Early screening method and system for osteoporosis based on multimodal data fusion

Abstract

The present invention relates to the field of medical informatics technology, and specifically to an early screening method and system for osteoporosis based on multimodal data fusion. The method comprises: obtaining the bone density, hormone content and calcium supplementation of sample personnel and test personnel; determining the potential impact of drugs on bone density based on the relative difference in bone density between sample personnel in the medication group and sample personnel in the non-medication group; if the test person is male, then determining the osteoporosis impact coefficient based on the changes in hormone content of the male sample personnel and the test person in two adjacent tests; if the test person is female, then performing differential analysis based on the stage of the test person and the calcium supplementation situation to determine the osteoporosis impact coefficient, and determining a multidimensional osteoporosis assessment coefficient based on the bone density of the test person. The present invention realizes a multidimensional and personalized assessment of bone condition, and improves the credibility of the assessment results.

Description

Technical Field

[0001] The present invention relates to the technical field of medical informatics, and in particular to an early screening method and system for osteoporosis based on multimodal data fusion. Background Art

[0002] Early screening for osteoporosis is of great clinical significance. Osteoporosis often has no obvious early symptoms, and many patients are unaware of the severity of the problem before a fracture occurs, significantly increasing the risk of osteoporotic fractures. Therefore, early screening is crucial to detect osteoporosis early and enable timely intervention to effectively reduce fracture risk and improve patients' quality of life.

[0003] Current diagnostic technology for osteoporosis primarily relies on dual-energy X-ray absorptiometry (DXA), which measures bone density to determine the severity of osteoporosis. However, DXA requires bone loss of at least 30% to detect an abnormality, resulting in insufficient sensitivity for diagnosing early osteoporosis. Consequently, this technology struggles to provide timely early warning of osteoporosis in its early stages, particularly in the elderly. Furthermore, osteoporosis screening in the elderly cannot be conducted using a single, standardized approach. Physiological differences between men and women, including the more pronounced effects of menopause in women, complicate matters, leading to further interference with early osteoporosis screening in both sexes. Furthermore, elderly individuals often take multiple medications for multiple chronic conditions, which can negatively impact bone metabolism and potentially interfere with early osteoporosis screening. Consequently, existing assessment methods offer limited reliability, impacting the accuracy of physicians' assessments of osteoporosis status. Summary of the Invention

[0004] In order to solve the problem of low reliability of the assessment results of existing methods when evaluating osteoporosis, the purpose of the present invention is to provide an early screening method and system for osteoporosis based on multimodal data fusion. The technical solutions adopted are as follows:

[0005] In a first aspect, the present invention provides an early screening method for osteoporosis based on multimodal data fusion, the method comprising the following steps:

[0006] Obtain bone density, hormone levels, and calcium supplementation status from multiple osteoporosis sample subjects and those to be tested;

[0007] Determine the potential effect of medication on bone density based on the relative difference in bone density between the sample of medication users and the sample of non-medication users;

[0008] If the subject is male, the osteoporosis influence coefficient of the subject's hormone factors is determined based on the changes in hormone levels of the male sample and the subject in two consecutive tests;

[0009] If the subject is female, when the subject is in the postmenopausal stage, the female sample personnel in the postmenopausal stage are clustered using the calcium supplementation status of the female sample personnel in the postmenopausal stage, the estrogen content of all tests, and the changes in hormone content between two adjacent tests; when the subject is in the menopausal transition stage, the female sample personnel in the menopausal transition stage are clustered using the fluctuation characteristics of the estradiol content of the female sample personnel in the menopausal transition stage; based on the overall distribution of bone density of the sample personnel in the cluster corresponding to the subject, the osteoporosis influence coefficient of the hormonal factor is determined; the cluster corresponding to the subject is determined by the cluster center method;

[0010] The multi-dimensional osteoporosis assessment coefficient of the person to be tested is determined by comprehensively considering the bone density of the person to be tested, the osteoporosis influence coefficient of hormonal factors and the potential influence of drugs on bone density.

[0011] Preferably, the determining of the potential impact of the drug on bone density based on the relative difference in bone density between the sample persons of the drug-using group and the sample persons of the non-drug-using group includes:

[0012] The average bone density of all sample persons in the medication group and all sample persons in the non-medication group is calculated, and recorded as the bone density characteristic value of the medication group and the bone density characteristic value of the non-medication group respectively;

[0013] The normalized result of the ratio between the bone density characteristic value of the drug-using group and the bone density characteristic value of the non-drug-using group is used as the potential impact of the drug on bone density.

[0014] Preferably, the method of determining the osteoporosis influence coefficient of the hormonal factors of the person to be tested based on the changes in hormone levels of the male sample person and the person to be tested in two consecutive tests includes:

[0015] For any male person: a first ratio of the androgen content of each test of the male person to that of the subsequent test is calculated, and a normalized result of the average of all first ratios is determined as the androgen decreasing trend degree of the male person; a second ratio of the estrogen content of each test of the male person to that of the previous test is calculated, and a normalized result of the average of all second ratios is determined as the estrogen increasing trend degree of the male person; the product of the androgen decreasing trend degree and the estrogen increasing trend degree is determined as the bone metabolism imbalance degree of the male person;

[0016] Obtain the osteoporosis influence coefficient of hormone factors for each male sample;

[0017] The least squares method was used to fit the bone metabolism imbalance of all male samples and the osteoporosis influence coefficient of hormonal factors to obtain the corresponding functional model relationship;

[0018] The bone metabolism imbalance degree of the person to be tested is input into the functional model relationship to obtain the osteoporosis influence coefficient of the hormone factor of the person to be tested.

[0019] Preferably, clustering the postmenopausal female sample persons using the calcium supplementation status of the postmenopausal female sample persons, all detected estrogen levels, and changes in hormone levels between two adjacent tests, includes:

[0020] For any female sample in the postmenopausal stage:

[0021] Calculating the total score of calcium-related supplementation of any female sample person, and determining the normalized result of the total score as the calcium supplement awareness of any female sample person; determining the average value of the estrone content of all the tests of any female sample person as the estrone basic performance value of any female sample person;

[0022] The mean of the inversely proportional normalized differences between the estrogen levels of all two consecutive tests of any female sample person is determined as the hormone stability coefficient of any female sample person;

[0023] Obtaining a comprehensive estrogen performance value of any female sample person according to the most recently detected estrogen content of any female sample person, the basic estrogen performance value, and the hormone stability coefficient;

[0024] Based on the similarities in calcium supplementation awareness and estrogen comprehensive performance values ​​of each pair of female sample personnel, all female sample personnel in the postmenopausal stage were clustered.

[0025] Preferably, obtaining the estrogen comprehensive performance value of any female sample person according to the most recently detected estrogen content of any female sample person, the estrogen basic performance value and the hormone stability coefficient includes:

[0026] Calculating a first product of the estrone content of the most recent test of any female sample person and the hormone stability coefficient;

[0027] Calculating the difference between the constant 1 and the hormone stability coefficient, and then multiplying the difference by the estrone comprehensive performance value of any female sample person to obtain a second product;

[0028] The sum of the first product and the second product is taken as the comprehensive estrone performance value of any female sample person.

[0029] Preferably, all female sample persons in the postmenopausal stage are clustered based on the similarities in calcium supplementation awareness and estrogen comprehensive performance values ​​of each pair of female sample persons, including:

[0030] The calcium supplement awareness and estrone comprehensive performance values ​​of each pair of female sample personnel were calculated by Euclidean distance, and the calculation results were used for clustering measurement. The DBSCAN clustering algorithm was used to cluster all female sample personnel in the postmenopausal stage.

[0031] Preferably, clustering female sample personnel in the menopausal transition stage by utilizing the fluctuation characteristics of estradiol levels in female sample personnel in the menopausal transition stage includes:

[0032] The normalized value of the standard deviation of the estradiol content of all the tests of the candidate sample person is used as the bone resorption-generation imbalance influence degree of the candidate sample person; the candidate sample person is any female sample person in the menopausal transition stage;

[0033] Based on the differences in the bone resorption-production imbalance influence among female samples at different stages of menopausal transition, the DBSCAN clustering algorithm was used to cluster all female samples at the menopausal transition stage.

[0034] Preferably, the step of determining the osteoporosis influence coefficient of hormonal factors based on the overall distribution of bone density of the sample persons in the cluster corresponding to the person to be tested includes:

[0035] The inverse proportional normalization result of the bone density of each sample person in the cluster corresponding to the tested person is determined as the osteoporosis factor of the corresponding sample person;

[0036] The average value of the osteoporosis factors of all sample persons in the cluster corresponding to the person to be tested is used as the osteoporosis influence coefficient of the hormone factor.

[0037] Preferably, the method comprehensively considers the bone density of the subject to be tested, the osteoporosis influence coefficient of hormone factors, and the potential influence of drugs on bone density to determine the multi-dimensional osteoporosis assessment coefficient of the subject to be tested, including:

[0038] Recording the difference between the constant 1 and the normalized value of the bone density of the person to be tested as a first difference, and calculating a third product of the first difference and the osteoporosis influence coefficient of the hormonal factor;

[0039] Calculating the fourth product of the normalized value of the bone density of the tested person and the potential effect of the drug on the bone density;

[0040] The sum of the third product and the fourth product is determined as the multi-dimensional osteoporosis assessment coefficient of the person to be tested.

[0041] In a second aspect, the present invention provides an early screening system for osteoporosis based on multimodal data fusion, the system being used to implement the above-described method, the system comprising:

[0042] The data collection module is used to obtain the bone density, hormone content and calcium supplementation status of multiple osteoporosis sample personnel and the persons to be tested;

[0043] A drug impact analysis module is used to determine the potential impact of drugs on bone density based on the relative difference in bone density between sample persons in the drug-using group and sample persons in the non-drug-using group;

[0044] The male assessment module is used to determine the osteoporosis influence coefficient of the hormonal factors of the tested person based on the changes in hormone levels of the male sample person and the tested person in two consecutive tests if the tested person is male;

[0045] The female assessment module is used to cluster female sample personnel in the postmenopausal stage using their calcium supplementation status, all estrogen levels tested, and changes in hormone levels between two consecutive tests if the subject is female; cluster female sample personnel in the postmenopausal stage using the fluctuation characteristics of estradiol levels in female sample personnel in the postmenopausal stage if the subject is female; determine the osteoporosis influence coefficient of hormonal factors based on the overall distribution of bone density of sample personnel in the cluster corresponding to the subject; the cluster corresponding to the subject is determined using the cluster center method;

[0046] The comprehensive assessment module is used to comprehensively consider the bone density of the person to be tested, the osteoporosis impact coefficient of hormonal factors, and the potential impact of drugs on bone density to determine the multi-dimensional osteoporosis assessment coefficient of the person to be tested.

[0047] The present invention has at least the following beneficial effects:

[0048] The present invention takes into account that the middle-aged and elderly population may be affected by long-term chronic diseases, and the drugs they take may affect osteoporosis. Therefore, the indirect factors of the drugs are quantified. In addition, since there are certain physiological differences between testers of different genders when testing for osteoporosis in the middle-aged and elderly, the gender of the testees is first distinguished, and different evaluation methods are used for people of different genders. If the testee is male, the hormone imbalance is evaluated based on the changes in hormone levels of male sample personnel and the testee in two adjacent tests, and then the osteoporosis influence coefficient of the testee's hormone factors is determined; if the testee is female, the hormone imbalance is evaluated based on the changes in hormone levels of male sample personnel and the testee in two adjacent tests, and then the osteoporosis influence coefficient of the testee's hormone factors is determined. Since women in the postmenopausal stage are mainly affected by the direct impact of decreased estrogen levels on osteoporosis, while women in the menopausal transition stage are mainly affected by the periodic fluctuations in estrogen levels, the imbalance between bone absorption and bone formation may still increase the risk of osteoporosis. Therefore, different assessment methods are adopted for the menopausal stage of the test subjects, and differentiated modeling analysis is carried out in combination with the test subjects' awareness of active calcium supplementation, which realizes a multi-dimensional and personalized assessment of the test subjects' bone condition, provides a highly reliable reference basis for doctors' subsequent diagnosis, and can achieve early warning and timely intervention for high-risk groups, which helps reduce the risk of fractures. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] In order to more clearly illustrate the technical solutions and advantages of the embodiments of the present invention or the prior art, the following briefly introduces the drawings required for use in the embodiments or the prior art descriptions. 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.

[0050] Figure 1 A flowchart of an early screening method for osteoporosis based on multimodal data fusion provided by an embodiment of the present invention;

[0051] Figure 2 This is a structural block diagram of the early osteoporosis screening system based on multimodal data fusion provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0052] In order to further illustrate the technical means and effects adopted by the present invention to achieve the predetermined purpose of the invention, the early screening method and system for osteoporosis based on multimodal data fusion proposed by the present invention are described in detail below in combination with the accompanying drawings and preferred embodiments.

[0053] 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.

[0054] The specific scheme of the early screening method and system for osteoporosis based on multimodal data fusion provided by the present invention is described in detail below with reference to the accompanying drawings.

[0055] Example of an early screening method for osteoporosis based on multimodal data fusion:

[0056] The specific scenario targeted by this embodiment is: when conducting a preliminary evaluation of osteoporosis in middle-aged and elderly people, due to certain physiological differences between men and women, such differences are not taken into account when using a single evaluation method. In order to provide more reliable reference information on the bone quality of the people to be tested and assist doctors in making subsequent judgments, this embodiment adopts different evaluation methods based on the different genders of the people to be tested and the characteristics of the stage they are in, thereby improving the accuracy and credibility of the evaluation results.

[0057] This embodiment proposes an early screening method for osteoporosis based on multimodal data fusion, such as Figure 1 As shown, the early screening method for osteoporosis based on multimodal data fusion of this embodiment includes the following steps:

[0058] Step S1, obtaining the bone density, hormone content and calcium supplementation status of multiple osteoporosis sample persons and persons to be tested.

[0059] As people age, they are often at risk of osteoporosis, especially those over 50. Bone density gradually decreases, and bone strength weakens, making them more susceptible to osteoporosis-related diseases and safety issues. This condition not only makes bones fragile and increases the risk of fractures, but can also lead to a series of serious consequences, such as long-term bed rest, decreased physical function, loss of self-care ability, and mental health issues. In short, appropriate early screening for osteoporosis in the middle-aged and elderly population not only helps to promptly identify potential safety hazards, but also allows for early warning and intervention, effectively reducing the risk of fractures and improving patients' overall health and quality of life.

[0060] This embodiment will provide a preliminary assessment method for osteoporosis in the middle-aged and elderly population to obtain comprehensive information for doctors' reference.

[0061] Specifically, the hospital database was used to obtain bone density, hormone levels, and calcium supplementation data from several individuals with a history of osteoporosis at each visit. Bone density, hormone levels, and calcium supplementation were also collected from the participants. Hormone levels included estrogen, testosterone, and estrone levels. Bone density was measured using dual-energy X-ray absorptiometry (DXA). Calcium supplementation was assessed using a questionnaire, which collected information on each participant's proactive calcium supplementation. The questionnaire included the following questions: "Do you regularly take calcium supplements?" (A. Yes, B. No); "Do you regularly consume foods high in calcium?" (A. Yes, B. No). In specific applications, implementers can customize the question types based on their specific circumstances. Answering "yes" to each question in the questionnaire awarded one point, while answering "no" awarded no point. The higher the participant's total score, the more effective their calcium supplementation.

[0062] It should be noted that the sample personnel include multiple male sample personnel and multiple female sample personnel. All data in this embodiment are obtained with full authorization. In this embodiment, the total number of sample personnel is 1000. In specific applications, the implementer can set it according to specific circumstances.

[0063] So far, this embodiment has obtained the bone density, hormone content and calcium supplementation status of each osteoporosis sample person.

[0064] Step S2: determining the potential impact of the drug on bone density based on the relative difference in bone density between the sample persons in the drug-using group and the sample persons in the non-drug-using group.

[0065] Existing dual-energy X-ray absorptiometry (DXA) methods for diagnosing osteoporosis through bone density testing are effective for people with significant bone density loss. However, they lack clear osteoporosis characteristics for early screening of osteoporosis in middle-aged and elderly individuals. This example considers extracting features associated with osteoporosis based on the hormone changes and lifestyle characteristics of middle-aged and elderly individuals.

[0066] Hormonal factors are the dominant factor in osteoporosis in middle-aged and elderly people during the aging process. However, these people may also be affected by long-term chronic diseases, and medications they take may have an indirect impact on osteoporosis. Therefore, this example uses relevant medical information to obtain a set of drugs that affect bone metabolism, such as glucocorticoids and anti-epileptic drugs (phenobarbital, carbamazepine). The group of sample participants who have taken any of these drugs within three consecutive months is considered the medication user group, and the group of all other sample participants is considered the non-medication user group.

[0067] Next, based on the medication group and the non-medication group, we will analyze whether the medication group will cause a more prominent osteoporosis-related impact.

[0068] Specifically, the mean bone density of all samples from the medication group, across all tests, is calculated and recorded as the bone density characteristic value for the medication group. The mean bone density of all samples from the non-medication group is calculated and recorded as the bone density characteristic value for the non-medication group. The normalized ratio of the bone density characteristic value of the medication group to the bone density characteristic value of the non-medication group is used as the potential impact of the drug on bone density. There are many data normalization methods, and implementers can select one based on their specific circumstances, so that the normalized result is in the range [0, 1].

[0069] Using the above method, the potential impact of drugs on bone density was obtained.

[0070] Step S3: If the person to be tested is male, the osteoporosis influence coefficient of the hormone factor of the person to be tested is determined based on the changes in hormone levels of the male sample person and the person to be tested in two consecutive tests.

[0071] Considering that middle-aged and elderly women experience endocrine disruptions and bone loss due to menopausal conditions, and that men experience a gradual decline in testosterone levels with aging, and due to differences in male and female physiology, female estrogen generally leads to a bidirectional regulation of bone resorption and formation, while male estrogen regulates bone resorption, while male bone formation is generally regulated by both androgens and estrogens.

[0072] Bone resorption and formation in men are driven by the combined effects of androgens and estrogens, with estrogen dominating bone resorption and testosterone and estrogen regulating bone formation. However, during male aging, androgen levels gradually decrease, while aromatase activity increases, which in turn drives more testosterone into estrogen. This causes estrogen levels to not decrease in tandem with androgen levels, disrupting the balance of bone metabolism. This imbalance leads to a continuous increase in osteoclast activity and a decrease in osteoblast repair capacity, ultimately causing damage to bone microstructure.

[0073] Therefore, this embodiment will analyze the person to be tested based on his or her gender. If the person to be tested is a male, this embodiment will conduct a relative balance analysis of male hormones and female hormones. If the two are in an unbalanced state, it indicates that the male hormone test results of multiple tests will show an overall downward trend and the female hormone will show a relatively upward trend.

[0074] Specifically, for any male person: calculate the ratio of the male hormone content in each test of the male person to the content in the next test, record the ratio as the first ratio, and determine the normalized result of the mean of all first ratios as the male person's male hormone decreasing trend degree; calculate the ratio of the female hormone content in each test of the male person to the content in the previous test, record the ratio as the second ratio, and determine the normalized result of the mean of all second ratios as the male person's female hormone increasing trend degree; and take the product of the male hormone decreasing trend degree and the female hormone increasing trend degree as the male person's bone metabolism imbalance degree.

[0075] Obtain the osteoporosis influence coefficient of hormonal factors for each male sample person; it should be noted that the osteoporosis influence coefficient of hormonal factors for male sample persons is obtained by doctors' evaluation based on their experience.

[0076] A least squares fit was performed on the bone metabolic imbalance scores of all male subjects and the osteoporosis influence coefficients of hormonal factors to obtain a corresponding functional model relationship. The bone metabolic imbalance scores of the subjects were then input into the functional model relationship to obtain the osteoporosis influence coefficients of the subjects' hormonal factors. The least squares method is a conventional technique and will not be further elaborated here.

[0077] Step S4, if the person to be tested is a female, then when the person to be tested is in the postmenopausal stage, the female sample persons in the postmenopausal stage are clustered using the calcium supplementation status of the female sample persons in the postmenopausal stage, all tested estrogen levels, and changes in hormone levels between two adjacent tests; when the person to be tested is in the menopausal transition stage, the female sample persons in the menopausal transition stage are clustered using the fluctuation characteristics of the estradiol levels of the female sample persons in the menopausal transition stage; based on the overall distribution of bone density of the sample persons in the cluster corresponding to the person to be tested, the osteoporosis influence coefficient of the hormonal factor is determined.

[0078] Considering that women are affected by menopausal hormone changes and have more complex hormonal manifestations, when the test subject is female, it is necessary to comprehensively evaluate the osteoporosis status of the test subject based on the situation of female sample persons.

[0079] First, according to the relevant menopausal standards for women, namely FSH>40U / L and E2<110pmol / L (≈30pg / mL), the hormone data of female personnel can be compared with the menopausal standards. Women who meet the menopausal standards can be classified as women in the postmenopausal stage, while women who do not meet the menopausal standards can be classified as women in the menopausal transition stage.

[0080] For women in the postmenopausal stage, although estrogen levels will decrease significantly, over time, the body's hormones will gradually adapt to the new levels and enter a relatively stable state. However, for women in the menopausal transition stage, ovarian hormone levels begin to fluctuate, and menstrual cycles may become irregular. At this time, fluctuating estrogen levels may lead to decreased bone density, and women may experience symptoms such as hot flashes, night sweats, and mood swings. This may indirectly affect bone health and lead to higher bone loss.

[0081] For middle-aged and elderly women, the impact on osteoporosis is divided according to the different stages of menopause. Women in the postmenopausal stage are mainly affected by the direct impact of decreased estrogen levels on osteoporosis; while women in the menopausal transition stage are mainly affected by the cyclical fluctuations in estrogen levels. The imbalance between bone absorption and bone formation may still indirectly increase the risk of osteoporosis.

[0082] Estrogen levels are generally relatively stable for women in the menopausal transition period. However, considering the single nature of hormonal changes during this period, which can be affected by conscious calcium supplementation, even if low hormone levels may lead to osteoporosis, conscious calcium supplementation can compensate for the hormone deficiency, resulting in different osteoporosis manifestations. Estrogen levels have different effects on bone density, and calcium supplementation can, to a certain extent, compensate for the impact of estrogen deficiency on bone density. Therefore, this example will determine the dynamic effects of estrogen and calcium supplementation on middle-aged and elderly postmenopausal women.

[0083] When the subjects were in the postmenopausal stage, the bone change characteristics of the females in the sample were analyzed.

[0084] Specifically, for any female sample person in the postmenopausal stage:

[0085] The total score of the female sample person's calcium-related supplementation is calculated, and the normalized result of the total score is determined as the female sample person's calcium supplement awareness; when normalizing the total score, an existing data normalization method is used, and the normalized result is a value of [0, 1]. The average value of the estrogen content of all the tests of the female sample person is determined as the female sample person's estrogen basic performance value.

[0086] For women in the postmenopausal stage of the middle-aged and elderly population, although the hormone level decreases, this state will gradually stabilize over time. Based on this feature, the absolute value of the difference between the estrogen content of each of the two adjacent tests of the female sample person is calculated as the difference between the estrogen content of the two tests. There is a difference between the estrogen content between each of the two adjacent tests, and each difference is inversely normalized; the mean of the inversely normalized results of the difference between the estrogen content of all two adjacent tests of the female sample person is determined as the hormone stability coefficient of the female sample person; in this embodiment, the method of inversely normalizing the difference is specifically: using the exponential function value with a natural constant as the base and the negative of the difference as the exponent as the inversely normalized result of the difference. As other implementation methods, other inversely normalized methods can also be used for processing, which will not be described in detail here.

[0087] The larger the hormone stability coefficient, the more stable the hormone status of the current women in the postmenopausal stage of the middle-aged and elderly group is, so the corresponding estrogen expression should be more inclined to the hormone value of the most recent test; if the hormone stability coefficient is smaller, it means that the current hormone is not in a relatively stable state, so it should be inclined to the basic estrogen expression value of women in the postmenopausal stage.

[0088] Therefore, this embodiment calculates the product of the most recently tested estrogen content of the female sample and the hormone stability coefficient, and records this product as the first product; calculates the difference between the constant 1 and the hormone stability coefficient, multiplies this difference by the female sample's estrogen comprehensive performance value, and records the resulting product as the second product; and the sum of the first product and the second product is used as the female sample's estrogen comprehensive performance value. The female sample's estrogen comprehensive performance value can be specifically expressed as:

[0089]

[0090] in, Indicates the comprehensive performance value of estrone of the female sample personnel, Indicates the hormone stability coefficient of the female sample, Indicates the estrone content of the female sample in the most recent test. It represents the comprehensive performance value of estrone of the female sample.

[0091] represents the first product, Represents the second product.

[0092] By using the above method, the comprehensive estrone performance value of each sample person can be obtained.

[0093] We calculated calcium supplement awareness and estrogen performance scores for each pair of female sample members using the Euclidean distance method. The results were then used to perform clustering. The DBSCAN clustering algorithm was then used to cluster all postmenopausal female sample members, resulting in multiple clusters. Each cluster reflects similar hormone profiles and proactive calcium supplementation awareness among women. The DBSCAN clustering algorithm is currently available and will not be further elaborated here.

[0094] For middle-aged and elderly women in the menopausal transition stage, the hormone fluctuations of women in this stage often correspond to multiple osteoporosis manifestations. Therefore, for their refined group division, it can be considered to deal with them from the perspective of hormone fluctuation manifestations.

[0095] Any female sample person in the menopausal transition stage is recorded as a candidate sample person, and the normalized value of the standard deviation of the estradiol content of all tests of the candidate sample person is used as the bone resorption-production imbalance influence degree of the candidate sample person, which is used to reflect that women in the menopausal transition stage are mainly affected by the cyclical fluctuation of estrogen levels.

[0096] Based on the differences in bone resorption-generation imbalance influence degrees among female sample personnel at different stages of menopausal transition, the DBSCAN clustering algorithm was used to cluster all female sample personnel at the menopausal transition stage. When clustering using the DBSCAN clustering algorithm, the absolute value of the difference between the bone resorption-generation imbalance influence degrees of female sample personnel at different stages of menopausal transition was used as the distance measurement method between the corresponding two female sample personnel. All female sample personnel at the menopausal transition stage were clustered using the DBSCAN clustering algorithm to obtain multiple clusters.

[0097] Then, the cluster type corresponding to the person to be tested is determined by the cluster center method. Specifically, the center of each cluster is first obtained. Then, for the person to be tested, the distance from the person to the center of all clusters is calculated, and the cluster type with the shortest distance is taken as the cluster type corresponding to the person to be tested.

[0098] The inverse proportional normalization result of the bone density of each sample person in the cluster corresponding to the person to be tested is determined as the osteoporosis factor of the corresponding sample person; the average value of the osteoporosis factor of all sample persons in the cluster corresponding to the person to be tested is used as the osteoporosis influence coefficient of the hormonal factor. It should be noted that: in this embodiment, the inverse proportional normalization method of bone density is: an exponential function with a natural constant as the base and a negative bone density as the exponent is used as the inverse proportional normalization result of bone density. As other implementation methods, other inverse proportional normalization methods can also be used for processing, which will not be described in detail here.

[0099] Step S5, comprehensively considering the bone density of the person to be tested, the osteoporosis influence coefficient of hormone factors, and the potential influence of drugs on bone density, to determine a multi-dimensional osteoporosis assessment coefficient of the person to be tested.

[0100] Through the above process, the corresponding osteoporosis impact coefficient is determined in different ways based on the individual situation of the person to be tested, and the synergistic effects of medical history, medication and lifestyle health are taken into consideration to determine the potential impact of the drug on bone density. Next, the osteoporosis impact coefficient and the potential impact of the drug on bone density will be comprehensively evaluated on the person to be tested.

[0101] When bone density is low, it means that bone health has been affected to a greater extent. That is, the elderly population often has a decrease in bone density due to changes in hormone levels in the body. At this time, more attention should be paid to the impact of hormones. On the contrary, when bone density is high, it means that the decrease in bone density caused by hormones is not significant. Therefore, we can focus on the potential impact of drugs on bone density.

[0102] Specifically, the difference between the constant 1 and the normalized value of the bone density of the person to be tested is recorded as the first difference, the product of the first difference and the osteoporosis influence coefficient of the hormonal factor is calculated, and the product is recorded as the third product; the product of the normalized value of the bone density of the person to be tested and the potential influence of the drug on bone density is calculated, and the product is recorded as the fourth product; the sum of the third product and the fourth product is determined as the multidimensional osteoporosis assessment coefficient of the person to be tested.

[0103] In this embodiment, a specific calculation formula for the multi-dimensional osteoporosis assessment coefficient of the person to be tested is given. The multi-dimensional osteoporosis assessment coefficient of the person to be tested can be expressed as:

[0104]

[0105] Among them, Q represents the multi-dimensional osteoporosis assessment coefficient of the tested person, represents the normalized value of the bone density of the person to be tested, Indicates the osteoporosis influence coefficient of hormone factors, Indicates the potential effect of a drug on bone density.

[0106] represents the first difference, represents the third product, When the bone density of the tested individual is low, it indicates that bone health is significantly affected, and in this case, greater attention should be paid to the influence of hormones.

[0107] It should be noted that the normalization method of bone density adopts the existing data normalization method to process, so that the value of the normalized result is [0, 1].

[0108] Through the above process, this embodiment obtains the multi-dimensional osteoporosis assessment coefficient of the person to be tested. Subsequently, the doctor can evaluate the osteoporosis condition of the person to be tested based on the multi-dimensional osteoporosis assessment coefficient of the person to be tested. When the risk of osteoporosis occurs, it should be monitored and managed more strictly, and appropriate drug intervention and treatment plans can be adopted if necessary.

[0109] This embodiment takes into account that the middle-aged and elderly population may be affected by long-term chronic diseases, and the drugs they take may affect osteoporosis. Therefore, the indirect factors of the drugs are quantified. In addition, because there are certain physiological differences between testers of different genders when testing for osteoporosis in the middle-aged and elderly, the gender of the testees is first distinguished, and different evaluation methods are used for people of different genders. If the testee is male, the hormone imbalance is evaluated based on the changes in hormone levels of male sample personnel and the testee in two adjacent tests, and then the osteoporosis influence coefficient of the testee's hormone factors is determined; if the testee is female, Since women in the postmenopausal stage are mainly affected by the direct impact of decreased estrogen levels on osteoporosis, while women in the menopausal transition stage are mainly affected by the periodic fluctuations in estrogen levels, the imbalance between bone absorption and bone formation may still increase the risk of osteoporosis. Therefore, different assessment methods are adopted for the menopausal stage of the test subjects, and differentiated modeling analysis is carried out in combination with the test subjects' awareness of active calcium supplementation, which realizes a multi-dimensional and personalized assessment of the test subjects' bone condition, provides a highly reliable reference basis for doctors' subsequent diagnosis, and can achieve early warning and timely intervention for high-risk groups, which helps reduce the risk of fractures.

[0110] Example of an early screening system for osteoporosis based on multimodal data fusion:

[0111] See Figure 2 , which shows a structural block diagram of an early screening system for osteoporosis based on multimodal data fusion provided by an embodiment of the present invention. The system may include a data acquisition module, a drug impact analysis module, a male evaluation module, a female evaluation module and a comprehensive evaluation module.

[0112] The data collection module is used to obtain the bone density, hormone content and calcium supplementation status of multiple osteoporosis sample personnel and the personnel to be tested;

[0113] A drug impact analysis module is used to determine the potential impact of drugs on bone density based on the relative difference in bone density between sample persons in the drug-using group and sample persons in the non-drug-using group;

[0114] The male assessment module is used to determine the osteoporosis influence coefficient of the hormonal factors of the tested person based on the changes in hormone levels of the male sample person and the tested person in two consecutive tests if the tested person is male;

[0115] The female assessment module is used to cluster female sample personnel in the postmenopausal stage using their calcium supplementation status, all estrogen levels tested, and changes in hormone levels between two consecutive tests if the subject is female; cluster female sample personnel in the postmenopausal stage using the fluctuation characteristics of estradiol levels in female sample personnel in the postmenopausal stage if the subject is female; determine the osteoporosis influence coefficient of hormonal factors based on the overall distribution of bone density of sample personnel in the cluster corresponding to the subject; the cluster corresponding to the subject is determined using the cluster center method;

[0116] The comprehensive assessment module is used to comprehensively consider the bone density of the person to be tested, the osteoporosis impact coefficient of hormonal factors, and the potential impact of drugs on bone density to determine the multi-dimensional osteoporosis assessment coefficient of the person to be tested.

[0117] It should be understood that Figure 2 The structural block diagram of the multimodal data fusion-based early osteoporosis screening system and its modules can be implemented in various ways. For example, in some embodiments, the system and its modules can be implemented using hardware, software, or a combination of software and hardware. The hardware portion can be implemented using dedicated logic, while the software portion can be stored in memory and executed by an appropriate instruction execution system, such as a microprocessor or dedicated hardware. Those skilled in the art will appreciate that the above-described methods and systems can be implemented using computer-executable instructions and / or contained in processor control code, for example, provided on a carrier medium such as a disk, CD, or DVD-ROM, a programmable memory such as read-only memory (firmware), or a data carrier such as an optical or electronic signal carrier. The systems and their modules described herein can be implemented not only using hardware circuits such as very large-scale integrated circuits or gate arrays, semiconductors such as logic chips or transistors, or programmable hardware devices such as field-programmable gate arrays or programmable logic devices, but can also be implemented using software executed by various types of processors, or a combination of the above hardware circuits and software (e.g., firmware).

[0118] For more details about the above modules, please refer to other places in this manual and will not be repeated here.

[0119] In other embodiments, a device for early screening of osteoporosis based on multimodal data fusion is provided, comprising a memory and a processor. The memory is used to store executable program code, and the processor is used to call and execute the executable program code from the memory, causing the device to execute the above-mentioned method for early screening of osteoporosis based on multimodal data fusion. The device can specifically be a chip, component, or module, and the chip can include a connected processor and memory; the memory is used to store instructions, and when the processor calls and executes the instructions, the chip can execute the method for early screening of osteoporosis based on multimodal data fusion provided in the above-mentioned embodiments.

[0120] In other embodiments, a computer program product is also provided. When the computer program product is run on a computer, the computer is caused to execute the above-mentioned related steps to implement the early screening method for osteoporosis based on multimodal data fusion provided in the above embodiment.

[0121] In other embodiments, a computer-readable storage medium is also provided, in which a computer program code is stored. When the computer program code is run on a computer, the computer executes the above-mentioned related method steps to implement the smart charging pile metering method based on Internet of Things technology provided in the above embodiment.

[0122] Among them, the provided systems, electronic devices, computer program products, and computer-readable storage media are all used to execute the corresponding methods provided above. Therefore, the beneficial effects that can be achieved can refer to the beneficial effects in the corresponding methods provided above, and will not be repeated here.

[0123] It should be noted that the above is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent replacements, 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 early screening method for osteoporosis based on multimodal data fusion, characterized in that: The method comprises the following steps: Obtain bone density, hormone levels, and calcium supplementation status from multiple osteoporosis sample subjects and those to be tested; The potential impact of drugs on bone density was determined based on the relative difference in bone density between the sample population of drug users and the sample population of non-drug users. The drug user group was composed of all sample populations whose medications had affected bone metabolism within three consecutive months, and the non-drug user group was composed of all sample populations except the drug user group. If the subject is male, the osteoporosis influence coefficient of the subject's hormone factors is determined based on the changes in hormone levels of the male sample and the subject in two consecutive tests; If the subject is female, when the subject is in the postmenopausal stage, the female sample personnel in the postmenopausal stage are clustered using the calcium supplementation status of the female sample personnel in the postmenopausal stage, the estrogen content of all tests, and the changes in hormone content between two adjacent tests; when the subject is in the menopausal transition stage, the female sample personnel in the menopausal transition stage are clustered using the fluctuation characteristics of the estradiol content of the female sample personnel in the menopausal transition stage; based on the overall distribution of bone density of the sample personnel in the cluster corresponding to the subject, the osteoporosis influence coefficient of the hormonal factor is determined; the cluster corresponding to the subject is determined by the cluster center method; The multi-dimensional osteoporosis assessment coefficient of the person to be tested is determined by comprehensively considering the bone density of the person to be tested, the osteoporosis influence coefficient of hormonal factors and the potential influence of drugs on bone density.

2. The method for early screening of osteoporosis based on multimodal data fusion according to claim 1, characterized in that: The method of determining the potential impact of the drug on bone density based on the relative difference in bone density between the sample persons in the drug-using group and the sample persons in the non-drug-using group includes: The average bone density of all sample persons in the medication group and all sample persons in the non-medication group is calculated, and recorded as the bone density characteristic value of the medication group and the bone density characteristic value of the non-medication group respectively; The normalized result of the ratio between the bone density characteristic value of the drug-using group and the bone density characteristic value of the non-drug-using group is used as the potential impact of the drug on bone density.

3. The method for early screening of osteoporosis based on multimodal data fusion according to claim 1, characterized in that: The method of determining the osteoporosis influence coefficient of the hormone factor of the person to be tested based on the changes in hormone levels of the male sample person and the person to be tested in two consecutive tests includes: For any male person: a first ratio of the androgen content of each test of the male person to that of the subsequent test is calculated, and a normalized result of the average of all first ratios is determined as the androgen decreasing trend degree of the male person; a second ratio of the estrogen content of each test of the male person to that of the previous test is calculated, and a normalized result of the average of all second ratios is determined as the estrogen increasing trend degree of the male person; the product of the androgen decreasing trend degree and the estrogen increasing trend degree is determined as the bone metabolism imbalance degree of the male person; Obtain the osteoporosis influence coefficient of hormone factors for each male sample; The least squares method was used to fit the bone metabolism imbalance of all male samples and the osteoporosis influence coefficient of hormonal factors to obtain the corresponding functional model relationship; The bone metabolism imbalance degree of the person to be tested is input into the functional model relationship to obtain the osteoporosis influence coefficient of the hormone factor of the person to be tested.

4. The method for early screening of osteoporosis based on multimodal data fusion according to claim 1, characterized in that: The method of clustering postmenopausal female sample personnel by using the calcium supplementation status of the postmenopausal female sample personnel, all estrogen levels tested, and changes in hormone levels between two adjacent tests, includes: For any female sample in the postmenopausal stage: Calculating the total score of calcium-related supplementation of any female sample person, and determining the normalized result of the total score as the calcium supplement awareness of any female sample person; determining the average value of the estrone content of all the tests of any female sample person as the estrone basic performance value of any female sample person; The mean of the inversely proportional normalized differences between the estrogen levels of all two consecutive tests of any female sample person is determined as the hormone stability coefficient of any female sample person; Obtaining a comprehensive estrogen performance value of any female sample person according to the most recently detected estrogen content of any female sample person, the basic estrogen performance value, and the hormone stability coefficient; Based on the similarities in calcium supplementation awareness and estrogen comprehensive performance values ​​of each pair of female sample personnel, all female sample personnel in the postmenopausal stage were clustered.

5. The method for early screening of osteoporosis based on multimodal data fusion according to claim 4, characterized in that: The method of obtaining the comprehensive estrogen performance value of any female sample person according to the most recently detected estrogen content of any female sample person, the basic estrogen performance value, and the hormone stability coefficient includes: Calculating a first product of the estrone content of the most recent test of any female sample person and the hormone stability coefficient; Calculating the difference between the constant 1 and the hormone stability coefficient, and then multiplying the difference by the estrone comprehensive performance value of any female sample person to obtain a second product; The sum of the first product and the second product is taken as the comprehensive estrone performance value of any female sample person.

6. The method for early screening of osteoporosis based on multimodal data fusion according to claim 4, characterized in that: Based on the similarities in calcium supplementation awareness and estrogen comprehensive performance values ​​of the two female sample members, all female sample members in the postmenopausal stage were clustered, including: The calcium supplement awareness and estrone comprehensive performance values ​​of each pair of female sample personnel were calculated by Euclidean distance, and the calculation results were used for clustering measurement. The DBSCAN clustering algorithm was used to cluster all female sample personnel in the postmenopausal stage.

7. The method for early screening of osteoporosis based on multimodal data fusion according to claim 1, characterized in that: The method of clustering female sample personnel in the menopausal transition stage by utilizing the fluctuation characteristics of estradiol levels in female sample personnel in the menopausal transition stage includes: The normalized value of the standard deviation of the estradiol content of all the tests of the candidate sample person is used as the bone resorption-generation imbalance influence degree of the candidate sample person; the candidate sample person is any female sample person in the menopausal transition stage; Based on the differences in the bone resorption-production imbalance influence among female samples at different stages of menopausal transition, the DBSCAN clustering algorithm was used to cluster all female samples at the menopausal transition stage.

8. The method for early screening of osteoporosis based on multimodal data fusion according to claim 1, characterized in that: The step of determining the osteoporosis influence coefficient of hormone factors based on the overall distribution of bone density of the sample persons in the cluster corresponding to the person to be tested includes: The inverse proportional normalization result of the bone density of each sample person in the cluster corresponding to the tested person is determined as the osteoporosis factor of the corresponding sample person; The average value of the osteoporosis factors of all sample persons in the cluster corresponding to the person to be tested is used as the osteoporosis influence coefficient of the hormone factor.

9. The method for early screening of osteoporosis based on multimodal data fusion according to claim 1, characterized in that: The multi-dimensional osteoporosis assessment coefficient of the subject to be tested is determined by comprehensively considering the bone density of the subject to be tested, the osteoporosis influence coefficient of hormone factors, and the potential influence of drugs on bone density, including: Recording the difference between the constant 1 and the normalized value of the bone density of the person to be tested as a first difference, and calculating a third product of the first difference and the osteoporosis influence coefficient of the hormonal factor; Calculating the fourth product of the normalized value of the bone density of the tested person and the potential effect of the drug on the bone density; The sum of the third product and the fourth product is determined as the multi-dimensional osteoporosis assessment coefficient of the person to be tested.

10. An early screening system for osteoporosis based on multimodal data fusion, the system being used to implement the method of claim 1, characterized in that: The system includes: The data collection module is used to obtain the bone density, hormone content and calcium supplementation status of multiple osteoporosis sample personnel and the persons to be tested; A drug impact analysis module is used to determine the potential impact of drugs on bone density based on the relative difference in bone density between sample persons in the drug-using group and sample persons in the non-drug-using group; The male assessment module is used to determine the osteoporosis influence coefficient of the hormonal factors of the tested person based on the changes in hormone levels of the male sample person and the tested person in two consecutive tests if the tested person is male; The female assessment module is used to cluster female sample personnel in the postmenopausal stage using their calcium supplementation status, all estrogen levels tested, and changes in hormone levels between two consecutive tests if the subject is female; cluster female sample personnel in the postmenopausal stage using the fluctuation characteristics of estradiol levels in female sample personnel in the postmenopausal stage if the subject is female; determine the osteoporosis influence coefficient of hormonal factors based on the overall distribution of bone density of sample personnel in the cluster corresponding to the subject; the cluster corresponding to the subject is determined using the cluster center method; The comprehensive assessment module is used to comprehensively consider the bone density of the person to be tested, the osteoporosis impact coefficient of hormonal factors, and the potential impact of drugs on bone density to determine the multi-dimensional osteoporosis assessment coefficient of the person to be tested.

Images