An auxiliary screening system and method for osteoporosis in elderly women

Abstract

The present invention relates to the field of medical care informatics technology, and in particular to a system and method for assisting elderly women with osteoporosis, comprising: obtaining the menopausal time and age of each elderly woman, dividing the clusters by clustering, analyzing the volatility of the menopausal time change of the elderly women of each cluster, the proximity concentration of the elderly women's internal distribution, and the uniformity of the elderly women's data, thereby determining the actual expansion demand of each cluster, thereby expanding the elderly women's data, and building a big data bone detection model for monitoring whether elderly women have osteoporosis. The present invention divides the clusters by clustering the menopausal time and age of elderly women, analyzing and determining the actual expansion demand of each cluster, thereby expanding each cluster, and building a more complete big data bone detection model with the expanded data, thereby improving the accuracy of the auxiliary detection of elderly women with osteoporosis.

Description

Technical Field

[0001] The present invention relates to the technical field of medical care informatics, and in particular to an auxiliary screening system and method for osteoporosis in elderly women. Background Art

[0002] In today's digital healthcare era, big data and advanced machine learning technologies can provide powerful support for osteoporosis risk assessment and management in older women. Specifically, by collecting and analyzing large amounts of bone density measurement data, a universal bone density change model can be trained. This model can accurately predict osteoporosis in older women based on their historical data characteristics.

[0003] When building bone density models for older women, this method relies on individual, multi-word bone density measurement data. However, since most women do not regularly test their bone density, the data sample is insufficient, affecting the accuracy of the bone density model. Therefore, it is necessary to expand the data based on the current distribution of older women, fill in the gaps in individual data, and improve the model's training dataset. However, since auxiliary screening for osteoporosis in older women is closely related to the time of menopause and also needs to consider the age changes of different older women, the bone density model built from big data of older women is inaccurate and difficult to fully reflect the osteoporosis status of older women. Summary of the Invention

[0004] The present invention provides an auxiliary screening system and method for osteoporosis in elderly women to solve the existing problems.

[0005] The present invention adopts the following technical solutions:

[0006] One embodiment of the present invention provides an auxiliary screening method for osteoporosis in elderly women, the method comprising the following steps:

[0007] Obtain the menopausal time and age of each elderly woman;

[0008] A scatter plot was constructed with each elderly woman as a data point, and menopausal time and age as the horizontal and vertical axes, respectively. All data points in the scatter plot were clustered to obtain several clusters. The volatility of menopausal time changes in elderly women in each cluster was determined based on the number of elderly women in each cluster, the slope of the fitted line of all data points in the scatter plot, and the slope of the fitted line of all data points in each cluster.

[0009] In each cluster, several elderly women with the closest neighbors of each elderly woman were obtained; based on the distance between elderly women in each cluster and their closest neighbors, the proximity concentration of elderly women within each cluster was determined; based on the number of times each elderly woman was a close neighbor of other elderly women, the uniformity of elderly women data in each cluster was determined;

[0010] The actual expansion requirement of each cluster is determined based on the number of elderly women in each cluster, the proximity concentration of the elderly women's internal distribution, the uniformity of the elderly women's data, and the volatility of the menopausal time of elderly women. Data expansion is performed on each cluster based on the actual expansion requirement, and a big data bone detection model is constructed with the expanded data to monitor whether elderly women have osteoporosis.

[0011] Furthermore, the method of determining the volatility of menopausal timing of the elderly women in each cluster based on the number of elderly women included in each cluster and the slope of the fitted line of all data points in the scatter plot and the slope of the fitted line of all data points in each cluster includes the following specific steps:

[0012] For all data points in each cluster, a linear fit is performed using the least squares method to obtain the slope of the straight line in each cluster;

[0013] For all data points in the scatter plot, a straight line is fitted using the least squares method to obtain the slope of the overall straight line;

[0014] The first The minimum bounding rectangle area of ​​the cluster is the same as the The number of elderly women in each cluster is compared to determine the The distribution looseness of the elderly women in each cluster;

[0015] The first The slope of the straight line in each cluster is subtracted from the slope of the overall straight line corresponding to all elderly women and the absolute value is taken to determine the first The deviation between the menopausal timing and age regularity of the elderly women in each cluster;

[0016] According to The looseness of the distribution of elderly women in each cluster and the deviation of the relationship between the menopausal time and age of elderly women are used to determine the Fluctuations in menopausal timing among older women in three clusters.

[0017] Furthermore, according to The looseness of the distribution of elderly women in each cluster and the deviation of the relationship between the menopausal time and age of elderly women are used to determine the The fluctuation of menopausal timing in each cluster of elderly women includes the following specific steps:

[0018] The first The looseness of the distribution of elderly women in the first cluster is related to the The normalized value of the deviation between the menopausal time and the age regular relationship of the elderly women in the cluster is multiplied as the first Fluctuations in menopausal timing among older women in three clusters.

[0019] Furthermore, the method of determining the proximity concentration of the elderly women within each cluster based on the distance between the elderly women in each cluster and their nearest elderly women includes the following specific steps:

[0020] In the In the clusters, get the An elderly woman and The first elderly women The Euclidean distance of the nearest elderly female An elderly woman and The sum of the Euclidean distances of all the nearest elderly women of an elderly woman is taken as the first The target distance of the elderly women is taken as the average of the target distances of all elderly women. The internal distribution of elderly women in each cluster is close to the concentration.

[0021] Furthermore, the method of determining the uniformity of the data of elderly women in each cluster according to the number of times each elderly woman is a neighboring elderly woman of other elderly women includes the following specific steps:

[0022] Get the Within a cluster The number of times an elderly woman is the nearest elderly woman of other elderly women is recorded as The first cluster The number of times an elderly woman belongs to the adjacent range;

[0023] Get the The average number of times all elderly women in the cluster belong to the adjacent range is recorded as The average number of times the elderly women in each cluster belong to the adjacent range;

[0024] The first The first cluster The number of times an elderly woman belongs to the adjacent range is the same as the The average number of times the elderly women in each cluster belong to the adjacent range is subtracted and the absolute value is taken, which is recorded as The first cluster The deviation of the number of times of close range of elderly women;

[0025] The first The deviation of the number of adjacent ranges of all senior women in the cluster is accumulated and summed up to be the first The distribution deviation of elderly women in each cluster;

[0026] According to The distribution deviation of elderly women in the cluster is determined The isolation degree of the distribution of elderly women in each cluster;

[0027] According to The distribution deviation and isolation of elderly women in each cluster are used to determine the The uniformity of data of elderly women in each cluster.

[0028] Furthermore, according to The distribution deviation of elderly women in the cluster is determined The isolation degree of the elderly female distribution of each cluster is calculated by the following specific steps:

[0029] In the In the cluster, the number of elderly women with the number of times in the adjacent range of 0 is Compare the number of elderly women in each cluster to determine the The isolation degree of the distribution of elderly women in each cluster.

[0030] Furthermore, according to The distribution deviation and isolation of elderly women in each cluster are used to determine the The specific steps for measuring the uniformity of the data of elderly women in each cluster are as follows:

[0031] The first The distribution deviation of the elderly women in the first cluster is The inverse proportional normalized value of the multiplication of the isolation degree of the elderly female distribution of the clusters is obtained The uniformity of data of elderly women in each cluster.

[0032] Furthermore, the actual expansion demand of each cluster is determined based on the number of elderly women in each cluster, the proximity concentration within the elderly women, the uniformity of elderly women's data, and the volatility of menopausal changes in elderly women. The specific steps include the following:

[0033] Compare the number of elderly women in each cluster with the maximum number of elderly women in all clusters to determine the initial expansion demand of each cluster;

[0034] Multiply the proximity concentration of the internal distribution of elderly women in each cluster and the uniformity of elderly women data, and record it as the expansion resistance of each cluster;

[0035] The expansion adjustment coefficient of each cluster is determined by comparing the normalized value of the fluctuation of menopausal time of elderly women in each cluster with the expansion resistance of each cluster;

[0036] The actual expansion requirement of each cluster is determined according to the expansion adjustment coefficient of each cluster.

[0037] Furthermore, the actual expansion requirement of each cluster is determined according to the expansion adjustment coefficient of each cluster, including the following specific steps:

[0038] The initial expansion requirement of each cluster is multiplied by the expansion adjustment coefficient of each cluster, and the result is added to the initial expansion requirement of each cluster to determine the actual expansion requirement of each cluster.

[0039] The present invention also proposes an auxiliary screening system for osteoporosis in elderly women, comprising a memory, a processor, and a computer program stored in the memory and runnable on the processor. The processor executes the computer program stored in the memory to implement the steps of the aforementioned auxiliary screening method for osteoporosis in elderly women.

[0040] The beneficial effects of the technical solution of the present invention are as follows: the present invention constructs age-menopause time coordinates, divides data of different elderly women into multiple clusters, determines the changes in age-menopause time of women in each cluster, and further comprehensively determines the actual expansion demand of each cluster based on the proximity concentration of the internal distribution of elderly women in each cluster, the uniformity of elderly women data, etc., to expand the data through the SMOTE algorithm, thereby establishing a big data bone detection model with more complete data, thereby improving the accuracy of auxiliary detection of osteoporosis in elderly women. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. 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 any creative work.

[0042] Figure 1 This is a flowchart of the steps of an auxiliary screening method for osteoporosis in elderly women according to the present invention;

[0043] Figure 2 This is a schematic diagram of a scatter plot of menopausal time and age in elderly women;

[0044] Figure 3 This is a schematic diagram of the clustering results for elderly women;

[0045] Figure 4 A schematic diagram of the relationship between menopausal time and age for each cluster of elderly women;

[0046] Figure 5 Schematic diagram of the relationship between menopausal time and age for all elderly women. DETAILED DESCRIPTION

[0047] To further illustrate the technical means and efficacy of the present invention to achieve the intended purpose, the following, in conjunction with the accompanying drawings and preferred embodiments, describes in detail the specific implementation, structure, features, and efficacy of an auxiliary screening system and method for osteoporosis in elderly women proposed by the present invention. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. In addition, specific features, structures, or characteristics of one or more embodiments may be combined in any suitable form.

[0048] 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.

[0049] The following describes in detail a specific scheme of an auxiliary screening system and method for osteoporosis in elderly women provided by the present invention with reference to the accompanying drawings.

[0050] See also Figure 1 , which shows a flowchart of a method for assisting in screening osteoporosis in elderly women according to an embodiment of the present invention, the method comprising the following steps:

[0051] Step S001: Obtain the menopausal time and age of each elderly woman.

[0052] Older women usually refer to women aged 65 and above. Currently, some hospitals and research institutions have established osteoporosis patient information collection and monitoring databases, covering patients' basic information (such as height and weight), osteoporosis risk assessment questionnaires and bone density reports.

[0053] Through the hospital physical examination system and related databases, the age, menopause time, bone density test results and related risk factor data of older women can be obtained, providing a basis for early screening and intervention of osteoporosis.

[0054] Step S002: construct a scatter plot with each advanced female as a data point and menopausal time and age as the horizontal and vertical axes respectively; cluster all data points in the scatter plot to obtain several clusters; determine the volatility of the menopausal time changes of the advanced females in each cluster based on the number of advanced females contained in each cluster and the slope of the fitted straight line of all data points in the scatter plot and the slope of the fitted straight line of all data points in each cluster.

[0055] Because the timing of menopause and age changes in older women affect bone mass, and the number, physical condition, and timing of menopause vary among older women, overfitting is prone to occur when building large models. Therefore, data expansion is necessary based on the timing of menopause and age of different older women.

[0056] Since the osteoporosis condition of elderly women is related to their age and the time of their menopause, and due to the different physical conditions of different women, the menopause time of the same age will vary greatly. However, the menopause time of most elderly women is directly proportional to their age, so the current elderly women can be clustered.

[0057] A scatter plot is constructed with each elderly woman as a data point, and menopausal time and age as the horizontal and vertical axes respectively. Figure 2 Schematic diagram of a scatter plot of menopausal time and age for elderly women.

[0058] DBSCAN clustering is performed on all data points in the scatter plot to obtain several clusters, thereby dividing women in different age ranges into different clusters. Figure 3 Schematic diagram of the clustering results for elderly women.

[0059] It should be noted that DBSCAN (Density-Based Spatial Clustering of Applications with Noise) is a density-based clustering algorithm that defines a cluster as the largest set of densely connected points. It can cluster areas with sufficiently high density and can find clusters of arbitrary shapes in noisy spatial databases. As a state-of-the-art technique, we will not elaborate on it here.

[0060] For all data points in each cluster, a straight line was fitted using the least squares method to determine the changing relationship between menopausal time and age in older women in each cluster, as well as the slope of the straight line in each cluster. Figure 4 Schematic diagram of the changing relationship between menopausal time and age for each cluster of elderly women.

[0061] At the same time, for all data points in the scatter plot, a straight line was fitted using the least squares method to determine the changing relationship between menopausal time and age for all elderly women, as well as the slope of the overall straight line. Figure 5 Schematic diagram of the relationship between menopausal time and age for all elderly women.

[0062] It should be noted that the least squares formula is a mathematical formula, which is called straight line fitting in mathematics. It not only includes the linear regression equation, but also includes the least squares method of the matrix. As a well-known technology, it will not be described in detail here.

[0063] When the slope of the straight line in each cluster is closer to the overall slope of the straight line of all elderly women, and the distribution of elderly women in each cluster is denser, it means that the menopausal time of elderly women in this cluster is relatively concentrated, and it also conforms to the changing pattern of overall menopausal time and age. This can further reflect that the fluctuation of the change in the menopausal time of elderly women in this cluster is smaller.

[0064] According to the distribution of elderly women in each cluster, the minimum circumscribed rectangular area of ​​each cluster and the number of elderly women contained in it are obtained;

[0065] The first The minimum bounding rectangle area of ​​the cluster is the same as the The number of elderly women in each cluster is compared to determine the The distribution looseness of the elderly women in each cluster;

[0066] The first The slope of the straight line in each cluster is subtracted from the slope of the overall straight line corresponding to all elderly women and the absolute value is taken to determine the first The deviation between the menopausal timing and age regularity of the elderly women in each cluster;

[0067] The first The looseness of the distribution of elderly women in the first cluster is related to the The deviation between the menopausal time and the age regular relationship of the elderly women in the cluster is multiplied and normalized as the first The fluctuation of menopausal time in elderly women of clusters; norm is a linear normalization function.

[0068] Step S003: In each cluster, obtain several neighboring elderly women of each elderly woman; determine the proximity concentration of the elderly women's internal distribution in each cluster based on the distance between the elderly women in each cluster and their neighboring elderly women; determine the uniformity of the elderly women data in each cluster based on the number of times each elderly woman is a neighboring elderly woman of other elderly women.

[0069] When expanding the data for older women within each cluster, it's generally done to the same fixed value based on the female data in each set. However, when expanding the data for older women in different clusters, the menopausal timing and age proximity of each cluster may vary. Furthermore, some clusters may have a relatively large number of outliers in the data for older women. This indicates that a larger number of sample points within the menopausal timing and age range of these older women are needed within this cluster. Therefore, the required expanded sample size for each set can be determined by considering the closeness and uniformity of the data points for older women within the current cluster.

[0070] For any elderly women in each cluster, if the differences in age and menopausal duration among the women in the current cluster are smaller, it means that the distribution density of elderly women in the current cluster is higher. If the distribution density of elderly women in the current cluster is higher, the current expansion degree of the interval needs to be lower, which indicates that the distribution of elderly women in the current cluster is in a relatively stable state.

[0071] The default k value in the k-nearest neighbor algorithm is 5;

[0072] It should be noted that the k-nearest neighbor algorithm is a basic classification and regression method. Its core idea is "birds of a feather flock together", that is, the category of a sample is determined by the categories of its k nearest neighbors.

[0073] In each cluster, the The elderly women are centered and the k-nearest neighbor algorithm is used to obtain the Several elderly female neighbors of an elderly female.

[0074] In the In the clusters, get the An elderly woman and The first elderly women The Euclidean distance of the nearest elderly female An elderly woman and The sum of the Euclidean distances of all the nearest elderly women of an elderly woman is taken as the first The target distance of the elderly women is taken as the average of the target distances of all elderly women. The internal distribution of elderly women in each cluster is close to the concentration.

[0075] It should be noted that in this embodiment, each elderly woman is regarded as a data point, that is, the Euclidean distance between elderly women is the Euclidean distance between data points in the scatter plot.

[0076] In this way, this embodiment can obtain the proximity concentration of the internal distribution of elderly women in each cluster.

[0077] After determining the proximity concentration of the internal distribution of elderly women in each cluster, if most elderly women are close to each other, while some elderly women are far away from most elderly women, the female data in this cluster will be selected multiple times during the k-nearest neighbor selection. If there are some female data with relatively far distances, their probability of being selected will be relatively low, or they will not be selected at all. If this situation occurs, it means that the distribution of elderly women in this cluster is uneven.

[0078] Therefore, in this example, we can consider analyzing the number of times each elderly woman is selected by the k nearest neighbors in each cluster. If the overall number of selections is relatively balanced, it means that the distribution of elderly women in the current cluster is more uniform; otherwise, it means that the distribution of elderly women in the current cluster is uneven, and expansion is needed within the cluster.

[0079] Get the Within a cluster The number of times an elderly woman is the nearest elderly woman of other elderly women is recorded as The first cluster The number of times an elderly woman belongs to the adjacent range;

[0080] Get the The average number of times all elderly women in the cluster belong to the adjacent range is recorded as The average number of times the elderly women in each cluster belong to the adjacent range;

[0081] The first The first cluster The number of times an elderly woman belongs to the adjacent range is the same as the The average number of times the elderly women in each cluster belong to the adjacent range is subtracted and the absolute value is taken, which is recorded as The first cluster The deviation of the number of times of close range of elderly women;

[0082] The first The deviation of the number of adjacent ranges of all senior women in the cluster is accumulated and summed up to be the first The distribution deviation of elderly women in each cluster;

[0083] In the In the cluster, the number of elderly women with the number of times in the adjacent range of 0 is Compare the number of elderly women in each cluster to determine the The isolation degree of the distribution of elderly women in each cluster;

[0084] The first The distribution deviation of the elderly women in the first cluster is Multiply the isolation degree of the elderly female distribution of the clusters, perform norm normalization, and subtract the norm normalized value from 1 (this operation is an inverse proportional normalization operation), so as to obtain the first The uniformity of data of elderly women in each cluster;

[0085] Generally speaking, the number of women in each age group should be balanced. Therefore, the fewer the number of elderly women in a certain age group, the more necessary it is to expand the data.

[0086] Compare the number of elderly women in each cluster with the maximum number of elderly women in all clusters to determine the initial expansion demand of each cluster;

[0087] Because, considering that when each cluster is actually expanded, it is also necessary to fully consider the comprehensive impact of the fluctuation of the menopausal time of elderly women in each cluster, the proximity concentration of the internal distribution of elderly women, and the uniformity of the data of elderly women. That is, when the fluctuation of the menopausal time of elderly women in each cluster is greater, and the proximity concentration of the internal distribution of elderly women in each cluster and the uniformity of the data of elderly women are smaller, it means that the data of elderly women in the current cluster is more in line with the trend of needing data expansion;

[0088] Multiply the proximity concentration of the internal distribution of elderly women in each cluster and the uniformity of elderly women data, and record it as the expansion resistance of each cluster;

[0089] The volatility of menopausal timing for older women in each cluster was compared with the expansion resistance of each cluster and softmax normalized to determine the expansion adjustment coefficient for each cluster.

[0090] It should be noted that the Softmax function is a normalized exponential function, which is mainly used to "compress" a K-dimensional vector containing arbitrary real numbers into another K-dimensional real vector, so that each element ranges between (0,1) and the sum of all elements is 1.

[0091] Based on the initial expansion demand of each cluster, the expansion adjustment coefficient of each cluster is used to adjust it, so as to comprehensively determine the actual expansion demand of each cluster;

[0092] Multiplying the initial expansion demand of each cluster by the expansion adjustment coefficient of each cluster, and adding the result to the initial expansion demand of each cluster to determine the actual expansion demand of each cluster;

[0093] Step S004: Determine the actual expansion requirement of each cluster based on the number of elderly women in each cluster, the proximity concentration of the elderly women's internal distribution, the uniformity of the elderly women's data, and the volatility of the menopausal time of elderly women; expand the data of each cluster based on the actual expansion requirement of each cluster, and construct a big data bone detection model with the expanded data to monitor whether elderly women have osteoporosis.

[0094] Through the above process, this example can determine the actual expansion requirement of each cluster. The SMOTE algorithm is used to perform different degrees of data expansion on the bone data of elderly women in each cluster based on the actual expansion requirement of each cluster, thereby obtaining the expanded data of elderly women.

[0095] The expanded data on elderly women can be trained through deep neural networks to establish a big data bone detection model to obtain osteoporosis screening results for elderly women.

[0096] It should be noted that SMOTE (Synthetic Minority Over-sampling Technique) is an algorithm used to handle class-imbalanced data. Deep neural networks are a complex form of machine learning, belonging to the broader category of artificial neural networks. Designed to mimic the processing methods of the human brain, they process data through a multi-layer (or "deep") neural structure, thereby solving various complex data-driven problems.

[0097] So far, the present invention is completed.

[0098] To sum up, in an embodiment of the present invention, the present invention also provides an auxiliary screening system for osteoporosis in elderly women, including a memory, a processor, and a computer program stored in the memory and runnable on the processor. The processor executes the computer program stored in the memory to implement the steps of the aforementioned auxiliary screening method for osteoporosis in elderly women.

[0099] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. An auxiliary screening method for osteoporosis in elderly women, characterized by: The method comprises the following steps: Obtain the menopausal time and age of each elderly woman; A scatter plot was constructed with each elderly woman as a data point, and menopausal time and age as the horizontal and vertical axes, respectively. All data points in the scatter plot were clustered to obtain several clusters. The volatility of menopausal time changes in elderly women in each cluster was determined based on the number of elderly women in each cluster, the slope of the fitted line of all data points in the scatter plot, and the slope of the fitted line of all data points in each cluster. In each cluster, several elderly women with the closest neighbors of each elderly woman were obtained; based on the distance between elderly women in each cluster and their closest neighbors, the proximity concentration of elderly women within each cluster was determined; based on the number of times each elderly woman was a close neighbor of other elderly women, the uniformity of elderly women data in each cluster was determined; The actual expansion requirement of each cluster is determined based on the number of elderly women in each cluster, the proximity concentration of the elderly women's internal distribution, the uniformity of the elderly women's data, and the volatility of the menopausal time of elderly women. Data expansion is performed on each cluster based on the actual expansion requirement, and a big data bone detection model is constructed with the expanded data to monitor whether elderly women have osteoporosis.

2. The auxiliary screening method for osteoporosis in elderly women according to claim 1, characterized in that: The method of determining the volatility of menopausal timing of the elderly women in each cluster based on the number of elderly women included in each cluster and the slope of the fitted straight line of all data points in the scatter plot and the slope of the fitted straight line of all data points in each cluster comprises the following specific steps: For all data points in each cluster, a linear fit is performed using the least squares method to obtain the slope of the straight line in each cluster; For all data points in the scatter plot, a straight line is fitted using the least squares method to obtain the slope of the overall straight line; The first The minimum bounding rectangle area of ​​the cluster is the same as the The number of elderly women in each cluster is compared to determine the The distribution looseness of the elderly women in each cluster; The first The slope of the straight line in each cluster is subtracted from the slope of the overall straight line corresponding to all elderly women and the absolute value is taken to determine the first The deviation between the menopausal timing and age regularity of the elderly women in each cluster; According to The looseness of the distribution of elderly women in each cluster and the deviation of the relationship between the menopausal time and age of elderly women are used to determine the Fluctuations in menopausal timing among older women in three clusters.

3. The auxiliary screening method for osteoporosis in elderly women according to claim 2, characterized in that: According to the The looseness of the distribution of elderly women in each cluster and the deviation of the relationship between the menopausal time and age of elderly women are used to determine the The fluctuation of menopausal timing in each cluster of elderly women includes the following specific steps: The first The looseness of the distribution of elderly women in the first cluster is related to the The normalized value of the deviation between the menopausal time and the age regular relationship of the elderly women in the cluster is multiplied as the first Fluctuations in menopausal timing among older women in three clusters.

4. The auxiliary screening method for osteoporosis in elderly women according to claim 1, characterized in that: The method of determining the proximity concentration of the elderly women within each cluster based on the distance between the elderly women in each cluster and their adjacent elderly women includes the following specific steps: In the In the clusters, get the An elderly woman and The first elderly women The Euclidean distance of the nearest elderly female An elderly woman and The sum of the Euclidean distances of all the nearest elderly women of an elderly woman is taken as the first The target distance of the elderly women is taken as the average of the target distances of all elderly women. The internal distribution of elderly women in each cluster is close to the concentration.

5. The auxiliary screening method for osteoporosis in elderly women according to claim 1, characterized in that: The method of determining the uniformity of the elderly female data of each cluster based on the number of times each elderly female is a neighboring elderly female of other elderly females includes the following specific steps: Get the Within a cluster The number of times an elderly woman is the nearest elderly woman of other elderly women is recorded as The first cluster The number of times an elderly woman belongs to the adjacent range; Get the The average number of times all elderly women in the cluster belong to the adjacent range is recorded as The average number of times the elderly women in each cluster belong to the adjacent range; The first The first cluster The number of times an elderly woman belongs to the adjacent range is the same as the The average number of times the elderly women in each cluster belong to the adjacent range is subtracted and the absolute value is taken, which is recorded as The first cluster The deviation of the number of times of close range of elderly women; The first The deviation of the number of adjacent ranges of all senior women in the cluster is accumulated and summed up to be the first The distribution deviation of elderly women in each cluster; According to The distribution deviation of elderly women in the cluster is determined The isolation degree of the distribution of elderly women in each cluster; According to The distribution deviation and isolation of elderly women in each cluster are used to determine the The uniformity of data of elderly women in each cluster.

6. The auxiliary screening method for osteoporosis in elderly women according to claim 5, characterized in that: According to the The distribution deviation of elderly women in the cluster is determined The isolation degree of the elderly female distribution of each cluster is calculated by the following specific steps: In the In the cluster, the number of elderly women with the number of times in the adjacent range of 0 is Compare the number of elderly women in each cluster to determine the The isolation degree of the distribution of elderly women in each cluster.

7. The auxiliary screening method for osteoporosis in elderly women according to claim 5, characterized in that: According to the The distribution deviation and isolation of elderly women in each cluster are used to determine the The specific steps for measuring the uniformity of the data of elderly women in each cluster are as follows: The first The distribution deviation of the elderly women in the first cluster is The inverse proportional normalized value of the multiplication of the isolation degree of the elderly female distribution of the clusters is obtained The uniformity of data of elderly women in each cluster.

8. The auxiliary screening method for osteoporosis in elderly women according to claim 1, characterized in that: The actual expansion demand of each cluster is determined based on the number of elderly women in each cluster, the proximity concentration of elderly women's internal distribution, the uniformity of elderly women's data, and the volatility of menopausal time of elderly women. The specific steps include the following: Compare the number of elderly women in each cluster with the maximum number of elderly women in all clusters to determine the initial expansion demand of each cluster; Multiply the proximity concentration of the internal distribution of elderly women in each cluster and the uniformity of elderly women data, and record it as the expansion resistance of each cluster; The expansion adjustment coefficient of each cluster is determined by comparing the normalized value of the fluctuation of menopausal time of elderly women in each cluster with the expansion resistance of each cluster; The actual expansion requirement of each cluster is determined according to the expansion adjustment coefficient of each cluster.

9. The auxiliary screening method for osteoporosis in elderly women according to claim 8, characterized in that: The specific steps of determining the actual expansion requirement of each cluster class according to the expansion adjustment coefficient of each cluster class are as follows: The initial expansion requirement of each cluster is multiplied by the expansion adjustment coefficient of each cluster, and the result is added to the initial expansion requirement of each cluster to determine the actual expansion requirement of each cluster.

10. An auxiliary screening system for osteoporosis in elderly women, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the computer program is executed by a processor, the steps of the auxiliary screening method for osteoporosis in elderly women as described in any one of claims 1 to 9 are implemented.

Images