A fracture risk prediction device, method, apparatus, and storage medium

Abstract

The application discloses a kind of bone fracture risk prediction equipment, method, device and storage medium.The equipment includes impedance measuring unit, impedance conversion unit and impedance analysis unit;The electrode end of impedance measuring unit is in contact with the skin at the bone of the area to be detected, the output end of impedance measuring unit is connected with the input end of impedance conversion unit, the output end of impedance conversion unit is connected with the input end of impedance analysis unit.The application passes weak current to the skin at the bone of the area to be detected, and the signal generated by the bone based on weak current is converted by impedance conversion unit to obtain impedance parameter;Finally, analysis and prediction are carried out to obtain the proportion of bone composition, and the bone fracture risk prediction result is obtained.The existing bone density evaluation prediction can only reflect 70% of bone strength.The application can be used as a supplement to traditional bone density determination evaluation, and can also be used to reflect osteoporosis according to the bone tissue composition of the measured part, and the bone fracture risk prediction accuracy is improved.

Description

Technical Field

[0001] This invention relates to the field of fracture risk prediction technology, and in particular to a fracture risk prediction device, method, apparatus and storage medium. Background Technology

[0002] With the increasing aging population, the burden on healthcare systems caused by fractures is becoming increasingly severe. Fragility fractures are the most serious complication of osteoporosis, with a very high incidence rate among elderly osteoporosis patients. Surveys show that the number of people suffering fractures due to osteoporosis is enormous. Osteoporosis (OP) is a systemic bone disease characterized by low bone mass, damage to bone microstructure, and increased bone fragility, making fractures more likely. Therefore, predicting bone risk is an important means of fracture prevention.

[0003] Currently, fracture risk is primarily predicted through assessment of the degree of osteoporosis. Bone mineral density (BMD) is the best quantitative indicator for diagnosing osteoporosis and predicting the risk of osteoporotic fractures. Dual-energy X-ray absorptiometry (DXA) is the gold standard for diagnosing osteoporosis, with a reduction ≥2.5 standard deviations considered osteoporosis. Quantitative ultrasound (QUS) and radiographic imaging can also be used for osteoporosis assessment. However, there is no unified diagnostic standard for QUS, and radiographic imaging has low sensitivity and accuracy in diagnosing osteoporosis; only a 30% reduction in bone mass will be visible on radiographs. All of the diagnostic methods for osteoporosis mentioned above have certain limitations. Even the gold standard, DXA, only reflects approximately 70% of bone strength, resulting in low sensitivity and accuracy in risk prediction.

[0004] The above content is only used to help understand the technical solution of the present invention and does not represent an admission that the above content is prior art. Summary of the Invention

[0005] The main objective of this invention is to provide a fracture risk prediction device, method, apparatus, and storage medium, aiming to solve the technical problem of low sensitivity and accuracy of traditional fracture risk prediction methods.

[0006] To achieve the above objectives, the present invention provides a fracture risk prediction device, which includes an impedance measurement unit, an impedance conversion unit, and an impedance analysis unit.

[0007] The impedance measurement unit has its electrode contacting the skin at the bone in the area to be tested. The output of the impedance measurement unit is connected to the input of the impedance conversion unit, and the output of the impedance conversion unit is connected to the input of the impedance analysis unit.

[0008] The impedance measurement unit is used to receive the weak current output by the impedance analysis unit, transmit the weak current to the skin of the bone in the area to be detected, and output the impedance signal generated by the bone based on the weak current to the impedance conversion unit.

[0009] The impedance conversion unit is used to convert the impedance signal to obtain impedance parameters and output the impedance parameters to the impedance analysis unit.

[0010] The impedance analysis unit is used to perform analysis and prediction based on the impedance parameters to obtain the fracture risk prediction result of the target to be detected.

[0011] Optionally, the impedance measurement unit includes a shielded chamber, a transmitting electrode, and a receiving electrode;

[0012] The transmitting electrode and the receiving electrode are disposed inside the shielding chamber; the input terminal of the transmitting electrode is connected to the output terminal of the impedance conversion unit, and the output terminal of the transmitting electrode is in contact with the skin at the bone of the area to be detected; the input terminal of the receiving electrode is in contact with the skin at the other end of the bone, and the output terminal of the receiving electrode is connected to the input terminal of the impedance conversion unit.

[0013] The shielding room is used to shield interference signals from the external environment;

[0014] The transmitting electrode is used to transmit the weak current transmitted by the impedance conversion unit to the bone.

[0015] The receiving electrode is used to receive the impedance signal generated by the bone and output it to the impedance conversion unit.

[0016] Optionally, the impedance conversion unit includes: a signal generator, a digital-to-analog converter, and an analog-to-digital converter;

[0017] The input terminal of the signal generator is connected to the output terminal of the impedance analysis unit, the output terminal of the signal generator is connected to the input terminal of the digital-to-analog converter, the output terminal of the digital-to-analog converter is connected to the input terminal of the transmitting electrode, the output terminal of the receiving electrode is connected to the input terminal of the analog-to-digital converter, and the output terminal of the analog-to-digital converter is connected to the input terminal of the impedance analysis unit.

[0018] The signal generator is used to receive the voltage signal from the impedance analysis unit, generate a current excitation signal of a specific frequency based on the voltage signal, and output the current excitation signal to the digital-to-analog converter.

[0019] The digital-to-analog converter is used to convert the current excitation signal into an analog signal and output the analog signal to the transmitting electrode;

[0020] The analog-to-digital converter is used to convert the impedance signal generated by the bone into impedance parameters, and output the impedance parameters to the impedance analysis unit.

[0021] Furthermore, to achieve the above objectives, the present invention also proposes a fracture risk prediction method, which is applied to the fracture risk prediction device described above, and the fracture risk prediction method includes:

[0022] Impedance detection is performed on the bones in the area to be detected to obtain the impedance parameters corresponding to the bones.

[0023] Based on the impedance parameter, the bone component parameters corresponding to the impedance parameter are obtained by traversing the preset impedance spectrum. The preset impedance spectrum includes the proportional relationship between the impedance parameter and each component in the bone.

[0024] The bone composition parameters, patient-related factors such as gender, age, and body mass index are input into a preset fracture risk prediction model to obtain the fracture risk prediction result of the target to be detected. The preset fracture risk prediction model is trained based on bone parameters and fracture risk level.

[0025] Optionally, before performing impedance detection on the bone in the area to be detected to obtain the impedance parameters corresponding to the bone, the method further includes:

[0026] The bone sample is weighed to obtain its initial mass;

[0027] Impedance detection is performed on the bone sample to measure the impedance parameters corresponding to the bone sample. The impedance parameters include the initial dielectric constant and conductivity of the bone sample.

[0028] The water content of the bone sample is removed, and the water content of the bone sample is determined based on the mass of the bone sample after water removal and the initial mass.

[0029] The dehydrated bone sample was incinerated, and the organic matter content of the bone sample was determined based on the mass of the incineration residue and the initial mass.

[0030] Data analysis is performed based on the water content, the organic matter content, and the impedance parameters corresponding to the bone sample, and a preset electrical impedance spectrum corresponding to the bone sample is constructed based on the analysis results.

[0031] Optionally, the step of performing impedance detection on the bone sample to obtain the impedance parameters corresponding to the bone sample includes:

[0032] Multiple impedance measurements were performed on the bone sample to obtain multiple impedance parameters.

[0033] Compare the impedance parameters and determine whether the comparison results exceed the preset coefficient of variation;

[0034] When the comparison result exceeds the preset coefficient of variation, return to the step of performing impedance detection on the bone sample multiple times to obtain multiple impedance parameters, until the comparison result does not exceed the preset coefficient of variation;

[0035] When the comparison result does not exceed the preset coefficient of variation, the average value of each impedance parameter corresponding to the comparison result not exceeding the preset coefficient of variation is taken as the impedance parameter corresponding to the bone sample.

[0036] Optionally, before performing impedance detection on the bone in the area to be detected to obtain the impedance parameters corresponding to the bone, the method further includes:

[0037] Preprocess the sample data required for model training to obtain the target sample data;

[0038] The fracture risk level and predictive factors were determined based on the sample data.

[0039] The target sample data, the fracture risk level, and the predictive factors are input into an initial model constructed based on a logistic regression algorithm for training, thereby obtaining a preset fracture risk prediction model.

[0040] Optionally, the step of inputting the target sample data, the fracture risk level, and the predictive factors into an initial model constructed based on a logistic regression algorithm for training to obtain a preset fracture risk prediction model includes:

[0041] The target sample data, the fracture risk level, and the predictive factors are input into an initial model built based on a logistic regression algorithm for training to obtain a trained model.

[0042] The test dataset from the sample data is input into the trained model, and it is determined whether the output of the trained model meets the preset performance index.

[0043] When the output result does not reach the preset performance index, the parameters of the trained model are adjusted according to the output result until the output result reaches the preset performance index.

[0044] When the output reaches the preset performance index, the trained model corresponding to the output reaching the preset performance index is used as the preset fracture risk prediction model.

[0045] Furthermore, to achieve the above objectives, the present invention also proposes a fracture risk prediction device, the device comprising:

[0046] Impedance detection module is used to perform impedance detection on the bone in the area to be detected, and obtain the impedance parameters corresponding to the bone.

[0047] The parameter acquisition module is used to traverse a preset electrical impedance spectrum according to the impedance parameter to obtain the bone component parameters corresponding to the impedance parameter. The preset electrical impedance spectrum includes the proportional relationship between the impedance parameter and each component in the bone.

[0048] The risk prediction module is used to input the bone composition parameters, patient-related factors such as gender, age, and body mass index into a preset fracture risk prediction model to obtain the fracture risk prediction result of the target to be detected. The preset fracture risk prediction model is trained based on bone parameters and fracture risk level.

[0049] In addition, to achieve the above objectives, the present invention also proposes a storage medium storing a fracture risk prediction program, which, when executed by a processor, implements the steps of the fracture risk prediction method as described above.

[0050] This invention proposes a fracture risk prediction device, comprising an impedance measurement unit, an impedance conversion unit, and an impedance analysis unit. The impedance measurement unit's electrode contacts the skin at the bone site in the area to be tested. The output of the impedance measurement unit is connected to the input of the impedance conversion unit, and the output of the impedance conversion unit is connected to the input of the impedance analysis unit. The impedance measurement unit receives a weak current output from the impedance analysis unit, transmits the weak current to the skin at the bone site in the area to be tested, and outputs the impedance signal generated by the bone based on the weak current to the impedance conversion unit. The impedance conversion unit converts the impedance signal to obtain impedance parameters and outputs the impedance parameters to the impedance analysis unit. The impedance analysis unit analyzes and predicts based on the impedance parameters to obtain a fracture risk prediction result for the target bone. This invention obtains the impedance parameters of the bone in the area to be tested through the impedance measurement unit and derives the bone composition ratio of the bone through the impedance analysis unit, thereby obtaining a fracture risk prediction result. Compared to existing bone density assessment predictions, which only reflect 70% of bone strength, this method offers a significant advantage. This invention can serve as a supplement to traditional bone mineral density (BMD) assessment, and can also be used directly to reflect osteoporosis based on the bone tissue composition of the tested site, thereby improving the accuracy of fracture risk prediction. Attached Figure Description

[0051] Figure 1 This is a structural block diagram of the first embodiment of the fracture risk prediction device of the present invention;

[0052] Figure 2This is a schematic diagram of the structure of the second embodiment of the fracture risk prediction device of the present invention;

[0053] Figure 3 This is a schematic diagram of the impedance conversion unit in the second embodiment of the fracture risk prediction device of the present invention;

[0054] Figure 4 This is a schematic diagram of the process of fracture risk prediction by the device in the second embodiment of the fracture risk prediction device of the present invention;

[0055] Figure 5 This is a flowchart illustrating the first embodiment of the fracture risk prediction method of the present invention;

[0056] Figure 6 This is a flowchart illustrating the second embodiment of the fracture risk prediction method of the present invention;

[0057] Figure 7 This is a structural block diagram of the first embodiment of the fracture risk prediction device of the present invention.

[0058] The realization of the objective, functional features and advantages of the present invention will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0059] It should be understood that the specific embodiments described herein are for illustrative purposes only and are not intended to limit the scope of the invention.

[0060] This invention provides a fracture risk prediction device, referring to... Figure 1 , Figure 1 This is a structural block diagram of the first embodiment of the fracture risk prediction device of the present invention.

[0061] In this embodiment, the fracture risk prediction device includes an impedance measurement unit 10, an impedance conversion unit 20, and an impedance analysis unit 30.

[0062] In this embodiment, the electrode of the impedance measurement unit 10 is in contact with the skin at the bone of the area to be tested, the output terminal of the impedance measurement unit 10 is connected to the input terminal of the impedance conversion unit 20, and the output terminal of the impedance conversion unit 20 is connected to the input terminal of the impedance analysis unit 30.

[0063] The impedance measurement unit 10 is used to receive the weak current output by the impedance analysis unit 30, transmit the weak current to the skin of the bone in the area to be detected, and output the impedance signal generated by the bone based on the weak current to the impedance conversion unit 20.

[0064] It should be noted that the impedance measurement unit described above is a device used to measure the impedance signal of human bones. When measuring human bones, the electrode of the impedance measurement unit contacts the skin at the bone site, and the impedance signal of the human bones is obtained by applying a weak current.

[0065] It should be understood that impedance signals are the impedance of human bones, containing both resistance and capacitance signals. Since the human body contains impedance, this impedance includes the total impedance of the skin, blood, muscles, cell tissues, and their junctions, encompassing both resistance and capacitance. Due to the characteristics of the human body, many freely moving electric charges are distributed internally. When the human body comes into contact with direct current, these free charges begin to move directionally, accumulating positive and negative charges on both sides of the cell membrane, similar to capacitance. The impedance measurement unit can measure the impedance signals of human bones at different frequencies through frequency sweeping. Based on these different impedance signals, the proportions of various components in the bone of the tested area can be accurately predicted, thereby assessing fracture risk.

[0066] In a specific implementation, the impedance measurement unit receives the weak current output by the impedance analysis unit and transmits the weak current to the bone in the area to be detected. When the human bone comes into contact with the weak current, the free charges in the human body begin to move in a direction. Positive and negative charges accumulate on both sides of the cell membrane, which is similar to a capacitor, thereby obtaining an impedance signal, and outputting the impedance signal to the impedance conversion unit.

[0067] The impedance conversion unit 20 is used to convert the impedance signal to obtain impedance parameters and output the impedance parameters to the impedance analysis unit 30.

[0068] The impedance analysis unit 30 is used to perform analysis and prediction based on the impedance parameters to obtain the fracture risk prediction result of the target to be detected.

[0069] It should be noted that the impedance transformation unit is a device used to generate a weak current and perform analog-to-digital signal conversion. The impedance analysis unit is a device used to analyze and predict human skeletal structure based on impedance parameters, and to assess skeletal risks.

[0070] In the specific implementation, when the impedance measurement unit outputs the impedance signal to the impedance conversion unit, the impedance conversion unit converts the impedance signal into impedance parameters, which include dielectric constant and conductivity, reflecting the state of human bones. Since the impedance analysis unit pre-stores relevant electrical impedance spectra, which contain the correspondence between the proportions of water, organic matter, and inorganic matter in the bone and their dielectric constant and conductivity, the impedance analysis unit searches through the electrical impedance spectrum based on the impedance parameters to determine the proportions of water, organic matter, and inorganic matter corresponding to the bone in the area to be detected. Then, based on the determined bone data, it predicts the risk of fracture.

[0071] In this embodiment, the impedance measurement unit receives a weak current output from the impedance analysis unit and transmits it to the bone in the area to be tested. When the human bone comes into contact with the weak current, free charges within the body begin to move directionally, accumulating positive and negative charges on both sides of the cell membrane, similar to a capacitor, thus generating an impedance signal. This impedance signal is then output to the impedance conversion unit. When the impedance measurement unit outputs the impedance signal to the impedance conversion unit, the unit converts the impedance signal into impedance parameters, including dielectric constant and conductivity, which reflect the state of the human bone. Since the impedance analysis unit pre-stores relevant electrical impedance spectra, which contain the correspondence between the proportions of water, organic matter, and inorganic matter in the bone and their dielectric constant and conductivity, the impedance analysis unit searches through the electrical impedance spectrum based on the impedance parameters to determine the proportions of water, organic matter, and inorganic matter corresponding to the bone in the area to be tested. Based on the determined bone data, it then predicts the risk of fracture. This embodiment acquires the impedance parameters of the bone in the test area through an impedance measurement unit, and then uses an impedance analysis unit to determine the proportions of water, organic matter, and inorganic matter in the bone, thereby obtaining a fracture risk prediction result. Compared to existing bone mineral density (BMD) assessments, which only reflect 70% of bone strength, this invention can serve as a supplement to traditional BMD assessments and can also directly reflect osteoporosis based on the bone tissue composition of the test site, thus improving the accuracy of fracture risk prediction.

[0072] refer to Figure 2 , Figure 2 This is a schematic diagram of the second embodiment of the fracture risk prediction device of the present invention.

[0073] Based on the first embodiment described above, in this embodiment, considering the accuracy of the impedance measurement unit, such as Figure 2 As shown, the impedance measurement unit 10 includes a shielded chamber 11, a transmitting electrode 12, and a receiving electrode 13.

[0074] The transmitting electrode 12 and the receiving electrode 13 are disposed inside the shielding chamber 11; the input terminal of the transmitting electrode 12 is connected to the output terminal of the impedance conversion unit 20, and the output terminal of the transmitting electrode 12 is in contact with the skin at the bone of the area to be detected; the input terminal of the receiving electrode 13 is in contact with the skin at the other end of the bone, and the output terminal of the receiving electrode 13 is connected to the input terminal of the impedance conversion unit 20.

[0075] The shielding chamber 11 is used to shield interference signals from the external environment;

[0076] The transmitting electrode 12 is used to transmit the weak current transmitted by the impedance conversion unit 20 to the bone.

[0077] The receiving electrode 13 is used to receive the impedance signal generated by the bone and output it to the impedance conversion unit 20.

[0078] It should be noted that a shielded room can shield external signals to improve the accuracy of measurements.

[0079] In practice, with the external environmental interference signals shielded in the shielded room, the transmitting electrode is brought into contact with the skin of the bone in the area to be tested, and a weak current is transmitted into the bone. Then, the receiving electrode receives the impedance signal generated by the bone in real time and outputs it to the impedance conversion unit for conversion. Subsequently, the impedance analysis unit analyzes and predicts the signal, thereby obtaining more accurate impedance parameters and improving the accuracy of fracture risk prediction.

[0080] Furthermore, considering the safety of the impedance transformation unit in converting current, refer to Figure 3 , Figure 3 This is a schematic diagram of the impedance conversion unit in the second embodiment of the fracture risk prediction device of the present invention. Figure 3 As shown, the impedance conversion unit 20 in this embodiment includes: a signal generator 21, a digital-to-analog converter 22, and an analog-to-digital converter 23;

[0081] The input terminal of the signal generator 21 is connected to the output terminal of the impedance analysis unit 30, the output terminal of the signal generator 21 is connected to the input terminal of the digital-to-analog converter 22, the output terminal of the digital-to-analog converter 22 is connected to the input terminal of the transmitting electrode 12, the output terminal of the receiving electrode 13 is connected to the input terminal of the analog-to-digital converter 23, and the output terminal of the analog-to-digital converter 23 is connected to the input terminal of the impedance analysis unit 30.

[0082] The signal generator 21 is used to receive the voltage signal from the impedance analysis unit 30, generate a current excitation signal of a specific frequency according to the voltage signal, and output the current excitation signal to the digital-to-analog converter 22.

[0083] The digital-to-analog converter 22 is used to convert the current excitation signal into an analog signal and output the analog signal to the transmitting electrode 12;

[0084] The analog-to-digital converter 23 is used to convert the impedance signal generated by the bone into impedance parameters and output the impedance parameters to the impedance analysis unit 30.

[0085] It should be noted that the signal generator is used to generate a current excitation signal of a specific frequency based on the voltage signal. Under safe current conditions, the impedance signal of human bones at different frequencies can be measured, thereby obtaining more realistic data on human bones.

[0086] Understandably, the analog-to-digital converter converts the analog form of the impedance signal into a digital form, wherein the output of the analog-to-digital converter is connected to the input of the impedance analysis unit via a signal generator, and the impedance analysis unit makes predictions using the dielectric constant and conductivity in the impedance parameters.

[0087] To facilitate understanding, the process of measurement and prediction using fracture risk prediction equipment is explained, but this does not limit the scope of this approach. (Refer to...) Figure 4 , Figure 4 This is a schematic diagram illustrating the fracture risk prediction process of the device in the second embodiment of the present invention. First, patient information, frequency parameters, and the site to be tested are input into the impedance analysis unit. The impedance analysis unit adjusts the output current based on these parameters to obtain more accurate data while ensuring safety. The voltage data is transmitted to the signal generator of the impedance conversion unit. The signal generator then generates a specific frequency current signal and transmits it to the digital-to-analog converter (DAC). The DAC converts the digital signal into an analog signal and injects a weak current into the bone through the transmitting electrode. Next, the current in the bone flows out from the receiving electrode and is transmitted to the analog-to-digital converter (ADC). The ADC converts the feedback current into a digital signal and sends it to the impedance analysis unit for analysis and prediction. Since a model is pre-built in the impedance analysis unit, the prediction model is obtained by inputting prediction factors, building the model, allocating data, and outputting the results. Therefore, the impedance analysis unit compares the bone electrical impedance parameters and electrical impedance spectrum to determine the proportions of water, organic matter, and inorganic matter within the bone, thereby predicting fracture risk.

[0088] In this embodiment, under conditions of shielding against external environmental interference signals in a shielded room, the transmitting electrode contacts the skin at the bone site in the area to be tested, transmitting a weak current into the bone. The receiving electrode then receives the impedance signal generated by the bone in real time and outputs it to an impedance conversion unit for conversion. The impedance analysis unit then analyzes and predicts the signal, thereby obtaining more accurate impedance parameters and improving the accuracy of fracture risk prediction. Furthermore, the impedance analysis unit adjusts the output current based on patient information, frequency parameters, and the test site, obtaining more accurate data while ensuring safety. The voltage data is then transmitted to the signal generator of the impedance conversion unit. The signal generator generates a specific frequency current signal and transmits it to a digital-to-analog converter (DAC). The DAC converts the digital signal into an analog signal and injects a weak current into the bone through the transmitting electrode. The current in the bone then flows out from the receiving electrode and is transmitted to an analog-to-digital converter (ADC). The ADC converts the feedback current into a digital signal and sends it to the impedance analysis unit for analysis and prediction. By comparing the bone impedance parameters and impedance spectrum, the proportions of water, organic matter, and inorganic matter within the bone are determined, thereby predicting fracture risk. This non-invasive method allows for safe, rapid, and more accurate prediction of fracture risk.

[0089] Reference Figure 5 The present invention provides a fracture risk prediction system and a method for predicting fracture risk. Figure 5 This is a schematic flowchart of the first embodiment of the fracture risk prediction method of the present invention. The fracture risk prediction method is applied to the fracture risk prediction device described above.

[0090] In this embodiment, the fracture risk prediction method includes the following steps:

[0091] Step S10: Perform impedance detection on the bones in the area to be detected to obtain the impedance parameters corresponding to the bones.

[0092] Step S20: Traverse the preset impedance spectrum according to the impedance parameter to obtain the bone component parameters corresponding to the impedance parameter. The preset impedance spectrum includes the proportional relationship between the impedance parameter and each component in the bone.

[0093] It should be noted that impedance parameters are parameters such as dielectric constant and conductivity generated when the human body comes into contact with an electric current signal. The preset impedance spectrum is a pre-set spectrum reflecting the relationship between impedance parameters and the proportions of various components in the bone. Bone composition parameters reflect the composition of human bone, such as water, organic matter, inorganic matter, and their corresponding proportions. Based on the measured impedance parameters, the preset impedance spectrum can be traversed to determine the proportions of water, organic matter, and inorganic matter in the bone of the target object, allowing for accurate acquisition of human bone data in a non-invasive manner.

[0094] Step S30: Input the bone composition parameters, patient-related factors such as gender, age, and body mass index into the preset fracture risk prediction model to obtain the fracture risk prediction result of the target to be detected. The preset fracture risk prediction model is obtained by training based on bone parameters and fracture risk level.

[0095] It should be noted that the preset fracture risk prediction model is a pre-built model used to predict fracture risk based on bone composition parameters. Predictive factors in the model may include frequency, the site of fracture, and factors associated with fracture risk, such as gender, age, and body mass index. The prediction results can be set with different risk levels based on different bone composition parameters, for example, including 0, 1, and 2. 0 represents low fracture risk, 1 represents osteoporosis, and 2 represents high fracture risk; other levels are also possible, but this embodiment does not impose any limitations on this.

[0096] In practice, bone composition parameters, patient-related factors such as gender, age, and body mass index are input into a preset fracture risk prediction model, and the fracture risk prediction result of the target to be detected can be obtained directly, such as the fracture risk level mentioned above.

[0097] This embodiment first performs impedance testing on the bone in the area to be tested to obtain the impedance parameters corresponding to the bone. Then, based on the impedance parameters, it traverses a preset impedance spectrum to obtain the bone composition parameters corresponding to the impedance parameters. Finally, the bone composition parameters, patient-related fracture risk factors such as gender, age, and body mass index are input into a preset fracture risk prediction model to obtain the fracture risk prediction result for the target. The preset fracture risk prediction model is trained based on bone parameters and fracture risk levels. This embodiment predicts fracture risk by measuring the dielectric constant and conductivity of human bone and using the impedance spectrum to derive the proportion of water, organic matter, and inorganic matter in that part of the bone based on the dielectric constant and conductivity. Compared to existing bone mineral density (BMD) assessments, which only reflect 70% of bone strength, this invention can serve as a supplement to traditional BMD assessments and can also directly reflect osteoporosis based on the bone tissue composition of the tested area, thus improving the accuracy of fracture risk prediction.

[0098] refer to Figure 6 , Figure 6 This is a flowchart illustrating the second embodiment of the fracture risk prediction method of the present invention.

[0099] Based on the above Figure 5 The first embodiment shown, in this embodiment, considering the accuracy of the preset impedance spectrum, includes the following step before step S10:

[0100] Step S01: Weigh the bone sample to obtain its initial mass.

[0101] It should be noted that the preset impedance spectrum can be obtained using bioelectrical impedance measurement techniques, such as measuring bone parameters of different bone samples in a laboratory. The initial mass of the bone sample includes the mass of different substances such as water, inorganic salts, and organic matter.

[0102] Step S02: Perform impedance detection on the bone sample to measure the impedance parameters corresponding to the bone sample. The impedance parameters include the initial dielectric constant and conductivity of the bone sample.

[0103] Step S03: Remove water from the bone sample and determine the water content of the bone sample based on the mass of the bone sample after water removal and the initial mass.

[0104] Step S04: Incinerate the dehydrated bone sample and determine the organic matter content of the bone sample based on the mass of the incineration residue and the initial mass.

[0105] It should be noted that impedance detection can be performed on bone samples at a certain frequency to measure the dielectric constant and conductivity of the bone samples. To ensure data accuracy, multiple measurements can be taken, and the precise dielectric constant and conductivity can be determined based on the average value. Other methods can also be used, and this embodiment does not limit this.

[0106] Understandably, to accurately determine the moisture content in bones, bone samples can be dried, and the moisture content of the bone sample can be determined by reverse measurement based on the remaining mass.

[0107] It should be understood that by incinerating bone samples, the weight ratio of organic to inorganic matter can be determined from the incineration residue.

[0108] Step S05: Perform data analysis based on the water content, the organic matter content, and the impedance parameters corresponding to the bone sample, and construct a preset electrical impedance spectrum corresponding to the bone sample based on the analysis results.

[0109] In practical implementation, the above steps can be repeated for multiple bone samples, and then data fitting can be performed based on the acquired data. The content of moisture, organic matter, and inorganic matter at a certain frequency is considered as three independent variables, while conductivity and dielectric constant are considered as two separate dependent variables. These are analyzed with the independent variables to establish a preset impedance spectrum that reflects the proportional relationship between dielectric constant and conductivity and the various components in the bone.

[0110] Furthermore, considering the accuracy of obtaining impedance parameters, step S02 in this embodiment includes: performing impedance detection on the bone sample multiple times to obtain multiple impedance parameters; comparing each impedance parameter and determining whether the comparison result exceeds a preset coefficient of variation; when the comparison result exceeds the preset coefficient of variation, returning to the step of performing impedance detection on the bone sample multiple times to obtain multiple impedance parameters, until the comparison result does not exceed the preset coefficient of variation; when the comparison result does not exceed the preset coefficient of variation, taking the average value of each impedance parameter corresponding to the comparison result not exceeding the preset coefficient of variation as the impedance parameter corresponding to the bone sample.

[0111] It should be noted that the preset coefficient of variation is a pre-set threshold. For example, due to the influence of the external environment and the accuracy of the experimental equipment, impedance testing of bone samples may yield different data. To ensure the accuracy of the impedance parameters, a preset coefficient of variation can be set. If the fluctuation range of the impedance parameters measured multiple times exceeds the preset coefficient of variation, the error is too large and remeasurement is required.

[0112] In the specific implementation, impedance detection is performed multiple times on the bone sample to obtain multiple impedance parameters. These parameters are then compared to determine if the fluctuation range exceeds a preset coefficient of variation. If the fluctuation range exceeds the preset coefficient of variation, the detection is repeated until the fluctuation range falls within the preset coefficient of variation. The average value of all impedance parameters is then used as the impedance parameter corresponding to the bone sample. This allows for the acquisition of more accurate impedance parameters such as dielectric constant and conductivity, improving the accuracy of the preset impedance spectrum.

[0113] For ease of understanding, the process of establishing a preset electrical impedance spectrum in the laboratory is used as an example, but this method is not limited. Measurements are performed using laboratory-extracted bone sample data, where cortical bone and cancellous bone can be measured separately. The main steps include:

[0114] (1) First, use a dental grinder to grind the bone sample to a width and height of 4mm and a length of 6mm. During the grinding process, use PBS solution to spray and wash to prevent the sample from overheating and losing moisture. Take out the sample and measure its weight, which is recorded as the initial mass.

[0115] (2) Measure the dielectric constant and conductivity of the bone sample: Measure the dielectric constant and conductivity at a frequency of 20Hz-120MHz, take three consecutive measurements and average the results, complete the measurement within 2 minutes, and the coefficient of variation of the three measurements should not exceed 5%.

[0116] (3) After the measurement is completed, place it in a 180℃ oven to dry for 3 hours, which is considered to remove the moisture.

[0117] (4) Weigh the water to obtain the weight of the water and divide it by the original weight to obtain the percentage of water. Then put it into a muffle furnace for incineration to obtain the weight of organic and inorganic matter. Divide it by the original weight to obtain the percentage of organic and inorganic matter.

[0118] (5) Repeat the above steps for multiple samples, and then fit the data. At a certain frequency, the content of moisture, organic matter, and inorganic matter is regarded as three independent variables, and the conductivity and dielectric constant are regarded as two separate dependent variables and analyzed with the independent variables respectively.

[0119] (6) Establish an impedance spectrum that reflects the relationship between dielectric constant and conductivity and the proportion of various components in the bone.

[0120] (7) Basic formula for impedance measurement:

[0121] Z = |Z|e jθ =|Z|cosθ+j|Z|sinθ;

[0122] Where Z represents the magnitude of the impedance and θ represents the phase of the impedance, the wet bone sample is observed to have capacitive properties from the obtained impedance measurements, which can be represented by an equivalent circuit.

[0123]

[0124] Where R and C represent the resistance and capacitance of the bone sample, respectively, and f represents the frequency.

[0125] R = |Z| / cosθ;

[0126] C = -sinθ / (2πf|Z|);

[0127] The electrical conductivity and dielectric constant of the bone sample can be obtained from the above calculations:

[0128]

[0129]

[0130] Where σ is the conductivity, ε ′ is the dielectric constant, L is the length of the measured bone sample, and A is the surface area of ​​the measured bone sample. A preset impedance spectrum is constructed based on the relationship between the content of organic matter, inorganic matter, and water in the bone sample and its conductivity and dielectric constant.

[0131] Furthermore, considering the efficiency of fracture risk prediction, this embodiment further includes the following steps before step S10: preprocessing the sample data required for model training to obtain target sample data; determining the fracture risk level and predictive factors based on the sample data; and inputting the target sample data, the fracture risk level, and the predictive factors into an initial model constructed based on a logistic regression algorithm for training to obtain a preset fracture risk prediction model.

[0132] It should be noted that the fracture risk level is determined based on different risk levels according to bone composition parameters, such as three levels: 0, 1, and 2. Among them, 0 represents low fracture risk, 1 represents osteoporosis, and 2 represents high fracture risk. Other levels are also possible, but this embodiment does not limit them.

[0133] Understandably, predictors are other factors that influence fracture risk, such as the frequency of impedance measurement, the site of the test, and factors associated with fracture risk, such as gender, age, and body mass index.

[0134] It should be understood that logistic regression is a classification machine learning algorithm. It is simple, efficient, fast in prediction, and easy to learn and understand. Logistic regression is a binary classification algorithm that can mainly solve discrete binary classification prediction problems.

[0135] In practical implementation, the sample data required for model training can be preprocessed, such as through data cleaning, feature extraction, and feature selection. Simultaneously, data denoising, outlier handling, and normalization can be performed to ensure data quality and consistency, yielding the target sample data. Then, based on the sample data, the fracture risk level and predictive factors are determined. The target sample data, fracture risk level, and predictive factors are then input into an initial model built based on a logistic regression algorithm for training, resulting in a preset fracture risk prediction model. This preset fracture risk prediction model can directly predict the risk level based on bone composition parameters, thereby improving the efficiency of fracture risk prediction.

[0136] Furthermore, considering the accuracy of the preset fracture risk prediction model, in this embodiment, the step of inputting the target sample data, the fracture risk level, and the predictive factors into an initial model constructed based on a logistic regression algorithm for training to obtain the preset fracture risk prediction model includes: inputting the target sample data, the fracture risk level, and the predictive factors into an initial model constructed based on a logistic regression algorithm for training to obtain a trained model; inputting the test dataset from the sample data into the trained model and determining whether the output result of the trained model reaches a preset performance index; if the output result does not reach the preset performance index, adjusting the parameters of the trained model according to the output result until the output result reaches the preset performance index; and when the output result reaches the preset performance index, using the trained model corresponding to the output result reaching the preset performance index as the preset fracture risk prediction model.

[0137] It should be noted that the test dataset is the dataset used to accurately test the trained model from the sample data. Preset performance metrics are parameters that measure the model's performance, such as accuracy, precision, recall, and F1 score, to evaluate the model's predictive performance.

[0138] In the specific implementation, the target sample data, fracture risk level, and predictive factors are input into an initial model built based on a logistic regression algorithm for training, resulting in a trained model. Then, the test dataset from the sample data is input into the trained model, and the output results are evaluated. Performance metrics such as accuracy, precision, recall, and F1 score are calculated to assess the model's predictive performance. If the output results do not meet the preset performance metrics, the parameters of the trained model can be adjusted, such as adjusting the parameters, trying different feature combinations, and addressing overfitting, to improve the model's performance. This process continues until the preset performance metrics are met. The trained model that achieves the preset performance metrics is then used as the preset fracture risk prediction model to improve its accuracy.

[0139] For ease of understanding, the process of building the model will be used as an example, but this approach is not limited. First, predictive factors are input, which may include frequency, the site of fracture, and factors associated with fracture risk such as gender, age, and body mass index. Then, the model is built and trained:

[0140] (1) Data preparation. Data were obtained from laboratory ex vivo bone samples and from medical literature and databases to ensure that the samples included sufficient fracture patients and non-patients.

[0141] (2) Data preprocessing. The collected data is preprocessed, including data cleaning, feature extraction, and feature selection. Simultaneously, noise reduction, outlier handling, and normalization are performed to ensure data quality and consistency.

[0142] (3) Data partitioning. The dataset is divided into training and testing sets for training and evaluating the model. Approximately 70-80% of the data is used for training, and the remaining 20-30% is used for testing.

[0143] (4) Model training. The logistic regression algorithm is used to input the features and labels of the training set into the model and train the model. The parameters of the model are estimated by the optimization algorithm to obtain a model that can predict the risk of fracture.

[0144] (5) Model evaluation. The trained model is evaluated using a test set, and performance metrics such as accuracy, precision, recall, and F1 score are calculated to assess the model's predictive performance.

[0145] (6) Model tuning. Based on the evaluation results, the model can be tuned, such as adjusting the model parameters, trying different feature combinations, and solving the overfitting problem, in order to improve the model performance.

[0146] (7) Model Validation. After the model training and tuning are completed, the model is validated using an independent validation set to test its generalization performance.

[0147] (8) Model Application. After the model is validated, it can be applied to actual fracture risk prediction. New patient information and features such as frequency and target site are input, and the model outputs the patient's fracture risk prediction results. The prediction results can contain two values: 0 and 1. 0 indicates low fracture risk, and 1 indicates high fracture risk. 0 and 1 are derived from the electrical impedance spectrum obtained based on the ratio thresholds of organic matter, inorganic matter, and water.

[0148] This embodiment first uses a frequency sweep method to measure the dielectric constant and conductivity of the bone sample. Simultaneously, it uses drying and calcination methods to measure the proportions of various components in the bone, fitting these measurements into an equation to obtain a preset impedance spectrum. Then, by measuring the dielectric constant and conductivity of human bone, the proportions of water, organic matter, and inorganic matter in that portion of the bone are derived from the preset impedance spectrum, thereby predicting fracture risk. This non-invasive method allows for safe, rapid, and more accurate prediction of fracture risk. Furthermore, multiple impedance tests can be performed on the bone sample to obtain multiple impedance parameters. These parameters are then compared to determine if the fluctuation range exceeds a preset coefficient of variation. If the fluctuation range exceeds the preset coefficient of variation, the test is repeated until the fluctuation range falls within the preset coefficient of variation. The average value of these impedance parameters is then used as the corresponding impedance parameter for the bone sample. This allows for obtaining more accurate impedance parameters such as dielectric constant and conductivity, improving the accuracy of the preset impedance spectrum.

[0149] Furthermore, this embodiment of the invention also proposes a storage medium storing a fracture risk prediction program, which, when executed by a processor, implements the steps of the fracture risk prediction method described above.

[0150] Reference Figure 7 , Figure 7 This is a structural block diagram of the first embodiment of the fracture risk prediction device of the present invention.

[0151] like Figure 7 As shown, the fracture risk prediction device proposed in this embodiment of the invention includes:

[0152] Impedance detection module 701 is used to perform impedance detection on the bone in the area to be detected and obtain the impedance parameters corresponding to the bone.

[0153] The parameter acquisition module 702 is used to traverse a preset impedance spectrum according to the impedance parameter to obtain the bone component parameter corresponding to the impedance parameter. The preset impedance spectrum includes the proportional relationship between the impedance parameter and each component in the bone.

[0154] The risk prediction module 703 is used to input the bone composition parameters, patient-related factors such as gender, age, and body mass index into a preset fracture risk prediction model to obtain the fracture risk prediction result of the target to be detected. The preset fracture risk prediction model is trained based on bone parameters and fracture risk level.

[0155] This embodiment first performs impedance testing on the bone in the area to be tested to obtain the impedance parameters corresponding to the bone. Then, based on the impedance parameters, it traverses a preset impedance spectrum to obtain the bone composition parameters corresponding to the impedance parameters. Finally, the bone composition parameters, patient-related fracture risk factors such as gender, age, and body mass index are input into a preset fracture risk prediction model to obtain the fracture risk prediction result for the target. The preset fracture risk prediction model is trained based on bone parameters and fracture risk levels. This embodiment predicts fracture risk by measuring the dielectric constant and conductivity of human bone and using the impedance spectrum to derive the proportion of water, organic matter, and inorganic matter in that part of the bone based on the dielectric constant and conductivity. Compared to existing bone mineral density (BMD) assessments, which only reflect 70% of bone strength, this invention can serve as a supplement to traditional BMD assessments and can also directly reflect osteoporosis based on the bone tissue composition of the tested area, thus improving the accuracy of fracture risk prediction.

[0156] Based on the first embodiment of the fracture risk prediction device of the present invention, a second embodiment of the fracture risk prediction device of the present invention is proposed.

[0157] In this embodiment, the fracture risk prediction device further includes an impedance spectrum acquisition module; the impedance spectrum acquisition module is used to weigh the bone sample to obtain the initial mass of the bone sample; perform impedance detection on the bone sample to obtain the impedance parameter corresponding to the bone sample; remove water from the bone sample and determine the water content of the bone sample based on the mass of the bone sample after water removal and the initial mass; incinerate the bone sample after water removal and determine the organic content of the bone sample based on the mass of the incineration residue and the initial mass; perform data analysis based on the water content, the organic content, and the impedance parameter corresponding to the bone sample, and construct a preset impedance spectrum corresponding to the bone sample based on the analysis results.

[0158] Furthermore, the impedance acquisition module is also used to perform impedance detection on the bone sample multiple times to obtain multiple impedance parameters; compare each impedance parameter and determine whether the comparison result exceeds a preset coefficient of variation; when the comparison result exceeds the preset coefficient of variation, return to the step of performing impedance detection on the bone sample multiple times to obtain multiple impedance parameters, until the comparison result does not exceed the preset coefficient of variation; when the comparison result does not exceed the preset coefficient of variation, take the average value of each impedance parameter corresponding to the comparison result not exceeding the preset coefficient of variation as the impedance parameter corresponding to the bone sample.

[0159] Furthermore, the fracture risk prediction device also includes a model building module; the model building module is used to preprocess the sample data required for model training to obtain target sample data; determine the fracture risk level and predictive factors based on the sample data; and input the target sample data, the fracture risk level and the predictive factors into an initial model built based on a logistic regression algorithm for training to obtain a preset fracture risk prediction model.

[0160] Furthermore, the model building module is also used to input the target sample data, the fracture risk level, and the predictive factors into an initial model built based on a logistic regression algorithm for training to obtain a trained model; input the test dataset from the sample data into the trained model and determine whether the output of the trained model reaches a preset performance index; if the output does not reach the preset performance index, adjust the parameters of the trained model according to the output until the output reaches the preset performance index; when the output reaches the preset performance index, use the trained model corresponding to the output reaching the preset performance index as the preset fracture risk prediction model.

[0161] Other embodiments or specific implementations of the fracture risk prediction device of the present invention can be referred to the above-described method embodiments, and will not be repeated here.

[0162] It should be noted that, in this document, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or system that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or system. Unless otherwise specified, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or system that includes that element.

[0163] The sequence numbers of the above embodiments of the present invention are for descriptive purposes only and do not represent the superiority or inferiority of the embodiments.

[0164] Through the above description of the embodiments, those skilled in the art can clearly understand that the methods of the above embodiments can be implemented by means of software plus necessary general-purpose hardware platforms. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product is stored in a storage medium (such as read-only memory / random access memory, magnetic disk, optical disk) and includes several instructions to cause a terminal device (which may be a mobile phone, computer, server, or network device, etc.) to execute the methods described in the various embodiments of the present invention.

[0165] The above are merely preferred embodiments of the present invention and do not limit the scope of the patent. Any equivalent structural or procedural transformations made based on the description and drawings of the present invention, or direct or indirect applications in other related technical fields, are similarly included within the scope of patent protection of the present invention.

Claims

1. A fracture risk prediction device, characterized in that, The fracture risk prediction device includes an impedance measurement unit, an impedance conversion unit, and an impedance analysis unit. The impedance measurement unit has its electrode contacting the skin at the bone in the area to be tested. The output of the impedance measurement unit is connected to the input of the impedance conversion unit, and the output of the impedance conversion unit is connected to the input of the impedance analysis unit. The impedance measurement unit is used to receive the weak current output by the impedance analysis unit, transmit the weak current to the skin of the bone in the area to be detected, and output the impedance signal generated by the bone based on the weak current to the impedance conversion unit. The impedance conversion unit is used to convert the impedance signal to obtain impedance parameters and output the impedance parameters to the impedance analysis unit. The impedance parameters include dielectric constant and conductivity, which are used to reflect the state of human bones. The impedance analysis unit is used to traverse a preset impedance spectrum according to the impedance parameter, obtain the bone composition parameter corresponding to the impedance parameter, and make a prediction based on the bone composition parameter to obtain the fracture risk prediction result; wherein, the bone composition parameter includes the ratio information of water, organic matter and inorganic matter in the bone; The process involves: weighing the bone sample to obtain its initial mass; performing impedance detection on the bone sample to measure its impedance parameters, including the initial dielectric constant and conductivity; removing water from the bone sample and determining its water content based on the mass of the dehydrated bone sample and the initial mass; incinerating the dehydrated bone sample and determining its organic content based on the mass of the incineration residue and the initial mass; performing data analysis based on the water content, organic content, and impedance parameters of the bone sample, and constructing a preset impedance spectrum based on the analysis results.

2. The fracture risk prediction device as described in claim 1, characterized in that, The impedance measurement unit includes a shielded chamber, a transmitting electrode, and a receiving electrode; The transmitting electrode and the receiving electrode are disposed inside the shielding chamber; the input terminal of the transmitting electrode is connected to the output terminal of the impedance conversion unit, and the output terminal of the transmitting electrode is in contact with the skin at the bone of the area to be detected; the input terminal of the receiving electrode is in contact with the skin at the other end of the bone, and the output terminal of the receiving electrode is connected to the input terminal of the impedance conversion unit. The shielding room is used to shield interference signals from the external environment; The transmitting electrode is used to transmit the weak current transmitted by the impedance conversion unit to the bone. The receiving electrode is used to receive the impedance signal generated by the bone and output it to the impedance conversion unit.

3. The fracture risk prediction device as described in claim 2, characterized in that, The impedance conversion unit includes: a signal generator, a digital-to-analog converter, and an analog-to-digital converter; The input terminal of the signal generator is connected to the output terminal of the impedance analysis unit, the output terminal of the signal generator is connected to the input terminal of the digital-to-analog converter, the output terminal of the digital-to-analog converter is connected to the input terminal of the transmitting electrode, the output terminal of the receiving electrode is connected to the input terminal of the analog-to-digital converter, and the output terminal of the analog-to-digital converter is connected to the input terminal of the impedance analysis unit. The signal generator is used to receive the voltage signal from the impedance analysis unit, generate a current excitation signal of a specific frequency based on the voltage signal, and output the current excitation signal to the digital-to-analog converter. The digital-to-analog converter is used to convert the current excitation signal into an analog signal and output the analog signal to the transmitting electrode; The analog-to-digital converter is used to convert the impedance signal generated by the bone into impedance parameters, and output the impedance parameters to the impedance analysis unit.

4. A computer-based method for predicting fracture risk, characterized in that, The fracture risk prediction method is applied to the fracture risk prediction device as described in any one of claims 1 to 3, and the fracture risk prediction method includes: Impedance detection is performed on the bones in the area to be detected to obtain the impedance parameters corresponding to the bones. Based on the impedance parameter, the bone component parameters corresponding to the impedance parameter are obtained by traversing the preset impedance spectrum. The preset impedance spectrum includes the proportional relationship between the impedance parameter and each component in the bone. The bone composition parameters and patient-related factors associated with fracture risk are input into a preset fracture risk prediction model to obtain the fracture risk prediction result of the area to be detected. The preset fracture risk prediction model is trained based on bone composition parameters and fracture risk level. The patient-related factors associated with fracture risk include gender, age and body mass index. Before performing impedance detection on the bone in the area to be detected to obtain the impedance parameters corresponding to the bone, the method further includes: The bone sample is weighed to obtain its initial mass; Impedance detection is performed on the bone sample to measure the impedance parameters corresponding to the bone sample. The impedance parameters include the initial dielectric constant and conductivity of the bone sample. The water content of the bone sample is removed, and the water content of the bone sample is determined based on the mass of the bone sample after water removal and the initial mass. The dehydrated bone sample was incinerated, and the organic matter content of the bone sample was determined based on the mass of the incineration residue and the initial mass. Data analysis is performed based on the water content, the organic matter content, and the impedance parameters corresponding to the bone sample, and the preset electrical impedance spectrum is constructed based on the analysis results.

5. The fracture risk prediction method as described in claim 4, characterized in that, The process of performing impedance detection on the bone sample to obtain the impedance parameters corresponding to the bone sample includes: Multiple impedance measurements were performed on the bone sample to obtain multiple impedance parameters. Compare the impedance parameters and determine whether the comparison results exceed the preset coefficient of variation; When the comparison result exceeds the preset coefficient of variation, return to the step of performing impedance detection on the bone sample multiple times to obtain multiple impedance parameters, until the comparison result does not exceed the preset coefficient of variation; When the comparison result does not exceed the preset coefficient of variation, the average value of each impedance parameter corresponding to the comparison result not exceeding the preset coefficient of variation is taken as the impedance parameter corresponding to the bone sample.

6. The fracture risk prediction method as described in claim 4, characterized in that, Before performing impedance detection on the bone in the area to be detected to obtain the impedance parameters corresponding to the bone, the method further includes: Preprocess the sample data required for model training to obtain the target sample data; The fracture risk level and predictive factors were determined based on the sample data. The target sample data, the fracture risk level, and the predictive factors are input into an initial model constructed based on a logistic regression algorithm for training, thereby obtaining a preset fracture risk prediction model.

7. The fracture risk prediction method as described in claim 6, characterized in that, The step of inputting the target sample data, the fracture risk level, and the predictive factors into an initial model constructed based on a logistic regression algorithm for training to obtain a preset fracture risk prediction model includes: The target sample data, the fracture risk level, and the predictive factors are input into an initial model built based on a logistic regression algorithm for training to obtain a trained model. The test dataset from the sample data is input into the trained model, and it is determined whether the output of the trained model meets the preset performance index. When the output result does not reach the preset performance index, the parameters of the trained model are adjusted according to the output result until the output result reaches the preset performance index. When the output reaches the preset performance index, the trained model corresponding to the output reaching the preset performance index is used as the preset fracture risk prediction model.

8. A fracture risk prediction device, characterized in that, The fracture risk prediction device performs the fracture risk prediction method as described in claim 4, the device comprising: Impedance detection module is used to perform impedance detection on the bone in the area to be detected, and obtain the impedance parameters corresponding to the bone. The parameter acquisition module is used to traverse a preset impedance spectrum according to the impedance parameter to obtain the bone component parameters corresponding to the impedance parameter. The preset impedance spectrum includes the proportional relationship between the impedance parameter and each component in the bone. The risk prediction module is used to input the bone composition parameters and patient-related factors into a preset fracture risk prediction model to obtain the fracture risk prediction result of the area to be detected. The preset fracture risk prediction model is trained based on bone composition parameters and fracture risk level. The patient-related factors include gender, age and body mass index. An impedance spectrum acquisition module is used to weigh the bone sample to obtain its initial mass; perform impedance detection on the bone sample to measure its corresponding impedance parameters, including the initial dielectric constant and conductivity of the bone sample; remove water from the bone sample and determine the water content of the bone sample based on the mass of the dehydrated bone sample and the initial mass; incinerate the dehydrated bone sample and determine the organic content of the bone sample based on the mass of the incineration residue and the initial mass; perform data analysis based on the water content, the organic content, and the corresponding impedance parameters of the bone sample, and construct the preset impedance spectrum based on the analysis results.

9. A storage medium, characterized in that, The storage medium stores a fracture risk prediction program, which, when executed by a processor, implements the steps of the fracture risk prediction method as described in any one of claims 4 to 7.

Images