A disease prognosis risk early warning method and system based on big data

Abstract

The application relates to the technical field of disease prognosis risk early warning, and particularly discloses a disease prognosis risk early warning method and system based on big data, which comprises the following steps: obtaining all reference prediction patients of a current patient based on all types of health data of the current patient; obtaining the evaluation and assignment of each type of health data of the current patient based on each type of health data of the current patient and an evaluation and assignment model of each type of health data; obtaining a disease early warning reference graph and a disease early warning comparison graph of the current patient based on the evaluation and assignment of all types of health data of all reference prediction patients of the current patient; and obtaining a disease prognosis risk estimation value of the current patient based on the disease early warning reference graph and the disease early warning comparison graph of the current patient, and then obtaining a disease prognosis risk early warning result of the current patient. The application realizes accurate quantification of the degree of future development and recovery risk of the disease of the current patient, improves the accuracy and reliability of disease early warning, and ensures effective use of resources.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of disease prognosis risk early warning, and in particular relates to a disease prognosis risk early warning method and system based on big data. BACKGROUND

[0002] At present, with the progress of medical technology and the accumulation of health data, medical institutions and patients increasingly expect to obtain personalized medical services and accurate disease prediction. The traditional "one-size-fits-all" medical model has been difficult to meet the current needs, especially in terms of disease prognosis and risk early warning. In the modern medical system, the types of health data generated by patients are diverse (such as physiological health data, physical measurement data, biochemical index data, etc.), and the quantity is huge. As a core hardware with efficient data processing, logical judgment and adaptive computing capabilities, intelligent chips can quickly process massive health data through parallel computing architecture, and the integrated intelligent processing unit can adapt to the multi-dimensional analysis needs of medical big data, providing computing power support for precise data value extraction. How to rely on the efficient computing power of intelligent chips to effectively process, analyze and utilize these data to extract valuable information for disease prognosis risk early warning is an important challenge in the field of medical big data. Traditional disease early warning systems are often based on group statistical data, making it difficult to achieve precise early warning for individual patients. With the development of personalized medicine, a more refined disease early warning system is needed that can conduct individualized disease prognosis risk early warning based on individual health data of patients.

[0003] However, there is currently no disease prognosis risk early warning method and system based on big data to accurately quantify the degree of future development and outcome risk of the current patient's disease, improve the accuracy and reliability of disease early warning, so as to help medical institutions more reasonably allocate medical resources and ensure effective use of resources.

[0004] Therefore, the present application proposes a disease prognosis risk early warning method and system based on big data. SUMMARY

[0005] The application provides a disease prognosis risk early warning method and system based on big data, which obtains all reference prediction patients of a current patient according to all types of health data of the current patient, realizes accurate screening of part of the patients having a prediction reference effect on disease prognosis risk early warning of the current patient from a database, obtains each type of health data evaluation assignment of the current patient according to each type of health data of the current patient and each type of health data evaluation assignment model, facilitates subsequent drawing of a disease early warning reference graph and a disease early warning comparison graph, and then obtains the disease early warning reference graph and the disease early warning comparison graph of the current patient according to all type of health data evaluation assignments of all reference prediction patients of the current patient, facilitates subsequent calculation of a disease prognosis risk estimation value of the current patient, obtains the disease prognosis risk estimation value of the current patient according to the disease early warning reference graph and the disease early warning comparison graph of the current patient, realizes accurate quantification of the degree of future development and prognosis risk of the current patient, and finally obtains a disease prognosis risk early warning result of the current patient according to the disease prognosis risk estimation value of the current patient, improves the accuracy and reliability of disease early warning, and helps medical institutions to more reasonably allocate medical resources and ensure effective use of resources.

[0006] The application provides a disease prognosis risk early warning method based on big data, which comprises the following steps:

[0007] S1: obtaining all reference prediction patients of a current patient based on all types of health data of the current patient;

[0008] S2: obtaining each type of health data evaluation assignment of the current patient based on each type of health data of the current patient and each type of health data evaluation assignment model;

[0009] S3: obtaining a disease early warning reference graph of the current patient based on all type of health data evaluation assignments of all reference prediction patients of the current patient, and obtaining a disease early warning comparison graph of the current patient based on all type of health data evaluation assignments of the current patient;

[0010] S4: obtaining a disease prognosis risk estimation value of the current patient based on the disease early warning reference graph and the disease early warning comparison graph of the current patient, and obtaining a disease prognosis risk early warning result of the current patient based on the disease prognosis risk estimation value of the current patient.

[0011] Preferably, the disease prognosis risk early warning method based on big data, S1: obtaining all reference prediction patients of a current patient based on all types of health data of the current patient, comprises the following steps:

[0012] acquire all kinds of health data of the current patient, wherein the all kinds of health data include physiological health data, physical measurement data, biochemical index data and living habit data, and the physiological health data include heart rate, blood pressure, blood sugar and blood lipid, the physical measurement data include age, waistline and body mass index, the biochemical index data include C-reactive protein amount, liver function index and kidney function index, and the living habit data include daily smoking amount, daily drinking amount and daily exercise amount;

[0013] based on the database and the all kinds of health data of the current patient, obtain all reference prediction patients of the current patient.

[0014] Preferably, the disease prognosis risk early warning method based on big data, based on the database and the all kinds of health data of the current patient, obtains all reference prediction patients of the current patient, including:

[0015] if there are two types of sub-data similar in each kind of health data of each patient in the database and the corresponding kind of health data of the current patient, it is determined that the corresponding patient in the database is similar to the corresponding kind of health data of the current patient;

[0016] if all kinds of health data of any patient in the database are similar to the current patient, the corresponding patient in the database is regarded as the reference prediction patient of the current patient.

[0017] Preferably, the disease prognosis risk early warning method based on big data, S2: based on the each kind of health data of the current patient and the each kind of health data evaluation assignment model, obtains the each kind of health data evaluation assignment of the current patient, including:

[0018] all sub-data of each kind of health data of the current patient are input into the corresponding kind of health data evaluation assignment model as model input, and the each kind of health data evaluation assignment of the current patient is obtained.

[0019] Preferably, the disease prognosis risk early warning method based on big data, S3: based on the all kinds of health data evaluation assignment of all reference prediction patients of the current patient, obtains the disease early warning reference graph of the current patient, and based on the all kinds of health data evaluation assignment of the current patient, obtains the disease early warning comparison graph of the current patient, including:

[0020] based on the all kinds of health data evaluation assignment of all reference prediction patients of the current patient, obtain the disease early warning reference graph of the current patient;

[0021] based on the disease early warning reference graph of the current patient and the all kinds of health data evaluation assignment, obtain the disease early warning comparison graph of the current patient.

[0022] Preferably, the disease prognosis risk early warning method based on big data, based on all reference prediction of the current patient of all class health data evaluation assignment, obtains the disease early warning reference graph of the current patient, including:

[0023] The quotient value between the mean value of all reference prediction of the current patient of physiological health data evaluation assignment and the maximum value in all reference prediction of the current patient of physiological health data evaluation assignment is taken as the first coefficient;

[0024] The quotient value between the mean value of all reference prediction of the current patient of body measurement data evaluation assignment and the maximum value in all reference prediction of the current patient of body measurement data evaluation assignment is taken as the second coefficient;

[0025] The quotient value between the mean value of all reference prediction of the current patient of biochemical index data evaluation assignment and the maximum value in all reference prediction of the current patient of biochemical index data evaluation assignment is taken as the third coefficient;

[0026] The quotient value between the mean value of all reference prediction of the current patient of life habit data evaluation assignment and the maximum value in all reference prediction of the current patient of life habit data evaluation assignment is taken as the fourth coefficient;

[0027] In the first quadrant of the preset rectangular coordinate system, a straight line is drawn from the origin of the preset rectangular coordinate system with the first coefficient as the slope to obtain the first ray, in the second quadrant of the preset rectangular coordinate system, a straight line is drawn from the origin of the preset rectangular coordinate system with the negative of the second coefficient as the slope to obtain the second ray, in the third quadrant of the preset rectangular coordinate system, a straight line is drawn from the origin of the preset rectangular coordinate system with the third coefficient as the slope to obtain the third ray, in the fourth quadrant of the preset rectangular coordinate system, a straight line is drawn from the origin of the preset rectangular coordinate system with the negative of the fourth coefficient as the slope to obtain the fourth ray;

[0028] Based on the first ray, the second ray, the third ray and the fourth ray, the disease early warning reference graph of the current patient is obtained.

[0029] Preferably, the disease prognosis risk early warning method based on big data, based on the first ray, the second ray, the third ray and the fourth ray, the disease early warning reference graph of the current patient is obtained, including:

[0030] Among all points on the first ray, the point with the distance from the end point of the first ray being the sum value of all reference prediction of the current patient of physiological health data evaluation assignment is taken as the first construction point of the disease early warning reference graph of the current patient;

[0031] connecting all points on the second ray with the end point of the second ray, and assigning a sum value to each point on the second ray, wherein the sum value is the evaluation of the physiological health data of the current patient, to obtain a second construction point of the disease early warning reference map of the current patient;

[0032] connecting all points on the third ray with the end point of the third ray, and assigning a sum value to each point on the third ray, wherein the sum value is the evaluation of the biochemical index data of the current patient, to obtain a third construction point of the disease early warning reference map of the current patient;

[0033] connecting all points on the fourth ray with the end point of the fourth ray, and assigning a sum value to each point on the fourth ray, wherein the sum value is the evaluation of the life habit data of the current patient, to obtain a fourth construction point of the disease early warning reference map of the current patient;

[0034] connecting the first construction point, the second construction point, the third construction point and the fourth construction point of the disease early warning reference map of the current patient, to obtain the disease early warning reference map of the current patient.

[0035] Preferably, the disease prognosis risk early warning method based on big data comprises the following steps:

[0036] connecting the geometric center of the disease early warning reference map of the current patient with the first construction point, the second construction point, the third construction point and the fourth construction point of the disease early warning reference map respectively and unidirectionally extending, to obtain a fifth ray, a sixth ray, a seventh ray and an eighth ray;

[0037] connecting all points on the fifth ray with the end point of the fifth ray, and assigning a sum value to each point on the fifth ray, wherein the sum value is the evaluation of the physiological health data of the current patient, to obtain a first construction point of the disease early warning comparison map of the current patient;

[0038] connecting all points on the sixth ray with the end point of the sixth ray, and assigning a sum value to each point on the sixth ray, wherein the sum value is the evaluation of the body measurement data of the current patient, to obtain a second construction point of the disease early warning comparison map of the current patient;

[0039] connecting all points on the seventh ray with the end point of the seventh ray, and assigning a sum value to each point on the seventh ray, wherein the sum value is the evaluation of the biochemical index data of the current patient, to obtain a third construction point of the disease early warning comparison map of the current patient;

[0040] connecting all points on the eighth ray with the end point of the eighth ray, and assigning a sum value to each point on the eighth ray, wherein the sum value is the evaluation of the life habit data of the current patient, to obtain a fourth construction point of the disease early warning comparison map of the current patient;

[0041] connecting the first construction point, the second construction point, the third construction point and the fourth construction point of the disease early warning comparison map of the current patient, to obtain the disease early warning comparison map of the current patient.

[0042] Preferably, the disease prognosis risk early warning method based on big data, S4: based on the disease early warning reference graph and the disease early warning comparison graph of the current patient, obtaining the disease prognosis risk estimation value of the current patient, and based on the disease prognosis risk estimation value of the current patient, obtaining the disease prognosis risk early warning result of the current patient, comprising:

[0043] Based on the disease early warning reference graph and the disease early warning comparison graph of the current patient, obtaining the disease prognosis risk estimation value of the current patient;

[0044] If the disease prognosis risk estimation value of the current patient is greater than the preset risk estimation value threshold, the need for early warning is regarded as the disease prognosis risk early warning result of the current patient, otherwise, the no need for early warning is regarded as the disease prognosis risk early warning result of the current patient.

[0045] The application provides a disease prognosis risk early warning system based on big data, which is used for executing any one of the disease prognosis risk early warning methods based on big data in embodiments 1-9, comprising:

[0046] The preprocessing module is used for obtaining all reference prediction patients of the current patient based on all types of health data of the current patient.

[0047] The assignment module is used for obtaining the evaluation value of each type of health data of the current patient based on each type of health data of the current patient and the evaluation assignment model of each type of health data.

[0048] The drawing module is used for obtaining the disease early warning reference graph of the current patient based on the evaluation value of all types of health data of all reference prediction patients of the current patient, and obtaining the disease early warning comparison graph of the current patient based on the evaluation value of all types of health data of the current patient.

[0049] The early warning module is used for obtaining the disease prognosis risk estimation value of the current patient based on the disease early warning reference graph and the disease early warning comparison graph of the current patient, and obtaining the disease prognosis risk early warning result of the current patient based on the disease prognosis risk estimation value of the current patient.

[0050] The beneficial effects generated by the present application relative to the prior art are: with the high-efficiency computing power support of the intelligent chip, all reference prediction patients of the current patient are obtained according to all kinds of health data of the current patient, and part of the patients that have a prediction reference effect on the disease prognosis risk warning of the current patient are accurately screened out from the database; based on the rapid data processing capability of the intelligent chip, each kind of health data evaluation assignment of the current patient is obtained according to each kind of health data of the current patient and each kind of health data evaluation assignment model, which is convenient for subsequent drawing of the disease warning reference graph and the disease warning comparison graph; then, relying on the parallel computing advantage of the intelligent chip on large-scale data, the disease warning reference graph and the disease warning comparison graph of the current patient are obtained according to all kinds of health data evaluation assignments of all reference prediction patients of the current patient, which is convenient for subsequent calculation of the disease prognosis risk estimate value; finally, based on the precise calculation capability of the intelligent chip, the disease prognosis risk estimate value of the current patient is obtained according to the disease warning reference graph and the disease warning comparison graph of the current patient, which realizes accurate quantification of the degree of the future development and the risk of the current patient, and finally the disease prognosis risk warning result of the current patient is obtained according to the disease prognosis risk estimate value of the current patient, which improves the accuracy and reliability of the disease warning, and helps medical institutions to more reasonably allocate medical resources and ensure effective use of resources.

[0051] Other features and advantages of the present application will be set forth in the following description, and in part will become apparent to those skilled in the art from the description, or can be learned by practice of the present application. The objects and other advantages of the present application can be realized and attained by the structure particularly pointed out in the written description and claims hereof.

[0052] The technical solutions of the present application will be further described in detail below with the help of the drawings and examples. BRIEF DESCRIPTION OF DRAWINGS

[0053] The accompanying drawings are included to provide a further understanding of the present application, and constitute a part of the specification, illustrate embodiments of the present application and explain the present application together with the embodiments, but do not constitute a limitation on the present application. In the drawings:

[0054] Figure 1 A flow chart of a disease prognosis risk warning method based on big data in an embodiment of the present application;

[0055] Figure 2 A schematic diagram of a disease prognosis risk warning system based on big data in an embodiment of the present application. DETAILED DESCRIPTION

[0056] The preferred embodiments of the present application will be described below in conjunction with the drawings, and it should be understood that the preferred embodiments described herein are only used to illustrate and explain the present application, and do not limit the present application.

[0057] Embodiment 1: The present application provides a disease prognosis risk early warning method based on big data, referring to Figure 1 , comprising:

[0058] S1: Based on all types of health data of the current patient, all reference prediction patients of the current patient are obtained;

[0059] S2: Based on each type of health data of the current patient and each type of health data evaluation assignment model, the evaluation assignment of each type of health data of the current patient is obtained;

[0060] S3: Based on the evaluation assignment of all types of health data of all reference prediction patients of the current patient, the disease early warning reference graph of the current patient is obtained, and based on the evaluation assignment of all types of health data of the current patient, the disease early warning comparison graph of the current patient is obtained;

[0061] S4: Based on the disease early warning reference graph and the disease early warning comparison graph of the current patient, the disease prognosis risk estimation value of the current patient is obtained, and based on the disease prognosis risk estimation value of the current patient, the disease prognosis risk early warning result of the current patient is obtained.

[0062] In this embodiment, the current patient is a patient who applies a disease prognosis risk early warning system based on big data to perform disease future development and outcome risk early warning.

[0063] In this embodiment, the health data is data related to the health condition of each patient, including physiological health data, body measurement data, biochemical index data and lifestyle habit data.

[0064] In this embodiment, the reference prediction patient is a part of the patients in the database that has a prediction reference effect on the disease prognosis risk early warning of the current patient.

[0065] In this embodiment, each type of health data evaluation assignment model is a model that inputs all sub-data of each type of health data of a large number of patients collected in advance as model input, and outputs the corresponding type of health data evaluation assignment of the corresponding patient by artificial annotation as model output, and is trained to input all sub-data of each type of health data of the current patient, and can output the corresponding type of health data evaluation assignment of the current patient.

[0066] In this embodiment, the evaluation assignment of each type of health data of the current patient is an assignment that can represent the state degree of each type of health data of the current patient, which is obtained by performing health evaluation on the current patient according to each type of health data of the current patient.

[0067] In this embodiment, the disease early warning reference graph of the current patient is a reference graph that is drawn based on the evaluation assignment of all types of health data of all reference prediction patients of the current patient, and is used to perform disease future development and outcome risk early warning on the current patient.

[0068] In this embodiment, the disease warning comparison chart of the current patient is drawn based on the evaluation assignment of all types of health data of the current patient, and is used to warn the future development and prognosis risk of the current patient.

[0069] In this embodiment, the disease prognosis risk estimation value of the current patient is obtained based on the disease warning reference chart and the disease warning comparison chart of the current patient, and is used to represent the degree of the future development and prognosis risk of the current patient.

[0070] In this embodiment, the disease prognosis risk warning result of the current patient is obtained based on the disease prognosis risk estimation value of the current patient, and is used to warn the future development and prognosis risk of the current patient.

[0071] The beneficial effects of the above technology are as follows: according to all types of health data of the current patient, all reference prediction patients of the current patient are obtained, which realizes accurate screening of part of the patients who have prediction reference effect on the disease prognosis risk warning of the current patient in the database, according to each type of health data of the current patient and each type of health data evaluation assignment model, each type of health data evaluation assignment of the current patient is obtained, which is convenient for subsequent drawing of the disease warning reference chart and the disease warning comparison chart, and then according to all types of health data evaluation assignments of all reference prediction patients of the current patient, the disease warning reference chart and the disease warning comparison chart of the current patient are obtained, which is convenient for subsequent calculation of the disease prognosis risk estimation value of the current patient, according to the disease warning reference chart and the disease warning comparison chart of the current patient, the disease prognosis risk estimation value of the current patient is obtained, which realizes accurate quantification of the degree of the future development and prognosis risk of the current patient, and finally according to the disease prognosis risk estimation value of the current patient, the disease prognosis risk warning result of the current patient is obtained, which improves the accuracy and reliability of disease warning, and helps medical institutions to more reasonably allocate medical resources and ensure effective use of resources.

[0072] Embodiment 2: Based on embodiment 1, the disease prognosis risk warning method based on big data, S1: based on all types of health data of the current patient, all reference prediction patients of the current patient are obtained, including:

[0073] All types of health data of the current patient are obtained, wherein all types of health data include physiological health data, body measurement data, biochemical index data and living habit data, and the physiological health data includes heart rate, blood pressure, blood sugar and blood lipid, the body measurement data includes age, waist circumference and body mass index, the biochemical index data includes C-reactive protein amount, liver function index and kidney function index, and the living habit data includes daily smoking amount, daily drinking amount and daily exercise amount.

[0074] Based on the database and all kinds of health data of the current patient, all reference prediction patients of the current patient are obtained.

[0075] In this embodiment, the body mass index is an important indicator for measuring the overall health of the human body calculated by weight and height.

[0076] In this embodiment, the C-reactive protein amount is the concentration or content of an acute phase response protein synthesized by liver cells and released into the plasma when the body is infected or tissue is damaged.

[0077] In this embodiment, the liver function index is a biochemical index for evaluating the health status of the liver, and the liver function index in this embodiment is specifically the alanine aminotransferase content.

[0078] In this embodiment, the kidney function index is a biochemical index for evaluating the health status and function of the kidney, and the kidney function index in this embodiment is specifically the serum creatinine content.

[0079] In this embodiment, the daily smoking amount is the number of cigarettes smoked by the patient per day.

[0080] In this embodiment, the daily drinking amount is the amount (ml) of alcohol consumed by the patient per day.

[0081] In this embodiment, the daily exercise amount is the duration (min) of exercise per day.

[0082] The beneficial effects of the above technology are: the specific items of all kinds of health data of the current patient are determined, and then all reference prediction patients of the current patient are obtained based on the database and all kinds of health data of the current patient.

[0083] Embodiment 3: Based on embodiment 2, the disease prognosis risk early warning method based on big data, all reference prediction patients of the current patient are obtained based on the database and all kinds of health data of the current patient, including:

[0084] If there are two types of sub-data similar in each kind of health data of each patient in the database and the corresponding kind of health data of the current patient, it is determined that the corresponding patient in the database is similar to the corresponding kind of health data of the current patient;

[0085] If all kinds of health data of any patient in the database are similar to the current patient, the corresponding patient in the database is regarded as a reference prediction patient of the current patient.

[0086] In this embodiment, the database is a pre-set database, and a large amount of all kinds of health data of patients are stored in the database.

[0087] In this embodiment, the sub-data is the sub-data contained in each kind of health data, for example, the heart rate is the sub-data of the physiological health data.

[0088] In this embodiment, one type of sub-data similarity is that the difference between the numerical values of one type of sub-data of each type of health data of two patients (each patient in the database and the current patient) is less than a preset difference value (a difference threshold value preset to determine whether the sub-data is similar, and each type of sub-data corresponds to a preset difference value).

[0089] The above technology has the beneficial effect that a specific method for obtaining all reference prediction patients of the current patient based on all types of health data of the database and the current patient is provided in detail, and the reference prediction patients with similar characteristics to the current patient are accurately screened, so as to generate a highly personalized disease prognosis risk warning result for the current patient.

[0090] Embodiment 4: Based on embodiment 1, the disease prognosis risk warning method based on big data, S2: obtaining the health data evaluation value of each type of health data of the current patient based on each type of health data of the current patient and the health data evaluation value model of each type of health data, including:

[0091] All sub-data of each type of health data of the current patient is input into the corresponding health data evaluation value model as a model input, and the health data evaluation value of each type of health data of the current patient is obtained.

[0092] The above technology has the beneficial effect that the health data evaluation value of each type of health data of the current patient is obtained according to each type of health data of the current patient and the health data evaluation value model of each type of health data, and the state degree of each type of health data of the current patient is quantified, which facilitates the drawing of the disease warning reference graph and the disease warning comparison graph.

[0093] Embodiment 5: Based on embodiment 1, the disease prognosis risk warning method based on big data, S3: obtaining the disease warning reference graph of the current patient based on the health data evaluation value of all types of health data of all reference prediction patients of the current patient, and obtaining the disease warning comparison graph of the current patient based on the health data evaluation value of all types of health data of the current patient, including:

[0094] The disease warning reference graph of the current patient is obtained based on the health data evaluation value of all types of health data of all reference prediction patients of the current patient;

[0095] The disease warning comparison graph of the current patient is obtained based on the disease warning reference graph of the current patient and the health data evaluation value of all types of health data.

[0096] The above technology has the beneficial effect that the disease warning reference graph and the disease warning comparison graph of the current patient are obtained according to the health data evaluation value of all types of health data of all reference prediction patients of the current patient, which facilitates the calculation of the disease prognosis risk estimate value.

[0097] In the embodiment 6, based on the embodiment 5, the disease prognosis risk early warning method based on big data, the mean value of the physiological health data evaluation assignment of all the reference predicted patients of the current patient is obtained, and the maximum value of the physiological health data evaluation assignment of all the reference predicted patients of the current patient is obtained.

[0098] The quotient value between the mean value of the physiological health data evaluation assignment of all the reference predicted patients of the current patient and the maximum value of the physiological health data evaluation assignment of all the reference predicted patients of the current patient is taken as the first coefficient.

[0099] The quotient value between the mean value of the body measurement data evaluation assignment of all the reference predicted patients of the current patient and the maximum value of the body measurement data evaluation assignment of all the reference predicted patients of the current patient is taken as the second coefficient.

[0100] The quotient value between the mean value of the biochemical index data evaluation assignment of all the reference predicted patients of the current patient and the maximum value of the biochemical index data evaluation assignment of all the reference predicted patients of the current patient is taken as the third coefficient.

[0101] The quotient value between the mean value of the life habit data evaluation assignment of all the reference predicted patients of the current patient and the maximum value of the life habit data evaluation assignment of all the reference predicted patients of the current patient is taken as the fourth coefficient.

[0102] In the first quadrant of the preset rectangular coordinate system, a straight line is drawn from the origin of the preset rectangular coordinate system with the first coefficient as the slope to obtain a first ray, in the second quadrant of the preset rectangular coordinate system, a straight line is drawn from the origin of the preset rectangular coordinate system with the negative of the second coefficient as the slope to obtain a second ray, in the third quadrant of the preset rectangular coordinate system, a straight line is drawn from the origin of the preset rectangular coordinate system with the third coefficient as the slope to obtain a third ray, and in the fourth quadrant of the preset rectangular coordinate system, a straight line is drawn from the origin of the preset rectangular coordinate system with the negative of the fourth coefficient as the slope to obtain a fourth ray.

[0103] Based on the first ray, the second ray, the third ray and the fourth ray, the disease early warning reference graph of the current patient is obtained.

[0104] In the embodiment, the preset rectangular coordinate system is a pre-set rectangular coordinate system.

[0105] The above technology has the beneficial effect that a specific method for evaluating all the reference predicted patients of the current patient based on all the health data evaluation assignments of the current patient is provided, and the first ray, the second ray, the third ray and the fourth ray are obtained, which facilitates the subsequent drawing of the disease early warning reference graph of the current patient.

[0106] Embodiment 7: Based on embodiment 6, the disease prognosis risk early warning method based on big data, based on the first ray, the second ray, the third ray and the fourth ray, obtains the disease early warning reference graph of the current patient, including:

[0107] Among all the points on the first ray, the point with the distance from the end point of the first ray being the sum of the physiological health data evaluation values of all the reference predicted patients of the current patient is taken as the first construction point of the disease early warning reference graph of the current patient;

[0108] Among all the points on the second ray, the point with the distance from the end point of the second ray being the sum of the body measurement data evaluation values of all the reference predicted patients of the current patient is taken as the second construction point of the disease early warning reference graph of the current patient;

[0109] Among all the points on the third ray, the point with the distance from the end point of the third ray being the sum of the biochemical index data evaluation values of all the reference predicted patients of the current patient is taken as the third construction point of the disease early warning reference graph of the current patient;

[0110] Among all the points on the fourth ray, the point with the distance from the end point of the fourth ray being the sum of the life habit data evaluation values of all the reference predicted patients of the current patient is taken as the fourth construction point of the disease early warning reference graph of the current patient;

[0111] The first construction point, the second construction point, the third construction point and the fourth construction point of the disease early warning reference graph of the current patient are connected to obtain the disease early warning reference graph of the current patient.

[0112] The beneficial technology of the above technology is that a specific method for obtaining the disease early warning reference graph of the current patient according to the first ray, the second ray, the third ray and the fourth ray is given in detail, which facilitates the calculation of the subsequent disease prognosis risk estimate value.

[0113] Embodiment 8: Based on embodiment 7, the disease prognosis risk early warning method based on big data, based on the disease early warning reference graph of the current patient and all the health data evaluation values, obtains the disease early warning comparison graph of the current patient, including:

[0114] The geometric center of the disease early warning reference graph of the current patient is connected to the first construction point, the second construction point, the third construction point and the fourth construction point of the disease early warning reference graph respectively and is unidirectionally extended to obtain the fifth ray, the sixth ray, the seventh ray and the eighth ray;

[0115] Among all the points on the fifth ray, the point with the distance from the end point of the fifth ray being the physiological health data evaluation value of the current patient is taken as the first construction point of the disease early warning comparison graph of the current patient;

[0116] all points on the sixth ray, the distance from the end point of the sixth ray is the evaluation of the body measurement data of the current patient, as the second construction point of the disease early warning comparison chart of the current patient;

[0117] all points on the seventh ray, the distance from the end point of the seventh ray is the evaluation of the biochemical index data of the current patient, as the third construction point of the disease early warning comparison chart of the current patient;

[0118] all points on the eighth ray, the distance from the end point of the eighth ray is the evaluation of the living habit data of the current patient, as the fourth construction point of the disease early warning comparison chart of the current patient;

[0119] connecting the first construction point, the second construction point, the third construction point and the fourth construction point of the disease early warning comparison chart of the current patient, to obtain the disease early warning comparison chart of the current patient.

[0120] In this embodiment, the geometric center is the centroid of the disease early warning reference chart of the current patient.

[0121] In this embodiment, the one-way extension is to extend one end of the construction point (the first construction point, the second construction point, the third construction point and the fourth construction point) and not to extend one end of the geometric center.

[0122] The beneficial effects of the above technology are: a specific method for obtaining the disease early warning comparison chart of the current patient according to the disease early warning reference chart of the current patient and all health data evaluation assignments is provided in detail, which is convenient for subsequent disease prognosis risk estimate value calculation.

[0123] Embodiment 9: Based on embodiment 1, the disease prognosis risk early warning method based on big data, S4: based on the disease early warning reference chart of the current patient and the disease early warning comparison chart, obtaining the disease prognosis risk estimate value of the current patient, and based on the disease prognosis risk estimate value of the current patient, obtaining the disease prognosis risk early warning result of the current patient, including:

[0124] obtaining the disease prognosis risk estimate value of the current patient based on the disease early warning reference chart of the current patient and the disease early warning comparison chart;

[0125] If the disease prognosis risk estimate value of the current patient is greater than the preset risk estimate value threshold, the need for early warning is regarded as the disease prognosis risk early warning result of the current patient, otherwise, the no need for early warning is regarded as the disease prognosis risk early warning result of the current patient.

[0126] In this embodiment, the disease prognosis risk estimate value of the current patient is obtained based on the disease early warning reference chart of the current patient and the disease early warning comparison chart, that is:

[0127] ;

[0128] wherein, β is a disease prognosis risk estimation value of the current patient, is an area of the disease early warning comparison graph of the current patient, is an area of the disease early warning reference graph of the current patient, is a length of a portion of the disease early warning comparison graph of the current patient in the first quadrant, is a length of a portion of the disease early warning comparison graph of the current patient in the second quadrant, is a length of a portion of the disease early warning comparison graph of the current patient in the third quadrant, is a length of a portion of the disease early warning comparison graph of the current patient in the fourth quadrant, is a length of a portion of the disease early warning reference graph of the current patient in the first quadrant, is a length of a portion of the disease early warning reference graph of the current patient in the second quadrant, is a length of a portion of the disease early warning reference graph of the current patient in the third quadrant, is a length of a portion of the disease early warning reference graph of the current patient in the fourth quadrant, n is a total number of all reference prediction patients of the current patient, is a distance length between a first construction point of the disease early warning comparison graph of the current patient and the geometric center, is a distance length between a second construction point of the disease early warning comparison graph of the current patient and the geometric center, is a distance length between a third construction point of the disease early warning comparison graph of the current patient and the geometric center, is a distance length between a fourth construction point of the disease early warning comparison graph of the current patient and the geometric center, ln is a natural logarithm, and the value of a natural constant e is 2.718.

[0129] In this embodiment, the preset risk estimation value threshold is a disease prognosis risk estimation threshold that is preset to judge the disease prognosis risk early warning result of the current patient.

[0130] The above technology has the beneficial effects that: according to the disease prognosis risk estimation value of the current patient, the disease prognosis risk early warning result of the current patient is obtained, a specific method for calculating the disease prognosis risk estimation value of the current patient is given in detail, the accuracy and reliability of disease early warning are improved, which helps medical institutions to more reasonably allocate medical resources and ensure effective use of resources.

[0131] Embodiment 10: The present application provides a disease prognosis risk early warning system based on big data, which is used to execute any one of the disease prognosis risk early warning methods based on big data in embodiments 1 to 9, comprising:

[0132] a preprocessing module, configured to obtain all reference prediction patients of the current patient based on all health-related data of the current patient;

[0133] The assignment module is configured to obtain each type of health data evaluation assignment of the current patient based on each type of health data of the current patient and the evaluation assignment model of each type of health data.

[0134] The drawing module is configured to obtain the disease early warning reference graph of the current patient based on the all-type health data evaluation assignments of the all-reference predicted patients of the current patient, and obtain the disease early warning comparison graph of the current patient based on the all-type health data evaluation assignments of the all-reference predicted patients of the current patient.

[0135] The early warning module is configured to obtain the disease prognosis risk estimation value of the current patient based on the disease early warning reference graph and the disease early warning comparison graph of the current patient, and obtain the disease prognosis risk early warning result of the current patient based on the disease prognosis risk estimation value of the current patient.

[0136] The above-mentioned technical effects are that: according to the all-type health data of the current patient, the all-reference predicted patients of the current patient are obtained, which realizes accurate screening of part of the patients in the database that have a prediction reference effect on the disease prognosis risk early warning of the current patient, according to the each type of health data of the current patient and the evaluation assignment model of each type of health data, the each type of health data evaluation assignment of the current patient is obtained, which is convenient for subsequent drawing of the disease early warning reference graph and the disease early warning comparison graph, and then according to the all-type health data evaluation assignments of the all-reference predicted patients of the current patient, the disease early warning reference graph and the disease early warning comparison graph of the current patient are obtained, which is convenient for subsequent calculation of the disease prognosis risk estimation value, according to the disease early warning reference graph and the disease early warning comparison graph of the current patient, the disease prognosis risk estimation value of the current patient is obtained, which realizes accurate quantification of the degree of the future development and the risk of the current patient, and finally according to the disease prognosis risk estimation value of the current patient, the disease prognosis risk early warning result of the current patient is obtained, which improves the accuracy and reliability of the disease early warning, and helps medical institutions to more reasonably allocate medical resources and ensure effective use of resources.

[0137] Obviously, those skilled in the art can make various modifications and variations to the present application without departing from the spirit and scope of the present application, and the present application also intends to include these modifications and variations.

Claims

1. A disease prognosis risk early warning method based on big data, characterized in that, include: S1: Based on all health data of the current patient, obtain all reference predicted patients for the current patient; S2: Based on the current patient's health data for each category and the health data assessment and assignment model for each category, obtain the health data assessment and assignment for the current patient for each category. S3: Based on all references of the current patient, evaluate and assign values ​​to all types of health data of the predicted patients to obtain the disease warning reference for the current patient, and based on all types of health data of the current patient, obtain the disease warning comparison chart for the current patient; S4: Based on the current patient's disease early warning reference chart and disease early warning comparison chart, obtain the current patient's disease prognosis risk prediction value, and based on the current patient's disease prognosis risk prediction value, obtain the current patient's disease prognosis risk early warning result; Specifically, based on the evaluation and assignment of all health data of all reference patients for the current patient, a disease warning reference image for the current patient is obtained, and based on the evaluation and assignment of all health data of the current patient, a disease warning comparison image for the current patient is obtained, including: Connect the first, second, third, and fourth construction points of the current patient's disease warning reference to obtain the current patient's disease warning reference; Connect the first, second, third, and fourth construction points of the current patient's disease warning comparison map to obtain the current patient's disease warning comparison map; Specifically, based on the current patient's disease early warning reference chart and disease early warning comparison chart, a disease prognostic risk prediction value for the current patient is obtained. Based on this prediction value, a disease prognostic risk early warning result for the current patient is obtained, including: Based on the current patient's disease early warning reference chart and disease early warning comparison chart, obtain the current patient's disease prognosis risk prediction value; ； Where β is the current prognostic risk estimate for the patient. The area of ​​the current patient's disease early warning comparison map. The area of ​​the reference diagram is used for the current patient's disease early warning. The length of the disease warning comparison chart for the current patient within the first quadrant. The length of the current patient's disease early warning comparison chart within the second quadrant. The length of the current patient's disease early warning comparison chart within the third quadrant. The length of the current patient's disease early warning comparison chart within the fourth quadrant. The disease warning reference for the current patient is shown in the first quadrant, representing a portion of the length of the reference area. For the current patient's disease warning, refer to the portion of the reference area in the second quadrant. For the current patient's disease warning, refer to the portion of the reference in the third quadrant. The disease warning reference for the current patient is shown in the fourth quadrant, where n is the total number of all reference predicted patients for the current patient. The distance between the first construction point and the geometric center of the current patient's disease early warning comparison map. The distance between the second construction point and the geometric center of the current patient's disease early warning comparison map. The distance between the third construction point and the geometric center of the current patient's disease early warning comparison map. Let ln be the distance between the fourth construction point of the current patient's disease warning comparison map and the geometric center, ln be the natural logarithm, and the natural constant e be 2.

718. If the current patient's estimated disease prognosis risk is greater than the preset risk prediction threshold, an alert will be issued as the current patient's disease prognosis risk warning result; otherwise, no alert will be issued as the current patient's disease prognosis risk warning result.

2. The disease prognosis risk early warning method based on big data according to claim 1, characterized in that, S1: Based on all health data of the current patient, obtain all reference predicted patients for the current patient, including: Obtain all types of health data for the current patient, including physiological health data, body measurement data, biochemical indicators data, and lifestyle data. Physiological health data includes heart rate, blood pressure, blood sugar, and blood lipids; body measurement data includes age, waist circumference, and body mass index; biochemical indicators include C-reactive protein levels, liver function indicators, and kidney function indicators; and lifestyle data includes daily smoking amount, daily alcohol consumption, and daily exercise amount. Based on the database and all health data of the current patient, obtain all reference predicted patients for the current patient.

3. The disease prognosis risk early warning method based on big data according to claim 2, characterized in that, Based on the database and all health data of the current patient, obtain all reference predicted patients for the current patient, including: If there are two types of sub-data similar to each type of health data of each patient in the database and the corresponding type of health data of the current patient, then the corresponding patient in the database is determined to be similar to the corresponding type of health data of the current patient. If any patient in the database has similar health data across all categories to the current patient, then the corresponding patient in the database will be used as the reference patient for prediction.

4. The disease prognosis risk early warning method based on big data according to claim 1, characterized in that, S2: Based on the current patient's health data for each category and the assessment and assignment model for each category of health data, obtain the assessment and assignment for each category of health data for the current patient, including: All sub-data of each type of health data for the current patient are used as model inputs and fed into the corresponding health data assessment and assignment model to obtain the assessment and assignment of each type of health data for the current patient.

5. The disease prognosis risk early warning method based on big data according to claim 1, characterized in that, S3: Based on the evaluation and assignment of all health data of all reference patients for the current patient, obtain the disease warning reference image for the current patient, and based on the evaluation and assignment of all health data of the current patient, obtain the disease warning comparison image for the current patient, including: Based on all references of the current patient, evaluate and assign values ​​to all types of health data of the predicted patients to obtain a disease warning reference for the current patient; Based on the current patient's disease warning reference and the evaluation and assignment of all types of health data, a disease warning comparison chart for the current patient is obtained.

6. The disease prognosis risk early warning method based on big data according to claim 5, characterized in that, Based on the current patient's reference data and all health data of all classes, an assessment and assignment is performed to obtain a disease early warning reference for the current patient, including: The first coefficient is the quotient between the mean of the physiological health data assessments assigned to all reference predictive patients for the current patient and the maximum value among the physiological health data assessments assigned to all reference predictive patients for the current patient. The quotient between the mean of the assessment values ​​assigned to the body measurement data of all reference predictive patients for the current patient and the maximum value among the assessment values ​​assigned to the body measurement data of all reference predictive patients for the current patient is used as the second coefficient. The quotient between the mean of the biochemical indicator data assessment and assignment of all reference predictive patients for the current patient and the maximum value among the biochemical indicator data assessment and assignment of all reference predictive patients for the current patient is used as the third coefficient. The quotient between the mean of the life habit data assessments assigned to all reference predictive patients for the current patient and the maximum value among the life habit data assessments assigned to all reference predictive patients for the current patient is used as the fourth coefficient. In the first quadrant of the preset rectangular coordinate system, a straight line is drawn from the origin of the preset rectangular coordinate system with the first coefficient as the slope to obtain the first ray. In the second quadrant of the preset rectangular coordinate system, a straight line is drawn from the origin of the preset rectangular coordinate system with the negative value of the second coefficient as the slope to obtain the second ray. In the third quadrant of the preset rectangular coordinate system, a straight line is drawn from the origin of the preset rectangular coordinate system with the third coefficient as the slope to obtain the third ray. In the fourth quadrant of the preset rectangular coordinate system, a straight line is drawn from the origin of the preset rectangular coordinate system with the negative value of the fourth coefficient as the slope to obtain the fourth ray. Based on the first, second, third, and fourth rays, a disease early warning reference diagram for the current patient is obtained.

7. The disease prognosis risk early warning method based on big data according to claim 6, characterized in that, Based on the first, second, third, and fourth rays, a disease early warning reference diagram for the current patient is obtained, including: Among all points on the first ray, the point whose distance from the endpoint of the first ray is equal to the sum of the physiological health data assessment values ​​assigned to all reference predicted patients for the current patient is taken as the first construction point of the disease early warning reference for the current patient; Among all points on the second ray, the point whose distance from the endpoint of the second ray is equal to the sum of the values ​​assigned by the body measurement data of all reference predicted patients for the current patient is used as the second construction point of the disease warning reference for the current patient; Among all points on the third ray, the point whose distance from the endpoint of the third ray is equal to the sum of the biochemical index data assessment values ​​of all reference predicted patients for the current patient is used as the third construction point of the disease early warning reference for the current patient; Among all points on the fourth ray, the point whose distance from the endpoint of the fourth ray is equal to the sum of the lifestyle data assessment values ​​assigned to all reference patients for the current patient is used as the fourth construction point of the disease early warning reference for the current patient.

8. The disease prognosis risk early warning method based on big data according to claim 7, characterized in that, Based on the current patient's disease early warning reference chart and the evaluation and assignment of all types of health data, a disease early warning comparison chart for the current patient is obtained, including: Connect the geometric center of the current patient's disease warning reference point to the first, second, third, and fourth construction points of the disease warning reference point, and extend them in one direction to obtain the fifth, sixth, seventh, and eighth rays; Among all points on the fifth ray, the point whose distance from the endpoint of the fifth ray is equal to the current patient's physiological health data assessment value is used as the first construction point of the current patient's disease warning comparison map; Among all points on the sixth ray, the point whose distance from the endpoint of the sixth ray is equal to the current patient's body measurement data assessment value is used as the second construction point of the current patient's disease warning comparison map; Among all points on the seventh ray, the point whose distance from the endpoint of the seventh ray is equal to the value assigned to the current patient's biochemical indicator data is used as the third construction point of the current patient's disease warning comparison map; Among all points on the eighth ray, the point whose distance from the endpoint of the eighth ray is the current patient's lifestyle data assessment value is used as the fourth construction point of the current patient's disease warning comparison map.

9. A disease prognosis risk early warning system based on big data, characterized in that, A method for implementing a big data-based disease prognosis risk early warning system as described in any one of claims 1 to 8, comprising: The preprocessing module is used to obtain all reference predicted patients for the current patient based on all types of health data for the current patient; The assignment module is used to obtain the assessment assignment of each type of health data for the current patient based on each type of health data and the assessment assignment model for each type of health data. The drawing module is used to evaluate and assign values ​​to all types of health data of the current patient based on all references of the current patient, to obtain a disease warning reference image for the current patient, and to obtain a disease warning comparison image for the current patient based on the evaluation and assignment of all types of health data of the current patient. The early warning module is used to obtain the current patient's disease prognosis risk prediction value based on the current patient's disease early warning reference map and disease early warning comparison map, and to obtain the current patient's disease prognosis risk early warning result based on the current patient's disease prognosis risk prediction value.

Images