A method, system and storage medium for constructing orthopedic database

Abstract

The present invention relates to the technical field of orthopedic data partitioning, and in particular to an orthopedic database construction method, system and storage medium. The method first obtains the initial age community of the patient through the SLPA algorithm; in the initial age community, the initial association weight is obtained according to the distribution of the patient and the adjacent node patients in bone density and bone metabolism markers; suspected abnormal patients are screened out according to the correlation change between age and bone density, and then the patients with bone abnormalities are screened out according to the difference in blood indicators; the initial association weight is adjusted according to the patient and the adjacent node bone abnormality patients, and the final association weight is obtained to update the initial age community to obtain the final age community for constructing an orthopedic database. In the process of partitioning, the present invention considers the abnormal degree of bone difference to adaptively adjust the association weight between ages, so that the partition result based on age and bone condition is highly robust, the database construction is of higher quality, and the analysis efficiency is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of orthopedic data partitioning, and in particular to an orthopedic database construction method, system and storage medium. Background Art

[0002] The orthopedic database can store a large amount of information on orthopedic patients. Through the analysis and comparison of orthopedic data in the database, it can effectively summarize and learn about similar orthopedic cases, providing a good data reference for subsequent learning and treatment, and providing necessary information for medical teaching and academic research.

[0003] When dividing the data structure of orthopedic data, the sub-data structure is usually divided based on orthopedic disease conditions. However, in the patient population, there are differences in bone development due to the different ages of patients, and different bone development conditions require different analyses and studies. For example, the bones of elderly patients are generally more porous, so patients need to be divided by age. However, there are also different bone conditions at similar ages. Therefore, general age divisions cannot truly reflect similar bone conditions. The inaccurate structural division of orthopedic data makes the orthopedic database inefficient in further data analysis and mining. Summary of the invention

[0004] In order to solve the technical problem that the structure division of orthopedic data in the prior art is not accurate enough, which makes the orthopedic database inefficient in further data analysis and mining, the purpose of the present invention is to provide an orthopedic database construction method, system and storage medium, and the technical solutions adopted are as follows:

[0005] An embodiment of the present invention provides a method for constructing an orthopedic database, the method comprising:

[0006] Obtain each patient's age, bone density, bone metabolism markers, and blood indicators from historical orthopedic medical data; use the SLPA algorithm to obtain the initial age community based on the age of all patients;

[0007] In each initial age community, the initial association weight between each patient and each adjacent node is determined according to the distribution of bone density and each bone metabolism marker between each patient and each adjacent node;

[0008] In each initial age community, based on the changes in the correlation between age and bone density, suspected abnormal patients are identified from all adjacent nodes of each patient; the degree of bone abnormality of each suspected abnormal patient is obtained based on the deviation of each suspected abnormal patient from all non-suspected abnormal patients in the initial age community in each blood index; patients with bone abnormality are screened out from suspected abnormal patients based on the degree of bone abnormality;

[0009] In the initial age community, the final association weight of each patient with each adjacent node is obtained according to the initial association weight of each patient with each adjacent node and the degree of bone abnormality of patients with bone abnormality in the adjacent node;

[0010] The initial age community of the SLPA algorithm is updated through the final association weight to obtain the final age community; and the orthopedic database is constructed based on the final age community.

[0011] Furthermore, the method for obtaining the initial association weight includes:

[0012] According to the correlation between each bone metabolism marker and bone density in different patients, the importance of the impact of each bone metabolism marker was determined;

[0013] In each initial age community, the initial association weight between each patient and each adjacent node is determined based on the difference in bone density between each patient and the patients at each adjacent node, as well as the difference and impact importance of each bone metabolism marker.

[0014] Furthermore, the method for obtaining the impact importance includes:

[0015] Arrange all patients in order of bone density from small to large to obtain a patient sequence; arrange the bone density corresponding to the patients in the order of the patient sequence to obtain a bone density sequence; arrange each bone metabolism marker corresponding to the patients in the order of the patient sequence to obtain a bone metabolism marker sequence of each bone metabolism marker;

[0016] According to the correlation change degree between the bone metabolism marker sequence and the bone density sequence of each bone metabolism marker, the change correlation of each bone metabolism marker is obtained;

[0017] The ratio of the change correlation of each bone metabolism marker to the maximum change correlation among all bone metabolism markers was taken as the impact importance of each bone metabolism marker.

[0018] Furthermore, the initial association weight between each patient and each adjacent node is determined according to the difference in bone density between each patient and the patient of each adjacent node, as well as the difference and impact importance of each bone metabolism marker, including:

[0019] According to the difference between each patient and the patients of each adjacent node in each bone metabolism marker, and the importance of the influence of each bone metabolism marker, the influence of each bone metabolism marker is obtained;

[0020] The influence degree of all bone metabolism markers of each patient and the patients of each adjacent node is combined to obtain the difference degree of influence between each patient and the patients of each adjacent node;

[0021] The difference in bone density between each patient and the patient at each adjacent node was taken as the bone density deviation between each patient and the patient at each adjacent node;

[0022] The initial association weight between each patient and each adjacent node was obtained based on the degree of difference in influence and the bone density deviation between each patient and each adjacent node; both the degree of difference in influence and the bone density deviation were negatively correlated with the initial association weight.

[0023] Furthermore, the method for obtaining the suspected abnormal patient includes:

[0024] For any initial age community, all patients in the initial age community are mapped to the bone density change coordinate system to obtain the change coordinate system of the initial age community; the horizontal axis of the bone density change coordinate system is age, and the vertical axis is bone density;

[0025] Performing straight-line fitting on the data of the initial age community in the change coordinate system to obtain a change straight line of the initial age community; for any patient in the initial age community, determining the adjacent bone density change range of the patient based on the change straight line and the age of the patient;

[0026] When the bone density of the patient at the adjacent node of the patient is not within the adjacent bone density variation range of the patient, the patient at the corresponding adjacent node is recorded as a suspected abnormal patient.

[0027] Furthermore, the method for obtaining the degree of bone abnormality includes:

[0028] For any suspected abnormal patient, the mean value of each blood index of all non-suspected abnormal patients in the initial age community where the suspected abnormal patient is located is used as the benchmark value of each blood index;

[0029] Obtaining a baseline deviation of each blood index of the suspected abnormal patient through the difference between each blood index of the suspected abnormal patient and the corresponding baseline value;

[0030] The degree of bone abnormality of the suspected abnormal patient is obtained by combining the baseline deviations of all blood indicators of the suspected abnormal patient.

[0031] Furthermore, the method for obtaining patients with bone abnormalities includes:

[0032] When the degree of bone abnormality of a suspected abnormal patient is greater than a preset abnormal threshold, the corresponding suspected abnormal patient is recorded as a bone abnormality patient.

[0033] Furthermore, the method for obtaining the final association weight includes:

[0034] For any patient, when the patient at the adjacent node of the patient is not a patient with bone abnormality, the initial association weight between the patient and the patient at the corresponding adjacent node is used as the final association weight between the patient and the corresponding adjacent node;

[0035] When the patient at the adjacent node of the patient is a patient with bone abnormality, the degree of bone abnormality of the corresponding patient with bone abnormality is negatively correlated to obtain the adjustment degree of the corresponding patient with bone abnormality; the product of the initial association weight between the patient and the corresponding patient with bone abnormality and the adjustment degree of the corresponding patient with bone abnormality is used as the final association weight between the patient and the corresponding adjacent node.

[0036] The present invention provides an orthopedic database construction system, the system comprising:

[0037] The data acquisition module is used to obtain the age, bone density, bone metabolism markers and blood indicators of each patient in the historical orthopedic medical data; the SLPA algorithm is used to obtain the initial age community based on the age of all patients;

[0038] An initial association analysis module is used to determine the initial association weight of each patient with each adjacent node in each initial age community according to the distribution of bone density and each bone metabolism marker between each patient and patients at each adjacent node;

[0039] The abnormal patient analysis module is used to determine suspected abnormal patients from all adjacent nodes of each patient in each initial age community based on the correlation changes between age and bone density; obtain the degree of bone abnormality of each suspected abnormal patient based on the deviation of each blood index between each suspected abnormal patient and all non-suspected abnormal patients in the initial age community; and screen out patients with bone abnormality from suspected abnormal patients based on the degree of bone abnormality;

[0040] The final association analysis module is used to obtain the final association weight of each patient with each adjacent node in the initial age community according to the initial association weight of each patient with each adjacent node and the degree of bone abnormality of the patients with bone abnormality in the adjacent node;

[0041] The database construction module is used to update the initial age community of the SLPA algorithm through the final association weight to obtain the final age community; and to construct an orthopedic database based on the final age community.

[0042] The present invention also provides a computer-readable storage medium, in which a computer program is stored. The computer program can be executed by at least one processor. When the computer program code runs on the processor of a computer, the computer executes the steps of an orthopedic database construction method as described above.

[0043] The present invention has the following beneficial effects:

[0044] The present invention considers the relevant situation of bone between similar ages to adjust the division situation, first obtains the initial age community of the patient by SLPA algorithm, and adjusts the association situation in the community detection algorithm. In the initial age community, according to the distribution of bone density and various bone metabolism markers of the patient and the patient of the adjacent node, the bone situation difference is reflected, as the initial association weight between the nodes, so as to adjust the association credibility of age by the bone situation. Further considering the association influence of the patient with abnormal bone situation on the adjacent bone association, such as the difference influence of the osteoporosis degree patient, first according to the general association change of age and bone density, screen out the suspected abnormal patient with suspected bone abnormality, and reflect the degree of abnormality of bone in the suspected abnormal patient by the difference situation in the blood index of the abnormal bone, obtain the degree of abnormal bone and then screen out the patient with abnormal bone, so that the adjacent patient with abnormal bone is adjusted more strongly in the follow-up, and the credibility of data structure division is guaranteed. Finally, the initial association weights between nodes in the initial age community are adjusted according to the degree of bone abnormality of patients with bone abnormalities to obtain the final association weights, thereby updating the initial age community of the SLPA algorithm to obtain the final age community for constructing an orthopedic database. In the process of age-based division, the present invention considers the bone differences and abnormality between orthopedic data to adaptively adjust the association weights between ages, so that the division results of the age and bone condition of orthopedic data are highly robust, the database construction is of higher quality, and the efficiency of further data analysis is improved.

[0045] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented in accordance with the contents of the specification. In order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are listed below. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] In order to more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the drawings required for use in the embodiments or the prior art descriptions are briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0047] Figure 1 A flow chart of a method for constructing an orthopedic database provided by one embodiment of the present invention;

[0048] Figure 2 A flow chart of a method for obtaining an initial association weight provided by an embodiment of the present invention;

[0049] Figure 3 A structural diagram of an orthopedic database construction system provided by one embodiment of the present invention. DETAILED DESCRIPTION

[0050] In order to further explain the technical means and effects adopted by the present invention to achieve the predetermined invention purpose, the following is a detailed description of an orthopedic database construction method, system and storage medium proposed by the present invention, its specific implementation, structure, features and effects, in conjunction with the accompanying drawings and preferred embodiments. In the following description, different "one embodiment" or "another embodiment" does not necessarily refer to the same embodiment. In addition, specific features, structures or characteristics in one or more embodiments may be combined in any suitable form.

[0051] Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.

[0052] The following is a detailed description of a method, system and storage medium for constructing an orthopedic database provided by the present invention in conjunction with the accompanying drawings.

[0053] See also Figure 1 , which shows a flow chart of a method for constructing an orthopedic database provided by an embodiment of the present invention, the method comprising the following steps:

[0054] S1: Obtain the age, bone density, bone metabolism markers and blood indicators of each patient in the historical orthopedic medical data; use the SLPA algorithm to obtain the initial age community based on the age of all patients.

[0055] Since patients of different age groups have different bone development, the corresponding treatment and prevention methods they need are also different. Therefore, in the orthopedic data structure, age can be used to divide patients into groups with similar data characteristics, thereby improving the data mining and analysis capabilities of the database. However, considering that different patients in the same age group have different bone conditions due to factors such as the growth environment, simple division is still not enough to reflect the characteristics of the sub-data groups. Therefore, further division and construction are performed based on multiple orthopedic data during analysis.

[0056] In a specific embodiment of the present invention, the hospital will perform multiple bone tests on the patient, so the age, bone density, bone metabolism markers and blood indicators of each patient can be obtained in the historical orthopedic medical data. Among them, bone density can directly characterize the patient's bone condition, and bone metabolism markers include a variety of substances, such as serum calcium, serum phosphorus, alkaline phosphatase, bone-specific alkaline phosphatase, osteocalcin, C-terminal peptide and N-terminal peptide, which can reflect bone development or bone health status. Blood indicators also include a variety of indicators, such as parathyroid hormone, prothrombin time, activated partial thromboplastin time, fibrinogen, hemoglobin, white blood cell count and red blood cell count, which can reflect the abnormality of bone quality. The specific data acquisition process will not be repeated here.

[0057] It is worth mentioning that in the specific implementation of this application, data related to patient age and medical data are involved. When the embodiments of this application are applied to specific products or technologies, the collection, use and processing of relevant data must comply with relevant laws, regulations and standards of relevant countries and regions.

[0058] The SLPA algorithm is a community discovery algorithm that reflects the close relationship between data nodes through the community, facilitates the analysis of the community structure of the data and improves the data mining ability. Communities are usually divided according to specific attributes to more accurately identify and analyze subgroups in the data set. In the embodiment of the present invention, considering that bone condition changes with age, patients are first divided into different communities according to age distribution, and the initial relationship in the community is determined by the SLPA algorithm with age proximity as the association path.

[0059] In a specific method of an embodiment of the present invention, patients are divided into five age communities at intervals of 20 years, that is, the age division range of the age community is {0-20 years old}, {21-40 years old}, {41-60 years old}, {61-80 years old} and {over 80 years old}. The implementer can make adjustments according to the specific situation. The SLPA algorithm can preliminarily determine the association path between patients in each age community to obtain the initial age community. Each node in the initial age community represents each patient, and the connection standard between nodes is that other patients whose age difference with the patient is no more than 1 year can be connected.

[0060] When the SLPA algorithm performs community updates, for the current node, it is necessary to keep the most popular label in memory, and generally, the node with the largest frequency of occurrence according to the number of adjacent nodes connected to each node is recorded as the label of the current node. For the current node, the weight of each adjacent node is consistent. However, for the orthopedic community, due to the difference in bone conditions between different adjacent nodes and the current node, even if the age is similar, the adjacent node intimacy of the current node is different, so when performing community update adjustment, the degree of association weight between nodes is considered simultaneously. It should be noted that the SLPA algorithm is a public technical means known to technicians in this neighborhood, and will not be further elaborated here.

[0061] S2: In each initial age community, the initial association weight between each patient and each adjacent node is determined according to the distribution of bone density and each bone metabolism marker between each patient and each adjacent node.

[0062] In general, people of similar age have similar bone density. However, due to development and other issues, although bone density may be similar, there are differences in bone metabolism markers, resulting in different bone health conditions. For example, bone-specific alkaline phosphatase and osteocalcin are usually related to bone formation, while C-terminal peptide and N-terminal peptide are related to bone resorption. The higher the difference in bone metabolism markers, the different degrees of bone health and bone loss risk in patients. It can also be understood that the deviations of different bone metabolism markers reflect the different development conditions of individual bones. Therefore, considering the influence of bone metabolism markers while considering bone density, the initial correlation weight is obtained to comprehensively reflect the similarity of bones between patients.

[0063] Preferably, in some possible implementations of the embodiments of the present invention, the method for obtaining the initial association weight includes:

[0064] First, considering that the degree of correlation between various bone metabolism markers and bone density changes varies, the degree of difference in the impact is also different. The correlation between each bone metabolism marker and bone density in all patients is analyzed, and the importance of the impact of each bone metabolism marker is determined based on the correlation between each bone metabolism marker and bone density in different patients.

[0065] Preferably, in some implementations of the embodiments of the present invention, the method for obtaining the influence importance of each bone metabolism marker includes:

[0066] Obtain a bone density sequence and a bone metabolism marker sequence for each bone metabolism marker. Arrange all patients in order of bone density from small to large to obtain a patient sequence; arrange the bone density corresponding to the patients in the order of the patient sequence to obtain a bone density sequence; arrange each bone metabolism marker corresponding to the patient in the order of the patient sequence to obtain a bone metabolism marker sequence for each bone metabolism marker. When analyzing bone density and bone metabolism markers, analyze the degree of correlation by approximating the trend of change. This embodiment analyzes the degree of correlation by the change of each bone metabolism marker when the bone density of all patients increases. In other embodiments of the present invention, the degree of correlation may also be distributed in a decreasing manner, which will not be elaborated here.

[0067] According to the degree of correlation change between the bone metabolism marker sequence and the bone density sequence of each bone metabolism marker, the change correlation of each bone metabolism marker is obtained. In the embodiment of the present invention, the Pearson correlation coefficient between the bone metabolism marker sequence and the bone density sequence of each bone metabolism marker can be calculated and the absolute value can be taken to obtain the change correlation of each bone metabolism marker. The greater the change correlation, the more correlation there is between the bone metabolism marker sequence and the bone density sequence.

[0068] In other embodiments of the present invention, the correlation between two sets of data sequences can also be obtained through Manhattan distance or dynamic time warping algorithm. It should be noted that Pearson correlation coefficient, Manhattan distance or dynamic time warping algorithm are technical means well known to technicians in this field and will not be elaborated here.

[0069] The ratio of the change correlation of each bone metabolism marker to the maximum change correlation among all bone metabolism markers is used as the influence importance of each bone metabolism marker. In the embodiment of the present invention, the influence of each bone metabolism marker is reflected by the maximum value normalization.

[0070] At this point, the importance of the impact of each bone metabolism marker has been obtained.

[0071] Then, the association weights between two adjacent nodes were analyzed comprehensively based on the differences in bone density and the differences and influences between bone metabolism markers. In each initial age community, the initial association weights between each patient and each adjacent node were determined based on the differences in bone density between each patient and each adjacent node, as well as the differences and influence importance of each bone metabolism marker.

[0072] Preferably, in some implementations of the embodiments of the present invention, the method for obtaining the initial association weight is as follows: Figure 2 , which shows a flow chart of a method for obtaining an initial association weight provided by an embodiment of the present invention, the method comprising the following steps:

[0073] S201: According to the difference degree of each bone metabolism marker between each patient and the patients of each adjacent node, and the influence importance of each bone metabolism marker, the influence degree of each bone metabolism marker is obtained; the influence degree of all bone metabolism markers between each patient and the patients of each adjacent node is combined to obtain the difference degree of influence between each patient and the patients of each adjacent node.

[0074] The impact importance can be used as the confidence weight when analyzing the differences in bone metabolism markers to adjust the analysis of the difference impact. The higher the correlation of the bone metabolism markers, the higher the degree of participation of the corresponding bone metabolism markers in the process of bone density changes, and the higher the credibility of the analysis of the difference impact.

[0075] Therefore, in some embodiments of the present invention, the numerical difference of each bone metabolism marker between each patient and each adjacent node is used as the influence deviation of each bone metabolism marker, and the deviation of the bone metabolism marker reflects the different bone development conditions that may exist. The product of the influence importance of each bone metabolism marker and the influence deviation is used as the influence of each bone metabolism marker between each patient and each adjacent node. The credibility of the influence deviation is adjusted by the influence importance to obtain the influence of each bone metabolism marker difference. The higher the influence, the greater the influence between nodes is affected by the difference of bone metabolism markers.

[0076] In some embodiments of the present invention, all bone metabolism markers are comprehensively considered, and the cumulative value of the influence of all bone metabolism markers is used as the difference in influence between each patient and each patient of the adjacent node, reflecting the overall difference in influence of the bone metabolism markers.

[0077] In other embodiments of the present invention, the average of the influence of all bone metabolism markers may be used as the difference in influence between each patient and each adjacent node patient to characterize the overall influence, which will not be elaborated here.

[0078] S202: The difference in bone density between each patient and each patient at an adjacent node is used as the bone density deviation between each patient and each patient at an adjacent node.

[0079] The difference in bone density can directly reflect the difference in bone quality between patients at different nodes. In some embodiments of the present invention, the absolute value of the difference in bone density is used as the bone density deviation between each patient and each patient at adjacent nodes. The greater the bone density deviation, the greater the difference in bone quality between patients at different nodes, and the lower the similarity of patients.

[0080] S203: According to the influence difference degree and bone density deviation between each patient and each adjacent node, the initial association weight between each patient and each adjacent node is obtained; the influence difference degree and bone density deviation are both negatively correlated with the initial association weight.

[0081] Finally, the difference in bone condition reflected by bone density deviation is adjusted through the influence reflected by bone metabolism markers to obtain a more accurate correlation. When the bone density deviation between adjacent nodes is smaller and the degree of difference in the influence of bone metabolism markers is also smaller, it means that the bone condition and potential bone development between patients are more similar, and the degree of correlation between patients is higher at this time.

[0082] In some embodiments of the present invention, the inverse proportional value of the multiplication of the influence difference degree and the bone density deviation of each patient and each adjacent node is used as the initial association weight of each patient and each adjacent node, and the inverse proportion reflects that the influence difference degree and the bone density deviation are negatively correlated with the initial association weight. In other embodiments of the present invention, other basic mathematical operations can also be used to reflect that the influence difference degree and the bone density deviation are negatively correlated with the initial association weight, such as a negative exponential function or subtraction, etc., which are not limited here.

[0083] At this point, the weight allocation of the association between each patient and adjacent nodes in each initial age community has been preliminarily completed based on bone conditions.

[0084] S3: In each initial age community, based on the changes in the correlation between age and bone density, suspected abnormal patients are determined from all adjacent nodes of each patient; the degree of bone abnormality of each suspected abnormal patient is obtained based on the deviation of each blood indicator between each suspected abnormal patient and all non-suspected abnormal patients in the initial age community; patients with bone abnormality are screened out from suspected abnormal patients based on the degree of bone abnormality.

[0085] When comprehensively considering the differences in bone conditions and potential developmental conditions between people of similar ages, it is often necessary to pay attention to patients with bone abnormalities. These patients are quite different from other people in the community, and their correlation requires higher adjustments to reduce the impact of patients with bone abnormalities on similar data structures.

[0086] In the same initial age community, the state of bone density changes is usually similar. For example, between 0-20 years old, bone density usually grows rapidly. Between 21-40 years old, the current bone density growth begins to slow down and usually reaches a peak. Between 41-60 years old, bone density begins to gradually decline, but the speed is relatively slow. Between 61-80 years old, bone density will accelerate the downward trend. When you are over 80 years old, bone density will slowly decline to a stable lower value. Therefore, if there are patients with similar age but large differences in bone density, it means that this patient is very likely due to diseases such as osteoporosis or bone inflammation. Compared with bone fractures caused by bone trauma, the degree of difference in these patients needs to be further adjusted. Therefore, first screen out suspected abnormal patients based on the changes in bone density in each initial age community.

[0087] Preferably, in some possible implementations of the embodiments of the present invention, the method for obtaining suspected abnormal patients includes:

[0088] For any initial age community, all patients in the initial age community are mapped to the bone density change coordinate system to obtain the change coordinate system of the initial age community. The change correlation between age and bone density is reflected by the coordinate distribution. The horizontal axis of the bone density change coordinate system is age, and the vertical axis is bone density. It should be noted that the coordinate system mapping is a technical means well known to technicians in this field and will not be elaborated here.

[0089] Further, the data of the initial age community in the change coordinate system is linearly fitted to obtain the change straight line of the initial age community. Since bone density and age growth are both linearly related, the bone density and age changes are generally consistent. In the embodiment of the present invention, the least squares method can be selected for linear fitting, which is not limited here.

[0090] For any patient in the initial age community, the adjacent bone density variation range of the patient is determined based on the variation line and the age of the patient. In the embodiment of the present invention, since the age difference between each patient and the patient at the adjacent node does not exceed 1 year, the distribution degree of the patients at the adjacent node in terms of bone density can be determined based on the age of each patient. The bone density corresponding to the patient's age when he is 1 year younger on the variation line is used as the lowest value of the adjacent bone density variation range of the patient, and the bone density corresponding to the patient's age when he is 1 year older is used as the highest value of the adjacent bone density variation range of the patient to determine the adjacent bone density variation range. In other embodiments of the present invention, an error range value can also be set, and the difference between the bone density corresponding to the patient's age when he is 1 year younger on the variation line and the error range value is used as the lowest value, and the sum of the bone density corresponding to the patient's age when he is 1 year older on the variation line and the error range value is used as the highest value, so as to increase the fault tolerance range and reduce the subsequent analysis rate. The setting of the error range value can be adjusted according to the specific implementation situation.

[0091] When the bone density of the patient at the adjacent node of the patient is not within the range of the adjacent bone density of the patient, it indicates that the patient at the corresponding adjacent node may have abnormal bone density, and the patient at the corresponding adjacent node is recorded as a suspected abnormal patient.

[0092] At this point, the preliminary screening of suspected abnormal patients is completed.

[0093] Further consideration is given to the changes in blood indicators caused by the disease. For example, osteoporosis patients may have abnormal coagulation function, which is manifested by abnormal changes in blood indicators such as prothrombin time, activated partial thromboplastin time and fibrinogen, and the changes in hemoglobin, white blood cell count and platelet count in blood indicators are quite different. At the same time, because parathyroid hormone can increase bone absorption and release calcium into the blood to maintain blood calcium levels, the parathyroid hormone level of osteoporosis patients may also increase. Similarly, if the patient's bones are accompanied by chronic inflammation or infection, it will lead to an increase in white blood cell count, etc. Therefore, the degree of blood indicators of the screened suspected abnormal patients is further analyzed to obtain the possible degree of bone abnormality of each suspected abnormal patient.

[0094] Preferably, in some possible implementations of the embodiments of the present invention, the method for obtaining the degree of bone abnormality includes:

[0095] For any suspected abnormal patient, the mean value of each blood indicator of all non-suspected abnormal patients in the initial age community of the suspected abnormal patient is used as the baseline value of each blood indicator, and the blood indicators of patients who have not been screened out are used as the baseline for judgment. The greater the deviation from the baseline value, the higher the possibility of the presence of the disease.

[0096] Further, by the difference between each blood index of the suspected abnormal patient and the corresponding baseline value, the baseline deviation of each blood index of the suspected abnormal patient is obtained. For each blood index, the higher the deviation, the more abnormal the situation. Finally, the baseline deviation of all blood indexes of the suspected abnormal patient is combined to obtain the degree of bone abnormality of the suspected abnormal patient. In the embodiment of the present invention, the accumulated values ​​of all baseline deviations are normalized to obtain the degree of bone abnormality, which comprehensively reflects the difference in blood indexes. It should be noted that normalization is a technical means well known to those skilled in the art. The selection of the normalization function can be linear normalization or standard normalization, etc. The specific normalization method is not limited here.

[0097] In other embodiments of the present invention, the baseline deviations of different blood indicators may be weighted and averaged to obtain the degree of bone abnormality in the suspected abnormal patient. For example, when focusing on the abnormality of osteoporosis patients, a higher weight may be given to the baseline deviation of parathyroid hormone, which is a more significant blood indicator. The implementer may adjust the weights by himself, and there is no limitation here.

[0098] Finally, patients with high bone abnormality characteristics are screened out by the degree of bone abnormality. In the association analysis, it is necessary to further reduce similar situations. In the embodiment of the present invention, when the degree of bone abnormality of the suspected abnormal patient is greater than the preset abnormal threshold, the corresponding suspected abnormal patient is recorded as a bone abnormal patient. In a specific implementation method of the embodiment of the present invention, the preset abnormal threshold is set to 0.3, and the specific value can be adjusted by the implementer according to the specific implementation situation, and is not limited here.

[0099] At this point, the screening of patients with bone abnormalities is completed.

[0100] S4: In the initial age community, the final association weight between each patient and each adjacent node is obtained according to the initial association weight between each patient and each adjacent node and the degree of bone abnormality of patients with bone abnormality in the adjacent node.

[0101] The initial association weights were further adjusted for the patients with bone abnormalities who were comprehensively screened to make the relationship of the orthopedic data structure more accurate. The initial association weights were obtained by considering the differences in bone quality between patients of similar age, and the association of patients with bone abnormalities was further reduced to obtain the final association weights.

[0102] In some possible implementations of the present invention, for any patient, when the patient at the adjacent node of the patient is not a patient with bone abnormalities, the initial association weight between the patient and the patient at the corresponding adjacent node is used as the final association weight between the patient and the corresponding adjacent node, and no further adjustment is required.

[0103] When the patient of the adjacent node of the patient is a patient with bone abnormality, the degree of bone abnormality of the corresponding patient with bone abnormality is negatively correlated to obtain the adjustment degree of the corresponding patient with bone abnormality. The greater the degree of bone abnormality, the higher the possible abnormality and the lower the degree of correlation. The negative correlation mapping can be in the form of inverse proportion or negative exponential power, etc., which is not limited here.

[0104] The product of the initial association weight between the patient and the corresponding bone abnormality patient and the adjustment degree of the corresponding bone abnormality patient is used as the final association weight between the patient and the corresponding adjacent node, and the association weight is further adjusted to improve the difference between bone abnormality patients.

[0105] S5: Update the initial age community of the SLPA algorithm through the final association weight to obtain the final age community; construct an orthopedic database based on the final age community.

[0106] After obtaining the final association weights between the patient and the adjacent nodes of each node, the node relationship is iteratively updated through the SLPA algorithm to obtain the final community division result as the final age community, considering the factors affecting the characteristics of bone development with age, and further adjusting in combination with individual bone development differences or abnormalities to obtain the final community division. The final age community can be used to classify and construct the patient's orthopedic data when the database is constructed, so as to facilitate subsequent data query, learning and analysis based on the orthopedic database, etc., which will not be elaborated here.

[0107] It should be noted that, in order to facilitate calculation, all data involved in the calculation in the embodiment of the present invention are preprocessed to eliminate the dimension effect. The specific means of eliminating the dimension effect are technical means well known to those skilled in the art and are not limited here.

[0108] In summary, the present invention considers the relevant situation of bone between similar ages to adjust the division situation, first obtains the initial age community of the patient by SLPA algorithm, and adjusts the association situation in the community detection algorithm. In the initial age community, according to the distribution of the patient and the patient of the adjacent node on bone density and various bone metabolism markers, the bone situation difference is reflected, as the initial association weight between the nodes, to adjust the association credibility of age by the bone situation. Further considering the patient of the abnormal bone situation on the association influence between adjacent bone associations, such as the difference influence of the osteoporosis degree patient, first according to the general association change of age and bone density, screen out the suspected abnormal patient suspected of having bone abnormality, by the difference situation of bone abnormality in blood index, reflect the degree of abnormality of bone in the suspected abnormal patient, obtain the degree of bone abnormality and then screen out the patient with bone abnormality, so that the adjacent patient with bone abnormality is adjusted more strongly in the follow-up, and the credibility of data structure division is guaranteed. Finally, the initial association weights between nodes in the initial age community are adjusted according to the degree of bone abnormality of patients with bone abnormalities to obtain the final association weights, thereby updating the initial age community of the SLPA algorithm to obtain the final age community for constructing an orthopedic database. In the process of age-based division, the present invention considers the bone differences and abnormality between orthopedic data to adaptively adjust the association weights between ages, so that the division results of the age and bone condition of orthopedic data are highly robust, the database construction is of higher quality, and the efficiency of further data analysis is improved.

[0109] The embodiment of the present invention provides an orthopedic database construction system, please refer to Figure 3, which shows a structural diagram of an orthopedic database construction system provided by an embodiment of the present invention, the system includes: a data acquisition module 701, an initial association analysis module 702, an abnormal patient analysis module 703, a final association analysis module 704 and a database construction module 705.

[0110] The data acquisition module 701 is used to obtain the age, bone density, bone metabolism markers and blood indicators of each patient in the historical orthopedic medical data; and to obtain the initial age community based on the age of all patients using the SLPA algorithm;

[0111] The initial association analysis module 702 is used to determine the initial association weight of each patient and each adjacent node in each initial age community according to the distribution of bone density and each bone metabolism marker between each patient and each adjacent node;

[0112] The abnormal patient analysis module 703 is used to determine suspected abnormal patients from all adjacent nodes of each patient in each initial age community based on the correlation change between age and bone density; obtain the degree of bone abnormality of each suspected abnormal patient based on the deviation of each blood index between each suspected abnormal patient and all non-suspected abnormal patients in the initial age community; and screen out patients with bone abnormality from the suspected abnormal patients based on the degree of bone abnormality;

[0113] The final association analysis module 704 is used to obtain the final association weight between each patient and each adjacent node in the initial age community according to the initial association weight between each patient and each adjacent node and the degree of bone abnormality of the bone abnormality patient in the adjacent node;

[0114] The database construction module 705 is used to update the initial age community of the SLPA algorithm through the final association weight to obtain the final age community; and to construct an orthopedic database based on the final age community.

[0115] It should be noted that the system provided in the above embodiment is only illustrated by the division of the above functional modules. In actual application, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the computer device can be divided into different functional modules to complete all or part of the functions described above. Since the specific implementation process of an orthopedic database construction system in this embodiment is the same as the specific implementation process of an orthopedic database construction method described above, it will not be repeated here.

[0116] The present invention also provides a computer-readable storage medium, in which a computer program is stored. The computer program can be executed by at least one processor. When the computer program code runs on the processor of a computer, the computer executes the steps of an orthopedic database construction method as described above.

[0117] It should be noted that the sequence of the above 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 accompanying 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.

[0118] The various embodiments in this specification are described in a progressive manner, and the same or similar parts between the various embodiments can be referenced to each other, and each embodiment focuses on the differences from other embodiments.

Claims

1. A method for constructing an orthopedic database, characterized in that: The method comprises: Obtain each patient's age, bone density, bone metabolism markers, and blood indicators from historical orthopedic medical data; use the SLPA algorithm to obtain the initial age community based on the age of all patients; In each initial age community, the initial association weight between each patient and each adjacent node is determined according to the distribution of bone density and each bone metabolism marker between each patient and each adjacent node; In each initial age community, based on the changes in the correlation between age and bone density, suspected abnormal patients are identified from all adjacent nodes of each patient; the degree of bone abnormality of each suspected abnormal patient is obtained based on the deviation of each suspected abnormal patient from all non-suspected abnormal patients in the initial age community in each blood index; patients with bone abnormality are screened out from suspected abnormal patients based on the degree of bone abnormality; In the initial age community, the final association weight of each patient with each adjacent node is obtained according to the initial association weight of each patient with each adjacent node and the degree of bone abnormality of patients with bone abnormality in the adjacent node; By using the final association weight, the initial age community of the SLPA algorithm is updated to obtain the final age community; an orthopedic database is constructed based on the final age community; The method for obtaining the initial association weight includes: According to the correlation between each bone metabolism marker and bone density in different patients, the importance of the impact of each bone metabolism marker was determined; In each initial age community, the initial association weight between each patient and each adjacent node is determined based on the difference in bone density between each patient and the patients at each adjacent node, as well as the difference and impact importance of each bone metabolism marker; The method for obtaining the impact importance includes: Arrange all patients in order of bone density from small to large to obtain a patient sequence; arrange the bone density corresponding to the patients in the order of the patient sequence to obtain a bone density sequence; arrange each bone metabolism marker corresponding to the patients in the order of the patient sequence to obtain a bone metabolism marker sequence of each bone metabolism marker; According to the correlation change degree between the bone metabolism marker sequence and the bone density sequence of each bone metabolism marker, the change correlation of each bone metabolism marker is obtained; The ratio of the change correlation of each bone metabolism marker to the maximum change correlation of all bone metabolism markers was taken as the impact importance of each bone metabolism marker; Determining the initial association weight between each patient and each adjacent node according to the difference in bone density between each patient and each adjacent node, as well as the difference and impact importance of each bone metabolism marker, includes: According to the difference between each patient and the patients of each adjacent node in each bone metabolism marker, and the importance of the influence of each bone metabolism marker, the influence of each bone metabolism marker is obtained; The influence degree of all bone metabolism markers of each patient and the patients of each adjacent node is combined to obtain the difference degree of influence between each patient and the patients of each adjacent node; The difference in bone density between each patient and the patient at each adjacent node was taken as the bone density deviation between each patient and the patient at each adjacent node; According to the difference in influence and the bone density deviation between each patient and each adjacent node, the initial association weight between each patient and each adjacent node is obtained; the difference in influence and the bone density deviation are both negatively correlated with the initial association weight; The method for obtaining the final association weight includes: For any patient, when the patient at the adjacent node of the patient is not a patient with bone abnormality, the initial association weight between the patient and the patient at the corresponding adjacent node is used as the final association weight between the patient and the corresponding adjacent node; When the patient at the adjacent node of the patient is a patient with bone abnormality, the degree of bone abnormality of the corresponding patient with bone abnormality is negatively correlated to obtain the adjustment degree of the corresponding patient with bone abnormality; the product of the initial association weight between the patient and the corresponding patient with bone abnormality and the adjustment degree of the corresponding patient with bone abnormality is used as the final association weight between the patient and the corresponding adjacent node.

2. The method for constructing an orthopedic database according to claim 1, characterized in that: The method for obtaining the suspected abnormal patient comprises: For any initial age community, all patients in the initial age community are mapped to the bone density change coordinate system to obtain the change coordinate system of the initial age community; the horizontal axis of the bone density change coordinate system is age, and the vertical axis is bone density; Performing straight-line fitting on the data of the initial age community in the change coordinate system to obtain a change straight line of the initial age community; for any patient in the initial age community, determining the adjacent bone density change range of the patient based on the change straight line and the age of the patient; When the bone density of the patient at the adjacent node of the patient is not within the adjacent bone density variation range of the patient, the patient at the corresponding adjacent node is recorded as a suspected abnormal patient.

3. The method for constructing an orthopedic database according to claim 1, characterized in that: The method for obtaining the degree of bone abnormality includes: For any suspected abnormal patient, the mean value of each blood index of all non-suspected abnormal patients in the initial age community where the suspected abnormal patient is located is used as the benchmark value of each blood index; Obtaining a baseline deviation of each blood index of the suspected abnormal patient through the difference between each blood index of the suspected abnormal patient and the corresponding baseline value; The degree of bone abnormality of the suspected abnormal patient is obtained by combining the baseline deviations of all blood indicators of the suspected abnormal patient.

4. The method for constructing an orthopedic database according to claim 1, characterized in that: The method for obtaining the patient with bone abnormality comprises: When the degree of bone abnormality of a suspected abnormal patient is greater than a preset abnormal threshold, the corresponding suspected abnormal patient is recorded as a bone abnormality patient.

5. An orthopedic database construction system, characterized in that: The system comprises: The data acquisition module is used to obtain the age, bone density, bone metabolism markers and blood indicators of each patient in the historical orthopedic medical data; the SLPA algorithm is used to obtain the initial age community based on the age of all patients; An initial association analysis module is used to determine the initial association weight of each patient with each adjacent node in each initial age community according to the distribution of bone density and each bone metabolism marker between each patient and patients at each adjacent node; The method for obtaining the initial association weight includes: According to the correlation between each bone metabolism marker and bone density in different patients, the importance of the impact of each bone metabolism marker was determined; In each initial age community, the initial association weight between each patient and each adjacent node is determined based on the difference in bone density between each patient and the patients at each adjacent node, as well as the difference and impact importance of each bone metabolism marker; The method for obtaining the impact importance includes: Arrange all patients in order of bone density from small to large to obtain a patient sequence; arrange the bone density corresponding to the patients in the order of the patient sequence to obtain a bone density sequence; arrange each bone metabolism marker corresponding to the patients in the order of the patient sequence to obtain a bone metabolism marker sequence of each bone metabolism marker; According to the correlation change degree between the bone metabolism marker sequence and the bone density sequence of each bone metabolism marker, the change correlation of each bone metabolism marker is obtained; The ratio of the change correlation of each bone metabolism marker to the maximum change correlation of all bone metabolism markers was taken as the impact importance of each bone metabolism marker; Determining the initial association weight between each patient and each adjacent node according to the difference in bone density between each patient and each adjacent node, as well as the difference and impact importance of each bone metabolism marker, includes: According to the difference between each patient and the patients of each adjacent node in each bone metabolism marker, and the importance of the influence of each bone metabolism marker, the influence of each bone metabolism marker is obtained; The influence degree of all bone metabolism markers of each patient and the patients of each adjacent node is combined to obtain the difference degree of influence between each patient and the patients of each adjacent node; The difference in bone density between each patient and the patient at each adjacent node was taken as the bone density deviation between each patient and the patient at each adjacent node; According to the difference in influence and the bone density deviation between each patient and each adjacent node, the initial association weight between each patient and each adjacent node is obtained; the difference in influence and the bone density deviation are both negatively correlated with the initial association weight; The abnormal patient analysis module is used to determine suspected abnormal patients from all adjacent nodes of each patient in each initial age community based on the correlation changes between age and bone density; obtain the degree of bone abnormality of each suspected abnormal patient based on the deviation of each blood index between each suspected abnormal patient and all non-suspected abnormal patients in the initial age community; and screen out patients with bone abnormality from suspected abnormal patients based on the degree of bone abnormality; A final association analysis module is used to obtain a final association weight between each patient and each adjacent node in the initial age community according to the initial association weight between each patient and each adjacent node and the degree of bone abnormality of patients with bone abnormality in the adjacent node; The method for obtaining the final association weight includes: For any patient, when the patient at the adjacent node of the patient is not a patient with bone abnormality, the initial association weight between the patient and the patient at the corresponding adjacent node is used as the final association weight between the patient and the corresponding adjacent node; When the patient at the adjacent node of the patient is a patient with bone abnormality, the degree of bone abnormality of the corresponding patient with bone abnormality is negatively correlated to obtain the adjustment degree of the corresponding patient with bone abnormality; the product of the initial association weight between the patient and the corresponding patient with bone abnormality and the adjustment degree of the corresponding patient with bone abnormality is used as the final association weight between the patient and the corresponding adjacent node; The database construction module is used to update the initial age community of the SLPA algorithm through the final association weight to obtain the final age community; and to construct an orthopedic database based on the final age community.

6. A computer-readable storage medium, wherein a computer program is stored in the computer-readable storage medium, wherein the computer program can be executed by at least one processor, and when the computer program code runs on the processor of a computer, the computer executes the steps of the orthopedic database construction method as described in any one of claims 1 to 4.

Images