be understood that the various parts of this application can be implemented using hardware, software, firmware, or a combination thereof. In the above embodiments, the N steps or methods can be implemented using software or firmware stored in memory and executed by a suitable instruction execution system. If implemented in hardware, as in another embodiment, it can be implemented using any one or a combination of the following techniques known in the art: discrete logic circuits having logic gates for implementing logical functions on data signals, application-specific integrated circuits (ASICs) having suitable combinational logic gates, programmable gate arrays (PGAs), field-programmable gate arrays (FPGAs), etc.

[0078] Those skilled in the art will understand that all or part of the steps of the methods described in the above embodiments can be implemented by a program instructing related hardware. The program can be stored in a computer-readable storage medium, and when executed, it includes one or a combination of the steps of the method embodiments.

[0079] Furthermore, the functional units in the various embodiments of this application can be integrated into a processing module, or each unit can exist physically separately, or two or more units can be integrated into a module. The integrated module can be implemented in hardware or as a software functional module. If the integrated module is implemented as a software functional module and sold or used as an independent product, it can also be stored in a computer-readable storage medium.

[0080] The storage medium mentioned above can be a read-only memory, a disk, or an optical disk, etc. Although embodiments of this application have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting this application. Those skilled in the art can make changes, modifications, substitutions, and variations to the above embodiments within the scope of this application.

Claims

1. A method for early warning of osteoarthritis risk based on dietary caffeine intake, characterized in that, The method includes: Obtain the original paired dataset, which includes multiple matching groups. Each matching group includes: a matching group number, a set of original fields for case knee joints, and a set of original fields for control knee joints. The original field set of the case knee joint and the original field set of the control knee joint are parsed to generate case knee joint records and control knee joint records corresponding to each matching group. The case knee joint records and control knee joint records include: matching group number, subject number, knee side, classification label, dietary caffeine intake, and model covariates. The pre-constructed risk association model is trained based on the knee joint records of the cases and the control knee joint records to obtain the trained risk association model. The data of features to be evaluated, including dietary caffeine intake, are input into a trained risk association model to obtain a comprehensive risk association index that characterizes the correlation between the data of features to be evaluated and the progression of knee joint degeneration.

2. The osteoarthritis risk warning method based on dietary caffeine intake according to claim 1, characterized in that, The steps for constructing the risk association model include: Using dietary caffeine intake as the core exposure variable in the risk association model and model covariates as covariates in the risk association model, a risk association model is constructed based on a generalized estimation equation using a binomial distribution and a logit link function.

3. The osteoarthritis risk warning method based on dietary caffeine intake according to claim 1, characterized in that, During the training of the risk association model, the subject ID was used as the cluster identifier, and robust covariance estimation was used to handle the correlation between the two knee records of the same subject. The classification label is a binary code, wherein the classification label in the case knee joint record is configured as the first code value, and the classification label in the control knee joint record is configured as the second code value.

4. The osteoarthritis risk warning method based on dietary caffeine intake according to claim 1, characterized in that, The data on the features to be evaluated, including dietary caffeine intake, are input into a trained risk association model to obtain a comprehensive risk association index characterizing the correlation between the data on the features to be evaluated and the progression of knee joint degeneration, including: Receive feature data to be evaluated, including dietary caffeine intake; The numerical input range and category code in the feature data to be evaluated are verified; If the verification passes, a comprehensive risk association index representing the correlation between the feature data to be evaluated and the progression of knee joint degeneration is obtained based on the trained risk association model.

5. The osteoarthritis risk warning method based on dietary caffeine intake according to claim 1, characterized in that, After inputting the feature data to be evaluated, including dietary caffeine intake, into a trained risk association model to obtain a comprehensive risk association index characterizing the correlation between the feature data to be evaluated and the progression of knee joint degeneration, the following steps are also taken: Several levels of boundary thresholds are preset, and graded early warning prompts are output based on the numerical relationship between comprehensive risk correlation indicators and boundary thresholds.

6. The osteoarthritis risk warning method based on dietary caffeine intake according to claim 5, characterized in that, Several threshold levels are pre-set, and graded early warning prompts are output based on the numerical relationship between comprehensive risk correlation indicators and threshold levels, including: Two levels of thresholds are preset, and three types of graded early warning prompts are output based on the numerical relationship between the comprehensive risk correlation index and the threshold. When the comprehensive risk correlation index is less than or equal to the first threshold of the two-level threshold, a risk warning that is lower than the research reference level will be output. When the comprehensive risk correlation index is between the two-level boundary threshold, a risk warning close to the research reference level is output. When the comprehensive risk correlation index is greater than or equal to the second boundary threshold in the two-level boundary threshold, a risk warning that is higher than the research reference level will be output. The two-level boundary thresholds can be adaptively adjusted.

7. The osteoarthritis risk warning method based on dietary caffeine intake according to claim 1, characterized in that, The model covariates include: age, sex, body mass index, race, depression score, physical activity score, comorbidity score, radiographic status of the contralateral knee joint, history of knee injury, and history of knee surgery. The data to be evaluated also include: age, sex, body mass index, race, depression score, physical activity score, comorbidity score, history of knee injury, and history of knee surgery.

8. A device for early warning of osteoarthritis risk based on dietary caffeine intake, characterized in that, The device includes: The acquisition module is used to acquire the original paired dataset, which includes multiple matching groups. Each matching group includes: a matching group number, a set of original fields for case knee joints, and a set of original fields for control knee joints. The generation module is used to parse the original field set of the case knee joint and the original field set of the control knee joint, and generate the case knee joint record and control knee joint record corresponding to each matching group. The case knee joint record and control knee joint record include: matching group number, subject number, knee side, classification label, dietary caffeine intake and model covariates. The training module is used to train a pre-built risk association model based on the case knee joint records and control knee joint records to obtain a trained risk association model. The inference module is used to input the feature data to be evaluated, including dietary caffeine intake, into the trained risk association model to obtain a comprehensive risk association index that characterizes the correlation between the feature data to be evaluated and the progression of knee joint degeneration.

9. A device, characterized in that, include: The memory, the processor, and the osteoarthritis risk warning program based on dietary caffeine intake stored on the memory and executable on the processor, wherein when the osteoarthritis risk warning program based on dietary caffeine intake is executed by the processor, the steps of the osteoarthritis risk warning method based on dietary caffeine intake as described in any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that can be executed to implement the steps of the osteoarthritis risk warning method based on dietary caffeine intake as described in any one of claims 1 to 7.

Images