A risk assessment method for geriatric endocrine diseases based on big data

Abstract

The present invention relates to the technical field of medical information management for the elderly endocrine system, and specifically relates to a method for risk assessment of elderly endocrine diseases based on big data. The present invention screens out all the highest similarity vectors of the target medical vector; obtains the basic risk factors according to the detection time distribution and similarity degree between the two; obtains the self-risk factors by combining the detection times of all medical vectors; obtains the data change risk factors according to the number of similar patients with the disease and the data difference and detection time difference between the medical vector and the first medical vector; obtains the endocrine system stability factor by using the similarity degree between medical vectors, combines them to obtain the dynamic risk factor, and further conducts a risk assessment of elderly endocrine diseases for the person to be evaluated. The present invention takes into account the physiological changes of the elderly population and the timeliness of risk factors, making the risk assessment of elderly endocrine diseases more accurate.

Description

Technical Field

[0001] The present invention relates to the field of medical information management for the elderly endocrine system, and particularly to a method for risk assessment of elderly endocrine diseases based on big data. Background Art

[0002] Elderly endocrine diseases refer to endocrine system diseases that occur with age, such as diabetes, hypothyroidism or hyperthyroidism, osteoporosis, etc. These diseases pose a great threat to the physical health of the elderly. Through disease screening and risk assessment in hospitals, potential endocrine diseases can be detected and prevented early, preventing the aggravation of the condition and reducing the impact of the disease on daily life.

[0003] In the current medical field, big data technology that integrates and analyzes a large amount of medical data and extracts disease-related features and risk factors from it is a commonly used and accurate disease risk assessment technology. However, most of the existing big data-based risk assessment methods for elderly endocrine diseases rely on static data and empirical models, and find similar characteristic manifestations by matching with historical patient data to infer the disease risk of patients. However, this method fails to fully utilize dynamic data, does not consider the physiological changes of the elderly population and the timeliness of risk factors, and often ignores the data change correlation between patients. Therefore, the risk assessment of endocrine diseases is often inaccurate, missing the best prevention opportunity. Summary of the Invention

[0004] To solve the technical problem that the conventional big data model cannot take into account the physiological changes and timeliness of risk factors of the elderly population, and will ignore the data change correlation between patients, resulting in inaccurate risk assessment for elderly endocrine diseases, the purpose of the present invention is to provide a risk assessment method for elderly endocrine diseases based on big data. The specific technical solution adopted is as follows: A risk assessment method for elderly endocrine diseases based on big data, the method includes: obtaining the medical vectors of the person to be evaluated at different detection times and the first medical vectors of each treated patient at different detection times; taking the medical vector of the first detection time of the person to be evaluated as the target medical vector; screening each treated patient according to the cosine similarity between the target medical vector and all the first medical vectors of each treated patient to obtain the similar diseased patients of the person to be evaluated and the highest similarity vector among all the first medical vectors of each similar diseased patient; obtaining the basic risk factors of the person to be evaluated according to the number of similar diseased patients, the similarity degree between the target medical vector and the highest similarity data of each similar diseased patient, and the detection time distribution where the highest similarity data is located; obtaining the self-risk factors of the person to be evaluated according to the basic risk factors and the detection time distribution of all the medical vectors of the person to be evaluated; obtaining the data change risk factors of the person to be evaluated according to the number of similar diseased patients, the data difference degree and the detection time difference degree between all the medical vectors of the person to be evaluated and all the first medical vectors of each similar diseased patient; obtaining the endocrine system stability factor of the person to be evaluated according to the similarity degree between the medical vectors of every two adjacent detection times of the person to be evaluated; obtaining the dynamic risk factor of the person to be evaluated according to the endocrine system stability factor, the self-risk factor and the data change risk factor of the person to be evaluated; and performing risk assessment for elderly endocrine diseases on the person to be evaluated according to the dynamic risk factor.

[0005] Further, the method for obtaining the similar diseased patients of the person to be evaluated and the highest similarity vector among all the first medical vectors of each similar diseased patient includes: calculating the cosine similarity between the target medical vector and all the first medical vectors of each treated patient, and screening out the first medical vector with the highest cosine similarity as the highest similarity vector; when the cosine similarity between the target medical vector and the highest similarity vector is not less than a preset first threshold, taking the treated patient corresponding to the highest similarity vector as a similar patient; and taking the similar patient with a history of endocrine diseases as a similar diseased patient.

[0006] Further, the method for obtaining the basic risk factors includes: obtaining the basic risk factors according to the basic risk factor calculation formula, and the basic risk factor calculation formula is as follows: In the formula, represents the basic risk factor of the person to be evaluated; Indicates the number of similar patients with the disease; Indicates the interval between the detection time corresponding to the highest similarity vector of the th similar patient with the disease and the disease onset time, where the disease onset time of the th similar patient with the disease can be directly obtained; Indicates the target medical vector of the person to be evaluated; Indicates the highest similarity vector of the th similar patient with the disease; Indicates the cosine similarity between the target medical vector and the highest similarity vector of the th similar patient with the disease; Indicates the number of similar patients.

[0007] Furthermore, the method for obtaining the self-risk factor includes: obtaining the self-risk factor according to the self-risk factor calculation formula, and the self-risk factor calculation formula is as follows: In the formula, Indicates the self-risk factor of the person to be evaluated; Indicates the number of medical vectors of the person to be evaluated; Indicates the basic risk factor of the th medical vector; Indicates the last detection time of the person to be evaluated; Indicates the detection time corresponding to the

[0008] th medical vector. Furthermore, the method for obtaining the data change risk factor includes: randomly selecting a similar patient with the disease as the target patient with the disease; in the first medical vector of the target patient with the disease, selecting the highest similarity vector corresponding to the first medical vector of the person to be evaluated as the first similar vector; after the detection time corresponding to the first similar vector, selecting the highest similarity vector corresponding to the second medical vector of the person to be evaluated as the second similar vector, until the highest similarity vectors corresponding to each medical vector of the person to be evaluated are obtained; forming a first matrix with each medical vector of the person to be evaluated, and forming a second matrix with the highest similarity vectors corresponding to each medical vector of the person to be evaluated in the first medical vector of each similar patient with the disease; obtaining the data change risk factor according to the data change risk factor calculation formula, and the data change risk factor calculation formula is as follows: In the formula, Indicates the data change risk factor of the person to be evaluated; Indicates the number of similar patients with the disease; Indicates the interval between the last detection time and the disease onset time of the th similar patient with the disease; Indicates each medical vector of the person to be evaluated in the first matrix; represent each highest similarity vector of similar diseased patients in the th second matrix; represent the number of detection times; represent the detection time difference between every two adjacent medical vectors of the person to be evaluated; represent the detection time difference between every two adjacent highest similarity vectors of the th similar diseased patient;

[0009] Furthermore, the method for obtaining the endocrine system stability factor includes: obtaining the endocrine system stability factor according to the endocrine system stability factor calculation formula, and the endocrine system stability factor calculation formula is as follows: In the formula, represent the endocrine system stability factor of the person to be evaluated; represent the number of detection times; represent the th medical vector of the person to be evaluated; represent the th medical vector of the evaluation patient; represent the cosine similarity between the th medical vector and the th medical vector of the person to be evaluated; represent the normalization function.

[0010] Furthermore, the method for obtaining the dynamic risk factor includes: obtaining the dynamic risk factor according to the dynamic risk factor calculation formula, and the dynamic risk factor calculation formula is as follows: In the formula, represent the dynamic risk factor of the person to be evaluated; represent the endocrine system stability factor of the person to be evaluated; represent the self-risk factor of the person to be evaluated; represent the data change risk factor of the person to be evaluated; represent the normalization function.

[0011] A big data-based risk assessment system for geriatric endocrine diseases, the system includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the steps of a big data-based risk assessment method for geriatric endocrine diseases as described above are implemented.

[0012] A computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps of a big data-based risk assessment method for geriatric endocrine diseases as described above are implemented.

[0013] A computer device includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the steps of the above-mentioned method for risk assessment of geriatric endocrine diseases based on big data are implemented.

[0014] The present invention has the following beneficial effects: The present invention obtains the medical vectors of the person to be evaluated at different detection times and the first medical vectors of each treated patient at different detection times. Since factors such as the physiological functions, lifestyles, and environmental conditions of different patients are different and can easily affect the pathogenesis of endocrine diseases, the first medical vectors of the person to be evaluated and the treated patients are compared and analyzed to obtain the basic risk factors of the person to be evaluated. Since the physical functions of the elderly population will undergo significant physiological changes over time, it is necessary to combine the characteristics of the changes in the medical data of the person to be evaluated with the characteristics of the data changes between the person to be evaluated and other similar diseased patients. Therefore, the self-risk factors and data change risk factors of the person to be evaluated are analyzed. Since the changes in the functions of the endocrine system of the elderly are unstable and may lead to irregular changes in their medical data, the endocrine system stability factors of the person to be evaluated are analyzed. The self-risk factors and data change risk factors of the person to be evaluated are combined and analyzed to obtain the dynamic risk factors of the person to be evaluated. The risk of geriatric endocrine diseases of the person to be evaluated is assessed based on the dynamic risk factors. The present invention takes into account the physiological changes of the elderly population and the timeliness of risk factors, making the risk assessment of geriatric endocrine diseases more accurate. Description of the Drawings

[0015] In order to more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for the description of the embodiments or the prior art. Obviously, the following-described drawings are only some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0016] Figure 1 It is a flowchart of a method for risk assessment of geriatric endocrine diseases based on big data provided by an embodiment of the present invention. Detailed Embodiments

[0017] To further elaborate on the technical means and effects adopted by the present invention to achieve the intended invention purpose, the following specifically describes, with reference to the accompanying drawings and preferred embodiments, a method for risk assessment of elderly endocrine diseases based on big data according to the present invention, including its specific implementation manner, structure, features and effects. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. In addition, the specific features, structures or characteristics in one or more embodiments can be combined in any suitable form.

[0018] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by those skilled in the technical field to which the present invention belongs.

[0019] The following specifically describes the specific solution of a method for risk assessment of elderly endocrine diseases based on big data provided by the present invention with reference to the accompanying drawings.

[0020] Please refer to Figure 1 , which shows a method for risk assessment of elderly endocrine diseases based on big data provided by an embodiment of the present invention. The method includes: Step S1: Obtain the medical vectors of the person to be evaluated at different detection times and the first medical vectors of each treated patient at different detection times.

[0021] The embodiments of the present invention are mainly applied to the prediction and assessment scenario of elderly endocrine diseases. In order to predict whether the elderly will suffer from endocrine diseases, it is first necessary to conduct medical tests on the elderly in different dimensions, such as blood glucose, blood lipids, renal function, urine protein, thyroid, sex hormones, etc. Integrate the medical data in different dimensions into a feature vector as the medical vector, and regard the elderly to be tested as the person to be evaluated. Since the physical state of the elderly may change greatly at different times, the medical vectors of the person to be evaluated at different detection times are obtained. Since the embodiments of the present invention analyze whether the person to be evaluated suffers from endocrine diseases based on a big data model, it is necessary to obtain the medical big data related to the endocrine of the elderly population through channels such as medical institutions, research institutions or public databases, and form the first medical vector with the medical data of the treated patients in the same detection dimensions as the person to be evaluated, that is, obtain the first medical vectors of each treated patient at different detection times.

[0022] In one embodiment of the present invention, the detection time is set to be weekly. It should be noted that in other embodiments of the present invention, the detection time can be set by oneself and is not limited herein.

[0023] Step S2: Use the medical vector of the first detection moment of the person to be evaluated as the target medical vector; screen each treated patient according to the cosine similarity between the target medical vector and all the first medical vectors of each treated patient to obtain the similar diseased patients of the person to be evaluated and the highest similarity vector among all the first medical vectors of each similar diseased patient; obtain the basic risk factors of the person to be evaluated according to the number of similar diseased patients, the similarity degree between the target medical vector and the highest similarity data of each similar diseased patient, and the detection moment distribution where the highest similarity data is located; obtain the self-risk factors of the person to be evaluated according to the basic risk factors and the detection moment distribution of all the medical vectors of the person to be evaluated; obtain the data change risk factors of the person to be evaluated according to the number of similar diseased patients, the data difference degree and the detection moment difference degree between all the medical vectors of the person to be evaluated and all the first medical vectors of each similar diseased patient.

[0024] In reality, there are many factors that cause geriatric endocrine diseases, and there are significant individual differences among elderly patients. Different patients have different physiological functions, lifestyles, environmental conditions, etc., which can easily affect the pathogenesis of endocrine diseases. Therefore, by comparing and analyzing the first medical vectors of the person to be evaluated and the treated patients, we can obtain the similar diseased patients of the person to be evaluated and the highest similarity vector among all the first medical vectors of each similar diseased patient, and obtain the basic risk factors of the person to be evaluated through the differences between the medical vectors of the person to be evaluated and the highest similarity vectors.

[0025] Preferably, in an embodiment of the present invention, the method for obtaining the similar diseased patients of the person to be evaluated and the highest similarity vector among all the first medical vectors of each similar diseased patient includes: calculating the cosine similarity between the target medical vector and all the first medical vectors of each treated patient, and screening out the first medical vector with the highest cosine similarity as the highest similarity vector.

[0026] When the cosine similarity between the target medical vector and the highest similarity vector is not less than a preset first threshold, the treated patient corresponding to the highest similarity vector is used as a similar patient. In an embodiment of the present invention, the preset first threshold is set to 0.6. It should be noted that in other embodiments of the present invention, the preset first threshold can be set by oneself and will not be limited here.

[0027] Since there may be different disease types among the treated patients, in order to analyze whether the person to be evaluated has an endocrine disease, the similar patients with a history of endocrine disease are used as the similar diseased patients, and the first medical vectors of the similar diseased patients are analyzed.

[0028] Preferably, in an embodiment of the present invention, the method for obtaining the basic risk factor includes: obtaining the basic risk factor according to the basic risk factor calculation formula, and the basic risk factor calculation formula is as follows: In the formula, represents the basic risk factor of the person to be evaluated; represents the number of similar diseased patients; represents the interval between the detection time corresponding to the highest similarity vector of the th similar diseased patient and the diseased time, where the diseased time of the th similar diseased patient can be directly obtained; represents the target medical vector of the person to be evaluated; represents the highest similarity vector of the th similar diseased patient; represents the cosine similarity between the target medical vector and the highest similarity vector of the th similar diseased patient; represents the number of similar patients.

[0029] In the basic risk factor calculation formula, the smaller the time interval between the detection time corresponding to the highest similarity vector of the th similar diseased patient and the diseased time is, the closer the target medical vector of the person to be evaluated is to the diseased condition of the th similar diseased patient. At this time, the risk of the person to be evaluated is greater, and the greater the cosine similarity between the target medical vector and the highest similarity vector of the th similar diseased patient is, the closer the target medical vector is to the first medical vector at the diseased time of the th similar diseased patient. At this time, the risk of the person to be evaluated is greater; compare each treated patient with the person to be evaluated to obtain the comprehensive risk index of the person to be evaluated; among similar patients, if the number of patients with endocrine diseases is larger, that is is larger, the risk of the person to be evaluated is greater at this time, that is, the basic risk factor of the person to be evaluated is larger.

[0030] In actual situations, the physical functions of the elderly population will undergo significant physiological changes over time. For example, the decline of liver and kidney functions and the change of drug metabolism ability will lead to changes in the incidence risk of endocrine diseases. In addition, the resistance of the elderly is relatively poor, and changes in risk factors such as lifestyle and environment are likely to affect the function of their endocrine system, which will also lead to changes in the incidence risk of endocrine diseases. Therefore, it is difficult to ensure the accuracy of the results of risk assessment for elderly endocrine diseases through static data, and it is necessary to combine the characteristics of the changes in the medical data of the person to be evaluated with the characteristics of the data changes between the person to be evaluated and other similar patients with the disease.

[0031] First, analyze the dynamic changes of the medical vectors of the person to be evaluated at different detection times to obtain the person's own risk factors.

[0032] Preferably, in an embodiment of the present invention, the method for obtaining the own risk factors includes: obtaining the own risk factors according to the own risk factor calculation formula, and the own risk factor calculation formula is as follows: In the formula, represents the own risk factor of the person to be evaluated; represents the number of medical vectors of the person to be evaluated; represents the basic risk factor of the th medical vector; represents the detection time corresponding to the

[0033] In the own risk factor calculation formula, the smaller the time interval between the th medical vector and the medical vector at the last detection time, the greater the time difference between the th medical vector and the medical vector at the first detection time, the greater the dynamic change. At this time, the basic risk factor of the th medical vector of the person to be evaluated occupies a greater weight, that is, is larger. Analyze each medical vector to obtain the own risk factor of the person to be evaluated.

[0034] Secondly, analyze the data change risk factors of the person to be evaluated according to the medical data changes between the similar patients with the disease and the person to be evaluated.

[0035] Preferably, in an embodiment of the present invention, the method for obtaining the data change risk factor includes: randomly selecting a similar diseased patient as the target diseased patient; in the first medical vector of the target diseased patient, selecting the highest similarity vector corresponding to the first medical vector of the person to be evaluated as the first similar vector; after the detection time corresponding to the first similar vector, selecting the highest similarity vector corresponding to the second medical vector of the person to be evaluated as the second similar vector, until the highest similarity vectors corresponding to each medical vector of the person to be evaluated are obtained; forming a first matrix with each medical vector of the person to be evaluated, and forming a second matrix with the highest similarity vectors corresponding to each medical vector of the person to be evaluated in the first medical vector of each similar diseased patient. An embodiment of the present invention provides a method for obtaining the first matrix and all second matrices, specifically including: 1. Denote that the current elderly individual to be evaluated has collected times of medical vectors, then the th collected medical vector is ; 2. For the first medical vector of any similar diseased patient, denote the first medical vector with the highest similarity to as ; 3. Calculate the eigenvector with the highest similarity to from all the first medical vectors after the detection time where is located, and denote it as ; 4. Repeat step 3 until all the first medical vectors are traversed, and combine , , …, in order and denote it as the first matrix , and combine , , …, in order and denote it as the second matrix .

[0036] Repeat steps 2 to 4, and construct the second matrix from the first medical vectors of the th similar diseased patient.

[0037] Obtain the data change risk factor according to the data change risk factor calculation formula. The data change risk factor calculation formula is as follows: In the formula, represents the data change risk factor of the person to be evaluated; represents the number of similar diseased patients; represents the interval between the last detection time and the diseased time of the th similar diseased patient; represents each medical vector of the person to be evaluated in the first matrix; represents the Each highest similarity vector of similar diseased patients in the second matrix; Indicates the number of detection times; Indicates the detection time difference between every two adjacent medical vectors of the person to be evaluated; Indicates the Detection time difference between every two adjacent highest similarity vectors of the Indicates the number of similar patients.

[0038] In the calculation formula of the data change risk factor, the smaller the interval between the last detection time and the diseased time of the th similar diseased patient, that is, the is larger, it indicates that the th similar diseased patient is more likely to be in a diseased state at the last detection time, and at this time, the diseased risk of the person to be evaluated is greater; Indicates the first matrix and the second matrix The Frobenius norm of the residual between them. The smaller this value is, the smaller the difference between the two matrices, indicating that the medical data of the th similar diseased patient and the person to be evaluated are more similar. At this time, the weight of the first medical vector of the th similar diseased patient for the person to be evaluated is greater; the detection time difference of the matrices is smaller, indicating that in the two matrices, the detection time difference between every two adjacent vectors is smaller. At this time, the diseased conditions of the th similar diseased patient and the person to be evaluated are closer. At this time, the weight of the first medical vector of the th similar diseased patient for the person to be evaluated is greater; if the number of patients with endocrine diseases is larger, that is, the is larger, at this time, the risk of the person to be evaluated is greater, and at this time, the data change risk factor of the person to be evaluated is larger. is larger, at this time, the risk of the person to be evaluated is greater, and at this time, the data change risk factor of the person to be evaluated is larger.

[0039] Step S3: Obtain the endocrine system stability factor of the person to be evaluated according to the similarity degree between every two adjacent medical vectors of the person to be evaluated; obtain the dynamic risk factor of the person to be evaluated according to the endocrine system stability factor, the self-risk factor and the data change risk factor of the person to be evaluated.

[0040] Due to the instability of the changes in the endocrine system function of the elderly, it may lead to irregular changes in their medical data. Therefore, in the embodiments of the present invention, the endocrine system stability factor of the person to be evaluated is obtained according to the similarity degree between every two adjacent medical vectors of the person to be evaluated.

[0041] Preferably, in one embodiment of the present invention, the method for obtaining the endocrine system stability factor includes: obtaining the endocrine system stability factor according to the endocrine system stability factor calculation formula, and the endocrine system stability factor calculation formula is as follows: In the formula, represents the endocrine system stability factor of the person to be evaluated; represents the number of detection times; represents the th medical vector of the person to be evaluated; represents the th medical vector of the evaluated patient; represents the th medical vector and the th medical vector of the person to be evaluated; represents the cosine similarity between them;

[0042] In the endocrine system stability factor calculation formula, the greater the cosine similarity between every two adjacent medical vectors of the person to be evaluated, the smaller the change in the endocrine system of the person to be evaluated. At this time, the greater it is, the greater the endocrine system stability factor of the person to be evaluated.

[0043] In the above steps, the self-risk factor and the data change risk factor of the person to be evaluated can be obtained. Combining with the endocrine system stability factor of the person to be evaluated, they are combined and analyzed to obtain the dynamic risk factor of the person to be evaluated.

[0044] Preferably, in one embodiment of the present invention, the method for obtaining the dynamic risk factor includes: obtaining the dynamic risk factor according to the dynamic risk factor calculation formula, and the dynamic risk factor calculation formula is as follows: In the formula, represents the dynamic risk factor of the person to be evaluated; represents the endocrine system stability factor of the person to be evaluated; represents the self-risk factor of the person to be evaluated; represents the data change risk factor of the person to be evaluated; represents the normalization function.

[0045] In the dynamic risk factor calculation formula, the greater the endocrine system stability factor of the person to be evaluated, that is, the stronger the endocrine system stability. At this time, the higher the credibility of the data change risk factor of the person to be evaluated, and the higher the weight of the data change risk factor; if the endocrine system stability factor of the person to be evaluated is smaller, that is, the weaker the endocrine system stability. At this time, the higher the credibility of the self-risk factor of the person to be evaluated, and the higher the weight of the self-risk factor.

[0046] So far, the dynamic risk factors of the person to be evaluated are obtained.

[0047] Step S4: Conduct a risk assessment of geriatric endocrine diseases for the person to be evaluated based on the dynamic risk factors.

[0048] In an embodiment of the present invention, a preset second threshold is set to 0.5, and a preset third threshold is set to 0.7. A risk assessment method provided by an embodiment of the present invention for assessing the endocrine disease risk of the person to be evaluated specifically includes: If , it is considered that the person to be evaluated has a certain risk of endocrine diseases and needs to receive medical guidance; if , it is considered that the person to be evaluated has a relatively high risk of endocrine diseases, and it is necessary to give an early warning to the patient and his / her family members in a timely manner and let them receive appropriate treatment.

[0049] It should be noted that the preset second threshold and the preset third threshold can be set by oneself and are not limited here.

[0050] So far, the risk assessment of geriatric endocrine diseases is completed.

[0051] In summary, obtain the medical vectors of the person to be evaluated at different detection times and the first medical vectors of each treated patient at different detection times; use the medical vector of the person to be evaluated at the first detection time as the target medical vector; screen each treated patient according to the cosine similarity between the target medical vector and all the first medical vectors of each treated patient to obtain the similar diseased patients of the person to be evaluated and the highest similarity vector among all the first medical vectors of each similar diseased patient; obtain the basic risk factors of the person to be evaluated according to the number of similar diseased patients, the similarity degree between the target medical vector and the highest similarity data of each similar diseased patient, and the detection time distribution where the highest similarity data is located; obtain the own risk factors of the person to be evaluated according to the basic risk factors and detection time distribution of all the medical vectors of the person to be evaluated; obtain the data change risk factors of the person to be evaluated according to the number of similar diseased patients, the data difference degree and detection time difference degree between all the medical vectors of the person to be evaluated and all the first medical vectors of each similar diseased patient; obtain the endocrine system stability factor of the person to be evaluated according to the similarity degree between the medical vectors of the person to be evaluated at every two adjacent detection times; obtain the dynamic risk factors of the person to be evaluated according to the endocrine system stability factor, own risk factors and data change risk factors of the person to be evaluated; conduct a risk assessment of geriatric endocrine diseases for the person to be evaluated based on the dynamic risk factors.

[0052] An embodiment of the present invention provides a risk assessment system for elderly endocrine diseases based on big data. The system includes a memory, a processor, and a computer program. The memory is used to store the corresponding computer program, and the processor is used to run the corresponding computer program. When the computer program runs in the processor, it can implement the methods described in steps S1 - S4.

[0053] The third object of the embodiments of the present invention is to provide a computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the steps of the method described in the above steps S1 - S4.

[0054] The fourth object of the embodiments of the present invention is to provide a computer-readable storage medium. The computer-readable storage medium stores a computer program. When the computer program is executed by the processor, it implements the steps of the method described in the above steps S1 - S4.

[0055] It should be noted that: the above sequence of the embodiments of the present invention is only for description and does not represent the advantages and disadvantages of the embodiments. The processes depicted in the drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0056] Each embodiment in this specification is described in a progressive manner. The same or similar parts between the embodiments can be referred to each other. Each embodiment focuses on the differences from other embodiments.

Claims

1. A risk assessment method for geriatric endocrine diseases based on big data, characterized in that, The method includes: obtaining the medical vectors of the person to be evaluated at different detection times and the first medical vectors of each treated patient at different detection times; taking the medical vector of the person to be evaluated at the first detection time as the target medical vector; screening each treated patient according to the cosine similarity between the target medical vector and all the first medical vectors of each treated patient, obtaining the similar diseased patients of the person to be evaluated and the highest similarity vector among all the first medical vectors of each similar diseased patient; obtaining the basic risk factors of the person to be evaluated according to the number of similar diseased patients, the similarity degree between the target medical vector and the highest similarity vector of each similar diseased patient, and the detection time distribution where the highest similarity vector is located; obtaining the self-risk factors of the person to be evaluated according to the basic risk factors and the detection time distribution of all the medical vectors of the person to be evaluated; obtaining the data change risk factors of the person to be evaluated according to the number of similar diseased patients, the data difference degree and the detection time difference degree between all the medical vectors of the person to be evaluated and all the first medical vectors of each similar diseased patient; obtaining the endocrine system stability factor of the person to be evaluated according to the similarity degree between the medical vectors of the person to be evaluated at every two adjacent detection times; performing risk assessment of geriatric endocrine diseases on the person to be evaluated according to the endocrine system stability factor, the self-risk factors and the data change risk factors of the person to be evaluated; when the cosine similarity between the target medical vector and the highest similarity vector is not less than a preset first threshold, taking the treated patient corresponding to the highest similarity vector as a similar patient; taking the similar patient with a history of endocrine diseases as a similar diseased patient; arbitrarily selecting one similar diseased patient as the target diseased patient; in the first medical vector of the target diseased patient, selecting the highest similarity vector corresponding to the first medical vector of the person to be evaluated as the first similar vector; after the detection time corresponding to the first similar vector, selecting the highest similarity vector corresponding to the second medical vector of the person to be evaluated as the second similar vector, until obtaining the highest similarity vector corresponding to each medical vector of the person to be evaluated; forming a first matrix with each medical vector of the person to be evaluated, and forming a second matrix with the highest similarity vectors corresponding to each medical vector of the person to be evaluated in the first medical vectors of each similar diseased patient; the calculation formula of the data change risk factor is as follows: In the formula, represents the data change risk factor of the person to be evaluated; represents the number of similar diseased patients; represents the interval between the last detection time and the onset time of the th similar diseased patient; represents each medical vector of the person to be evaluated in the first matrix; represents the highest similarity vector of the th similar diseased patient in the second matrix; Indicates the number of detection times; Indicates the detection time difference between every two adjacent medical vectors of the person to be evaluated; Indicates the detection time difference between every two adjacent highest similarity vectors of the th similar diseased patient; Indicates the number of similar patients.

2. The risk assessment method for geriatric endocrine diseases based on big data according to claim 1, wherein The method for obtaining the basic risk factor includes: obtaining the basic risk factor according to the basic risk factor calculation formula, and the basic risk factor calculation formula is as follows: In the formula, represents the basic risk factor of the person to be evaluated; represents the number of similar patients with the disease; represents the interval between the detection time corresponding to the highest similarity vector of the th similar patient with the disease and the onset time, where the onset time of the th similar patient with the disease can be directly obtained; represents the target medical vector of the person to be evaluated; represents the highest similarity vector of the th similar patient with the disease; represents the cosine similarity between the target medical vector and the highest similarity vector of the th similar patient with the disease; represents the number of similar patients.

3. A method for risk assessment of geriatric endocrine diseases based on big data according to claim 1, characterized in that, The method for obtaining the self-risk factor includes: obtaining the self-risk factor according to the self-risk factor calculation formula, and the self-risk factor calculation formula is as follows: In the formula, represents the self-risk factor of the person to be evaluated; represents the number of medical vectors of the person to be evaluated; represents the basic risk factor of the represents the last detection time of the person to be evaluated; represents the detection time corresponding to the th medical vector.

4. The risk assessment method for geriatric endocrine diseases based on big data according to claim 1, wherein The method for obtaining the endocrine system stability factor includes: obtaining the endocrine system stability factor according to the endocrine system stability factor calculation formula, and the endocrine system stability factor calculation formula is as follows: In the formula, represents the endocrine system stability factor of the person to be evaluated; represents the number of detection times; represents the medical vector of the person to be evaluated at the th detection time; represents the medical vector of the evaluated patient at the th detection time; represents the cosine similarity between the medical vector of the person to be evaluated at the th detection time and the medical vector at the th detection time; represents the normalization function.

5. The method for risk assessment of geriatric endocrine diseases based on big data according to claim 4, characterized in that, Obtain a dynamic risk factor based on the endocrine system stability factor, the individual risk factor, and the data change risk factor of the person to be evaluated, and conduct a risk assessment of geriatric endocrine diseases for the person to be evaluated according to the dynamic risk factor, including: The calculation formula of the dynamic risk factor is as follows: In the formula, represents the dynamic risk factor of the person to be evaluated; represents the endocrine system stability factor of the person to be evaluated; represents the individual risk factor of the person to be evaluated; represents the data change risk factor of the person to be evaluated; represents a normalization function.

6. A risk assessment system for geriatric endocrine diseases based on big data, the system comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, the steps of a method for risk assessment of elderly endocrine diseases based on big data as described in any one of claims 1 to 5 are implemented.

7. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by a processor, the steps of a method for risk assessment of elderly endocrine diseases based on big data as described in any one of claims 1 to 5 are implemented.

8. A computer device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, the steps of a method for risk assessment of elderly endocrine diseases based on big data as described in any one of claims 1 to 5 are implemented.

Images