Bone mineral density prediction model construction method, prediction system, device and application

Abstract

The invention discloses a construction method of a bone mineral density prediction model, a prediction system, a device and application, and the construction method comprises the following steps: S1, obtaining age and bone mineral density data of all subjects, assuming that a set formed by bone mineral density of all subjects is X, a bone mineral density sample set of subjects at the age t is Xt = {xt1,... xti,... xtnt}, xti is the bone mineral density value of the ith subject at the age t, and if jgt, xt is the bone mineral density value of the ith subject at the age t; if i, xtj is greater than or equal to xti; for all ages t, any nt is larger than or equal to 3, and the minimum value of nt is recorded as min {nt}; s2, for each Xt, all elements of the Xt are divided into min {nt} groups from large to small, each group of elements form a set ZXt, p = {zxt, p, s}, and 1 < = p < = min {nt}; s3, all ZXt, p of different ages form a set CZXp = {czxp, d}, so that layering of the subject set X is realized, and P is the number of layers; and S4, fitting the elements of the pth layer to obtain a fitting equation, thereby obtaining P equations which are bone mineral density prediction models. The method can be used for bone mineral density prediction and bone mineral density metabolism research.

Description

Technical Field

[0001] The present invention relates to the field of bone density prediction, and particularly to a method for constructing a bone density prediction model, a prediction system, a device and an application thereof. Background Art

[0002] BMD (Bone Mineral Density) refers to the bone mineral content per unit area of bone. Bone density is an important indicator of human bone metabolism, which can reflect the change of bone mass in the human body, the degree of osteoporosis, and is also an important basis for predicting the risk of fractures. Clinically, two indicators, T value and Z value, are often used to evaluate bone density, and their clinical significance and applicable populations are essentially different. According to the different tested objects, usually one of the T value or Z value is used for bone density assessment. The bone density T value is one of the most critical indicators for evaluating the health status of adult bones. It measures the difference between the current bone density and the average bone density of healthy young adults (usually around 30 years old, reaching peak bone mass). The internationally recognized bone density T diagnostic criteria are: T value ≥ -1.0 indicates normal bone density; -2.5 < T value < -1.0 indicates osteopenia or hypo-osteogeny; T value ≤ -2.5 indicates osteoporosis, suggesting that the bone density is significantly lower than the average level of healthy young adults, and the risk of fracture will increase significantly. The bone density Z value is the core indicator for evaluating the health status of children's bones. It measures the difference between the current bone density and the average bone density value of healthy children of the same age, gender and race. When Z value ≥ -2.0, the bone density is within the normal range of the same age; when Z value < -2.0, the bone density is lower than the expected level of the same age.

[0003] Currently, the commonly used methods for measuring bone density clinically include dual-energy X-ray absorptiometry (DXA), quantitative CT (QCT), peripheral quantitative CT (pQCT) and ultrasonic bone densitometer (QUS). However, the above methods are highly dependent on detection equipment, and the detection results are only limited to the values of the current measurement, and it is impossible to predict bone density.

[0004] In recent years, with the rapid development of artificial intelligence technology, artificial intelligence models have gradually been applied to the evaluation of bone density health status. Based on medical images, using artificial intelligence methods to model image data can evaluate bone density values, but this method still depends on imaging instruments for the measurement of current bone density and cannot predict bone density.

[0005] Therefore, it is of great significance to construct an evaluation method that adopts a unified bone density evaluation standard and can predict and screen bone density risks to evaluate bone density metabolism and predict and screen bone density risks. Summary of the Invention

[0006] In view of the above background technology, the present invention provides a method, prediction system, device and application for constructing a bone mineral density prediction model. By utilizing existing clinical big data on age-bone mineral density T-scores, an age-bone mineral density mathematical model is constructed to predict bone mineral density at a target age and screen high-risk groups for bone mineral density, providing a brand-new method for clinical screening of bone mineral density and assessment of bone mineral density risk.

[0007] To achieve the above-mentioned objectives, the present invention provides the following technical solution:

[0008] This invention provides a method for constructing a bone mineral density prediction model, comprising the following steps:

[0009] S1. Obtain the age and bone mineral density data of all subjects. Assume that the set of bone mineral density data of all subjects is X, where the bone mineral density sample set of subjects aged t is X. t ={x t1 ,...x ti ,...x tn_t}, where x ti Let x be the bone mineral density value of the i-th subject at age t, and if j > i, then x tj ≥x ti For all ages t, for any n_t≥3, denote the minimum value of n_t as min{n_t}.

[0010] S2, for each X t Divide all its elements into min{n_t} groups in descending order, and each group forms a set ZX. t,p ={zx t,p,s}, where 1≤p≤min{n_t}.

[0011] S3, ZX of all ages t,p Constitute the set CZX p ={czx p,d This allows for the stratification of the subject set X, where P is the number of strata.

[0012] S4. Fit the elements of the p-th layer to obtain the fitting equation, thereby obtaining P equations, which are the bone density prediction model.

[0013] Existing research usually classifies bone mineral density according to the internationally accepted classification standards for research. For example, the bone mineral density T-value is classified according to the criteria of T-value ≥ -1.0, -2.5 < T-value < -1.0, and T-value ≤ -2.5 for research. When the inventor studied the relationship between bone mineral density and age, it was found that although the bone mineral density values of different ages were different, the changing trends of the high-value and low-value areas of bone mineral density were similar with the change of age. This suggests that the metabolic changes of bone mineral density are closely related to age. Therefore, the inventor pioneered the stratification of the bone mineral density T-values of the same age in the population according to high and low, and modeled the age-bone mineral density values of each layer to construct an age-bone mineral density model for different bone mineral density layers. This method no longer evaluates bone mineral density using T-values or Z-values for each age group, but uses bone mineral density T-values for all ages. In this way, an age-bone mineral density model from 0 years old to 100 years old can be constructed according to a unified standard, which can effectively observe the bone mineral density metabolism law of all ages and be more accurately used for predicting the trend of bone mineral density values based on age.

[0014] The above method for constructing a bone mineral density prediction model can stratify the existing large bone mineral density data to construct a model, making the sleeping clinical data become valuable scientific research data for studying the influence of different regions, living environments and other factors on bone mineral density metabolism.

[0015] Preferably, the method for constructing a bone mineral density prediction model further includes the following steps:

[0016] S5. For a subject M outside any sample set, assuming its age is t m , and the bone mineral density value is B m , where t m should be within the coverage of the aforementioned age t.

[0017] S6. For the sample set and the corresponding min{n_t} calculate the sum of the squares of the differences between B m and each element in it, and take the p with the smallest square as p m , then this subject belongs to the layer to which it belongs.

[0018] S7. Based on the fitting equation of the layer to which it belongs, predict the bone mineral density value of M's future age t' m (t' m > t m ).

[0019] The above-mentioned method for constructing a bone mineral density prediction model incorporates a step for predicting the bone mineral density value of subject M outside the sample set. This step can predict the bone mineral density value of a specific subject at a target age based on their previous bone mineral density test data through model calculation. For subjects who may have bone mineral density risks, intervention in bone mineral metabolism can be carried out as early as possible based on the results to reduce the occurrence of future bone mineral density risks.

[0020] Preferably, the method for constructing a bone mineral density prediction model further includes the following steps:

[0021] S8, when M grows to age t′ m The actual value of its bone mineral density was measured to be B′. m The same method as S6 is used to calculate its layer; if t′ m With t m If the bone mineral density (BMD) levels of the subjects are consistent, then the BMD metabolic level of the subject M is either unchanged or measurable based on age. If the BMD levels are inconsistent, then the BMD metabolic level of the subject M is not measurable based on age.

[0022] Analyzing changes in bone mineral density metabolism levels and assessing whether bone mineral density metabolism is normal can help detect diseases related to bone mineral density metabolism.

[0023] Preferably, each X in S2 t The method for dividing all elements into min{n_t} groups in descending order is as follows:

[0024] If n_t is an integer multiple of min{n_t}, then divide the elements into min{n_t} groups in descending order. If n_t is not an integer multiple of min{n_t}, then subtract the bone density values ​​of all elements pairwise, square the results, and combine the average of the two samples with the smallest squares into one sample, then update X. t And subtract 1 from n_t, in the updated X t Repeat this process multiple times until the new n_t becomes an integer multiple of min{n_t}, then divide the elements into min{n_t} groups in descending order.

[0025] By using the above method, adjacent elements are merged to make the grouping more scientific and reduce the errors introduced by the grouping.

[0026] Preferably, in S3, all ZX of different ages t,p Constitute the set CZX p ={czx p,d The method for dividing the subject set X into P layers is as follows:

[0027] Divide min{n_t} by P, i.e., min{n_t} / P = q... r, where r is the remainder and 0 ≤ r < P. Let each of the first r layers have (q + 1) numbers, and each of the subsequent (P - r) layers have q numbers.

[0028] Layer by using the above method to ensure that the number of groups divided in the same layer is the same, reducing the error introduced by uneven layering.

[0029] Preferably, the S4 fitting method uses the lsqcurvefit function of the nonlinear fitting method, and the fitting steps include:

[0030] Assume that the bone mineral density X of the p-th layer p varies with age t and satisfies a cubic polynomial:

[0031] X p (t) = a p1 t 3 + a p2 t 2 + a p3 t + a p4 ,

[0032] where a p1 , a p2 , a p3 and a p4 are the parameters to be fitted;

[0033] Input the modeling sample data into the polynomial, and fit to obtain a p1 , a p2 , a p3 and a p4 , and construct an age - bone mineral density prediction model.

[0034] Perform separate fitting for each bone mineral density layer at different ages, and independently construct the age - bone mineral density models for each layer, which can be used to explore the similarities and differences in metabolism between different bone mineral density layers.

[0035] To achieve the above invention purpose, the present invention also provides a bone mineral density prediction system, including an input module, an operation module, and an output module.

[0036] The input module is used to input the set X composed of all subjects and / or the bone mineral density sample data of subject M; the operation module extracts the data input by the input module and uses the construction method of any one of the above bone mineral density prediction models to construct an age - bone mineral density prediction model; the output module is used to output the constructed age - bone mineral density prediction model.

[0037] Bone mineral density (BMD) prediction systems can utilize existing large datasets of age-bone mineral density (AMD) samples to construct age-bone mineral density (BMD) models that fit the samples. This system can be used to simulate the age-related trends of BMD in a modeling region, constructing an AMD model that conforms to local conditions and predicting BMD for the local population. It can also be used to compare BMD levels between regions, identifying regional differences in bone mineral density metabolism. Furthermore, it can establish separate AMD models for populations with different living environments and lifestyles, comparing the impact of these factors on bone mineral density metabolism. The system can also use pre-constructed AMD models, inputting the subject's historically measured age and BMD values, to predict the subject's BMD at a target age, assessing whether further examination is needed. Alternatively, by comparing the predicted BMD value at the subject's target age with the actual value, it can determine the degree of deviation between the actual and predicted BMD, assisting in the examination of diseases related to bone mineral density metabolism.

[0038] To achieve the above-mentioned objectives, the present invention also provides a bone density prediction device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it executes any of the above-mentioned methods for constructing a bone density prediction model, constructs a bone density-age prediction model, and runs the above-mentioned bone density prediction system.

[0039] Bone density prediction devices can be used in hospitals, health check centers, and other institutions that require bone density screening; they can also be used in research institutions that conduct bone density studies.

[0040] To achieve the above-mentioned objectives, the present invention also provides a method for constructing a bone mineral density prediction model, a prediction system, and a prediction device, which are applied in the following ways: implementing the method for constructing a bone mineral density prediction model as described in any one of the claims, running the bone mineral density prediction system, and using the bone mineral density prediction device; and in the application of bone mineral density metabolism statistical tools, predicting the bone mineral density metabolism level of subject M, and assessing abnormal bone mineral density metabolism in subject M.

[0041] Compared with the prior art, the beneficial effects of the present invention include at least the following:

[0042] 1. The construction method, prediction system and device of the bone mineral density prediction model are used to construct and predict age-related bone mineral density models. No imaging equipment is required. The current and future bone mineral density of the subject can be estimated based on historical data, which reduces the use of medical care and equipment resources in bone mineral density screening and avoids unnecessary radiation to the subject. It can also be used to assess whether the subject's actual bone mineral density metabolism is abnormal by comparing the predicted bone mineral density with the actual bone mineral density.

[0043] 2. The construction method, prediction system and device of the bone mineral density prediction model are used to construct and predict age-related bone mineral density models. By inputting historical test data, the bone mineral density risk of the subjects can be screened, and the subjects can be given a bone mineral density risk warning. Intervention before the bone mineral density risk occurs can reduce the probability of osteoporosis or delay the onset of osteoporosis.

[0044] 3. Methods, prediction systems, and devices for constructing bone mineral density (BMD) prediction models: Utilizing clinical big data, age-bone mineral density (BMD) models can be established under the influence of factors such as different regions and populations. The constructed models, prediction systems, and devices can not only be used to study the influencing factors of BMD across regions and populations, but also to construct personalized BMD models based on local conditions, making the model construction and prediction results more closely reflect the actual situation of specific populations. Age-bone mineral density (BMD) metabolic models can also be constructed for continuous ages from young to old in a population to study the changes in BMD with age across all age groups. By constructing age-bone mineral density (BMD) models for the extremely high and extremely low BMD levels in a population and comparing them with age-bone mineral density (BMD) models for the intermediate levels, the reasons for the differences between individuals with very good and very poor BMD and the general population can be explored, and key points for intervening in BMD metabolism can be investigated. Attached Figure Description

[0045] The accompanying drawings, which are included to provide a further understanding of this application and form part of this application, do not constitute a limitation thereof. The components in the drawings are not drawn to scale but are merely for illustrating the principles of this application. For ease of illustration and description of certain parts of this application, corresponding portions in the drawings may be enlarged, i.e., may appear larger relative to other components in an exemplary device actually manufactured according to this application. In the drawings:

[0046] Figure 1 This is a flowchart illustrating a method for constructing a bone mineral density prediction model in one embodiment of this application;

[0047] Figure 2 This is a scatter plot of the average age-bone mineral density and a curve of the prediction model in one embodiment of this application;

[0048] Figure 3 This is a bone mineral density prediction system in one embodiment of this application. Detailed Implementation

[0049] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0050] The technical solution of the present invention will be described in detail below with reference to specific embodiments. These specific embodiments can be combined with each other, and the same or similar concepts or processes may not be described again in some embodiments.

[0051] The present invention will now be described in further detail with reference to the accompanying drawings:

[0052] Example 1

[0053] See Figure 1-2 This embodiment provides a method for constructing a bone mineral density prediction model, including:

[0054] S1, assume that the set of bone mineral density samples from all subjects is X, where the set of bone mineral density samples from subjects aged t is X. t ={x t1 ,...x t,i ,...x t,n_t}, where x ti Let x be the bone mineral density value of the i-th subject at age t, and satisfy the condition that if j>i, then x tj ≥x ti Furthermore, for all ages t, for any n_t≥3, the minimum value of n_t is denoted as min{n_t}.

[0055] Based on the age-related bone mineral density data in Table 1, select an age range t with at least 3 data points. The data in the table shows that min{n_t} is 3.

[0056] Table 1. Subject age and bone mineral density data

[0057]

[0058]

[0059]

[0060]

[0061]

[0062] S2, for each X tDivide all its elements into min{n_t} groups in descending order. If n_t is an integer multiple of min{n_t}, divide the elements into min{n_t} groups in descending order. If n_t is not an integer multiple of min{n_t}, subtract the bone density values ​​of all elements pairwise and square them. Take the mean of the bone density values ​​of the two samples with the smallest squares and merge them into one sample. Update Xt and subtract 1 from n_t. Repeat this process multiple times in the updated Xt until the new n_t becomes an integer multiple of min{n_t}. Then divide the elements into min{n_t} groups in descending order, where min{n_t} = 3.

[0063] Each group of elements constitutes a set ZX t,p ={zx t,p,s}, where p is 3.

[0064] S3, ZX of all ages t,p Constitute the set CZX p ={czx p,d The subject set X was divided into P layers, where P = min{n_t} = 3 layers, and the bone mineral density T-value data of each layer are shown in Table 2.

[0065] Table 2 Bone mineral density values ​​at different ages and bone layers

[0066]

[0067]

[0068] S4, using these data to fit and obtain the 3-layer bone mineral density X p The relationship between (p=1,2,3) and age t yields three equations, which constitute a bone mineral density prediction model. The model curves are shown below. Figure 2 As shown.

[0069] First layer:

[0070] X1 = 2.7073 × 10 -5 t 3 -0.0069t 2 +0.4324t-7.0480 (0.1)

[0071] Second layer:

[0072] X2 = 3.9394 × 10 -5 t 3 -0.0084t 2 +0.4800t-8.5074 (0.2)

[0073] Third layer:

[0074] X3 = 2.5455×10 -5 t 3 -0.0053t 2 +0.2773t - 5.7103 (0.3)

[0075] In some embodiments, if P≠min{n_t}, then when stratifying, min{n_t} should be divided by P, that is, min{n_t} / P = q...r, where r is the remainder and 0≤r<P, so that each of the first r layers has (q + 1) numbers and each of the subsequent (P - r) layers has q numbers.

[0076] In some other embodiments, the bone density value can also be modeled using the absolute value of the directly measured bone density.

[0077] Example 2

[0078] Based on the bone density prediction model constructed in Example 1, the following steps are implemented:

[0079] S5. The age of subject M is within the age range in Table 1 where the bone density sample size is more than 3.

[0080] S6. For the sample set and the corresponding min{n_t} Calculate the sum of the distances between B m and each element in

[0081]

[0082] Denote the p with the minimum sum of distances as p m , then this subject belongs to the <00002​​​​​​​​​​​​​​​​​​​​​​​​​​1,40 =2.9, S 2,40 =10.0 and S 3,40 =21.8, min{S 1,40 ,S 2,40 ,S 3,40}=2.9=S 1,40 At this point, the subject can be classified into category 1. Since their group is consistent at age 32 and age 40, the bone mineral density metabolic level of subject M1 is measurable based on age. Therefore, it can be inferred that M1's bone mineral density metabolism is normal, and no further examination is necessary.

[0086] See Figure 3 The present invention also discloses a bone density prediction system, comprising an input module, a calculation module and an output module.

[0087] The input module is used to input the subject's previous bone mineral density samples, including the sample age and the corresponding bone mineral density value;

[0088] The calculation module constructs an age-bone mineral density prediction model using the bone mineral density prediction model construction method described in the above embodiments, and calculates the predicted bone mineral density value for the subject's target age based on the data input by the input module.

[0089] The output module is used to output the predicted bone mineral density value for the subject's target age.

[0090] The present invention also discloses a bone mineral density prediction device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it executes the bone mineral density prediction model construction method in the above embodiments to construct a bone mineral density-age prediction model; and runs the above bone mineral density prediction system to predict the bone mineral density of the subject.

[0091] If the bone density prediction system is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium, namely the memory in the bone density prediction device.

[0092] The present invention also discloses a method for constructing a bone mineral density prediction model, a prediction system, and an application of a prediction device. The method for constructing the bone mineral density prediction model, the system for predicting bone mineral density, and the device for predicting bone mineral density are applied in bone mineral density metabolism statistical tools, predicting the bone mineral density metabolism level of subjects, and assessing abnormal bone mineral density metabolism in subjects.

[0093] The application of the above-mentioned bone mineral density prediction model construction method, prediction system and prediction device in bone mineral density research can effectively improve the efficiency of bone mineral density research, prediction and evaluation, and is conducive to exploring and discovering unrevealed patterns.

[0094] The above are merely preferred embodiments of the present invention and are not intended to limit the invention. Various modifications and variations can be made to the present invention by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A method for constructing a bone mineral density prediction model, characterized in that, The steps include the following: S1. Obtain the age and bone mineral density data of all subjects. Assume that the set of bone mineral density data of all subjects is X, where the bone mineral density sample set of subjects aged t is X. t ={x t1 ,...x ti ,...x tn_t }, where x ti Let x be the bone mineral density value of the i-th subject at age t, and if j > i, then x tj ≥x ti For all ages t, for any n_t≥3, denote the minimum value of n_t as min{n_t}; S2, for each X t Divide all its elements into min{n_t} groups in descending order, and each group forms a set ZX. t,p ={zx t,p,s }, where 1≤p≤min{n_t}; S3, ZX of all ages t,p Constitute the set CZX p ={czx p,d This allows for the stratification of the subject set X, where P is the number of strata. S4. Fit the elements of the p-th layer to obtain a fitting equation, and thus obtain P equations, which are the bone density prediction models.

2. The method for constructing the bone mineral density prediction model according to claim 1, characterized in that, The steps also include the following: S5, for a subject M outside of any sample set, assume their age is t. m Bone mineral density value B m , where t m It should be within the range covered by the aforementioned age t; S6, for the sample set and the corresponding min{n_t} Calculate B m With each The square of the difference between the elements is taken as p, and the smallest square of this p is denoted as p. m Then the subject belongs to To which layer; S7, based on The fitting equation for the layer to which it belongs, for the future age t′ of M m (t′ m >t m Bone mineral density values ​​are used to predict bone mineral density.

3. The method for constructing the bone mineral density prediction model according to claim 2, characterized in that, The steps also include the following: S8, when M grows to age t′ m The actual value of its bone mineral density was measured to be B′. m The same method as S6 is used to calculate its layer; if t′ m With t m If the bone mineral density (BMD) levels of the subjects are consistent, then the BMD metabolic level of the subject M is either unchanged or measurable based on age. If the BMD levels are inconsistent, then the BMD metabolic level of the subject M is not measurable based on age.

4. The method for constructing the bone mineral density prediction model according to claim 1, characterized in that, Each X in S2 t The method for dividing all elements into min{n_t} groups in descending order is as follows: If n_t is an integer multiple of min{n_t}, then divide the elements into min{n_t} groups in descending order. If n_t is not an integer multiple of min{n_t}, then subtract the bone density values ​​of all elements pairwise, square the results, and combine the average of the two samples with the smallest squares into one sample, then update X. t And subtract 1 from n_t, in the updated X t Repeat this process multiple times until the new n_t becomes an integer multiple of min{n_t}, then divide the elements into min{n_t} groups in descending order.

5. The method for constructing the bone mineral density prediction model according to claim 1, characterized in that, The S3 contains all ZX of different ages. t,p Constitute the set CZX p ={czx p,d The method for dividing the subject set X into P layers is as follows: Divide min{n_t} by P, that is, min{n_t} / P = q...r, where r is the remainder, 0 ≤ r < P, so that each of the first r layers has (q + 1) numbers, and each group of the subsequent (P - r) layers has q numbers.

6. The method for constructing the bone mineral density prediction model according to claim 1, characterized in that, The fitting method in the S4 adopts the lsqcurvefit function of the non-linear fitting method, and the fitting steps include: Assume the bone density X of layer p is... p The change with age t satisfies a cubic polynomial: X p (t)=a p1 t 3 +a p2 t 2 +a p3 t+a p4 , Where a p1 ,a p2 ,a p3 and a p4 These are the parameters to be fitted; Input the modeling sample data into the polynomial to obtain a. p1 ,a p2 ,a p3 and a p4 We constructed an age-bone mineral density prediction model.

7. A bone mineral density prediction system, characterized in that, It includes an input module, an operation module and an output module. The input module is used to input the set X composed of all subjects and / or the bone density sample data of the subject M. The operation module extracts the data input by the input module and constructs an age-bone density prediction model by using the construction method of the bone density prediction model described in any one of claims 1 to 6. The output module is used to output the constructed age-bone density prediction model.

8. A bone mineral density prediction device, characterized in that, It includes a memory, a processor and a computer program stored on the memory and executable on the processor. It is characterized in that when the processor executes the computer program, it executes the construction method of the bone density prediction model described in any one of claims 1 to 6 to construct a bone density-age prediction model and runs the bone density prediction system described in claim 7.

9. A method for constructing a bone mineral density prediction model, a prediction system, and an application of a prediction device, characterized in that, The application of executing the construction method of the bone density prediction model described in any one of claims 1 to 6, running the bone density prediction system described in claim 7, and adopting the bone density prediction device described in claim 8 in a bone density metabolism statistical tool, predicting the bone density metabolism level of the subject M, and evaluating the bone density metabolism abnormality of the subject M.

Images