Construction method and equipment of fracture risk prediction model, medium and program product

Abstract

The invention provides a construction method and equipment of a fracture risk prediction model, a medium and a program product, and relates to the field of intelligent medical treatment. The method comprises the following steps: acquiring basic data of a T2DM osteoporosis patient as a training set sample; the training set samples are divided into a fracture group and a fracture-free group; performing statistical analysis on the basic data, representing continuous variables conforming to normal distribution by using mean + / -standard deviation, and comparing the difference between the two groups through independent sample t inspection; adopting median and quartile to describe continuous variables which do not conform to normal distribution; through single-factor and multi-factor analysis, screening fracture influence characteristics of the osteoporosis patient from the basic data; and constructing a prediction model of the fracture risk of the osteoporosis patient based on the prediction variables. A special group of postmenopausal female diabetes is used as a research object, difference characteristics are analyzed, a prediction model is constructed, and early prevention and treatment of osteoporosis fracture patients are achieved.

Description

Technical Field

[0001] The present invention relates to the field of intelligent medicine, and more particularly, to a method, device, medium and program product for constructing a fracture risk prediction model. Background Art

[0002] Osteoporotic fractures belong to brittle fractures and occur when bone strength decreases and bone brittleness increases. Osteoporotic fractures usually occur under low-energy external forces or during daily activities and are serious consequences of osteoporosis. The incidence of osteoporotic fractures is higher in postmenopausal women and diabetic patients. Once an osteoporotic fracture occurs, the prognosis of the patient is significantly worse than that of ordinary traumatic fractures. Therefore, correctly assessing the risk of osteoporotic fractures is beneficial to the prevention and treatment of osteoporotic fractures.

[0003] Risk assessment is an important means of identifying high-risk populations of osteoporotic fractures. Currently commonly used self-assessment tools include the Fracture Risk Assessment Tool (FRAX), the Garvan Fracture Risk Calculator (Garvan), the QFracture Scores (QFracture), etc. The screening AUC of these tools is 0.58 - 0.82 and they are widely used in population primary screening. These tools do not take into account the obvious regional differences in economic levels, environmental conditions, ethnic compositions, incidence rates and mortality rates of various fractures in different countries and regions. Moreover, due to the limited global epidemiological data in this regard, not all regions can apply them. Garva is a fracture risk assessment model established by the Bone and Mineral Research Project Team of the Garvan Institute of Medical Research in Australia based on DOES data using the Cox proportional hazards model analysis method. This model is based on a single cohort study of people over 60 years old in Australia only. Therefore, relevant validity studies should be carried out before applying it in regions other than Australia, which limits the wide applicability of this tool. The QFracture scoring tool is mainly applicable to the UK population, thus also limiting the wide use of this tool.

[0004] In summary, on the one hand, the above-mentioned evaluation models are for the risk assessment of osteoporosis patients in some foreign countries, not for the risk assessment of osteoporotic fracture patients in the domestic population, especially not for the risk assessment of osteoporotic fractures in postmenopausal women and diabetic patients. On the other hand, these evaluations are all self-assessment scales for patients, and the evaluation results may be inaccurate due to some subjective factors during the self-assessment process of patients. Therefore, it is of great significance to develop a new evaluation model for the risk of osteoporotic fractures. Considering the special group of postmenopausal women with diabetes, establishing an evaluation model for the risk of osteoporotic fractures in postmenopausal women with diabetes is of great significance for the prevention and treatment of osteoporotic fracture patients. Summary of the Invention

[0005] The present invention aims to solve at least one of the technical problems existing in the prior art. To this end, the present invention provides a method, device, medium and program product for constructing a fracture risk prediction model; the method of the present invention takes the special group of postmenopausal women with diabetes as the research object, analyzes the differential characteristics to construct a prediction model, and realizes the early prevention and treatment of osteoporotic fracture patients.

[0006] The first aspect of the present application discloses a method for constructing a fracture risk prediction model, the method comprising:

[0007] S101, obtaining the basic data of T2DM osteoporosis patients as the training set samples; the training set samples are divided into a fracture group and a non-fracture group;

[0008] S102, performing statistical analysis on the basic data. For continuous variables that conform to the normal distribution, they are expressed by the mean ± standard deviation, and the differences between the two groups are compared by independent sample t-test; for continuous variables that do not conform to the normal distribution, the median (M) and quartiles (Q1, Q3) are used to describe;

[0009] S103, screening the influencing characteristics of osteoporotic fractures in patients from the basic data through univariate and multivariate analysis;

[0010] S104, using the Akaike information for backward stepwise selection to select the prediction variables;

[0011] S105, constructing a prediction model for the risk of osteoporotic fractures in patients based on the prediction variables.

[0012] In some embodiments, the prediction model is in the form of a nomogram.

[0013] In some embodiments, the inclusion criteria for the training set samples include: women of any age who have been naturally postmenopausal for more than 1 year; meeting the diagnostic criteria for type 2 diabetes; BMI between 18.5 and 24.

[0014] In some embodiments, the exclusion criteria for the training set samples include: unnatural postmenopausal women who have undergone hysterectomy; patients suffering from diseases affecting bone metabolism; patients taking drugs affecting bone metabolism; type 1 diabetes, special types of diabetes.

[0015] In some embodiments, the diseases affecting bone metabolism include any one or more of the following: Cushing's syndrome, hyperthyroidism, primary hyperparathyroidism, rheumatoid arthritis, leukemia, multiple myeloma, subtotal gastrectomy, chronic hepatitis, congenital bone metabolism disorders and other endocrine diseases.

[0016] The second aspect of the present application discloses a method for predicting the risk of fracture in osteoporosis patients, the method comprising:

[0017] S201, obtaining predictors of the subject including age, menopause age, diabetes duration, PINP, CTX, FF score; the subject being a T2DM osteoporosis patient;

[0018] S202, calculating a risk score based on the predictors, and outputting an auxiliary prediction result of the probability of the subject having a fracture risk according to the risk score; when the risk score is greater than the first threshold, outputting an auxiliary prediction result that the subject has a high probability of having a fracture risk; when the risk score is less than the first threshold, outputting an auxiliary prediction result that the subject has a low probability of having a fracture risk.

[0019] In some embodiments, the method further includes: recommending a treatment suggestion of combined bone density and TBS detection every 6 months for patients with age > 60 years, early menopause age < 47 years, diabetes duration > 10 years, CTX > 500 pg / mL and FF > 75%.

[0020] The third aspect of the present application discloses a computer device, the device comprising: a memory and a processor; the memory is used for storing a computer program; the processor executes the computer program to implement the steps of the above method.

[0021] The fourth aspect of the present application discloses a computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, the steps of the above method are implemented.

[0022] The fifth aspect of the present application discloses a computer program product, including a computer program, and when the computer program is executed by a processor, the steps of the above method are implemented.

[0023] The present application has the following beneficial effects: The present application innovatively discloses a method for constructing a fracture risk prediction model and a method for predicting the fracture risk of osteoporosis patients for the postmenopausal T2DM population, and finds that age, menopause age, diabetes duration, PINP, CTX, and FF score are independent risk factors for osteoporosis-related fractures in patients. A nomogram is established based on factors such as age, menopause age, diabetes duration, PINP, CTX, and FF score. This model has a high evaluation value and can be used to evaluate the osteoporosis-related fracture risk of this special population of postmenopausal women and diabetic patients. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following will briefly introduce the drawings required for the description of the embodiments. Obviously, the following drawings are only some embodiments of the present invention. For those skilled in the art, without creative efforts, other drawings can be obtained based on these drawings.

[0025] Figure 1 It is a schematic flowchart of the method provided in the first aspect of the embodiment of the present invention;

[0026] Figure 2 It is a schematic flowchart of the method provided in the second aspect of the embodiment of the present invention;

[0027] Figure 3 It is a schematic diagram of the computer device provided in the embodiment of the present invention;

[0028] Figure 4 It is a schematic diagram of the architecture of an exemplary computing device provided in the embodiment of the present invention;

[0029] Figure 5 It is a schematic diagram of the storage medium provided in the embodiment of the present invention;

[0030] Figure 6 It is a flowchart of case screening and model establishment provided in the embodiment of the present invention;

[0031] Figure 7 It is the nomogram provided in the embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0032] In order to enable those skilled in the art to better understand the solution of the present invention, the following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the drawings in the embodiments of the present invention.

[0033] In some of the processes described in the specification, claims, and the above-mentioned drawings of the present invention, a plurality of operations appear in a specific order. However, it should be clearly understood that these operations can be executed not in the order in which they appear herein or in parallel. The serial numbers of the operations, such as 101, 102, etc., are only used to distinguish different operations, and the serial numbers themselves do not represent any execution order. In addition, these processes may include more or fewer operations, and these operations can be executed in sequence or in parallel. It should be noted that the descriptions such as "first" and "second" in this article are used to distinguish different messages, devices, modules, etc., do not represent a sequence, and do not limit that "first" and "second" are of different types.

[0034] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative efforts belong to the scope of protection of the present invention.

[0035] Figure 1 It is a schematic flowchart of a method for constructing a fracture risk prediction model provided by an embodiment of the present invention. Specifically, the method includes the following steps:

[0036] S101, obtaining the basic data of T2DM osteoporosis patients as training set samples; the training set samples are divided into a fracture group and a non-fracture group;

[0037] In some embodiments, the inclusion criteria for the training set samples include: women of any age who have been naturally menopausal for more than 1 year; meeting the diagnostic criteria for type 2 diabetes; BMI is between 18.5 and 24.

[0038] In some embodiments, the exclusion criteria for the training set samples include: non-naturally menopausal women including those after hysterectomy; patients suffering from diseases affecting bone metabolism; patients taking drugs affecting bone metabolism; type 1 diabetes, special types of diabetes.

[0039] In some embodiments, the diseases affecting bone metabolism include any one or more of the following: Cushing's syndrome, hyperthyroidism, primary hyperparathyroidism, rheumatoid arthritis, leukemia, multiple myeloma, subtotal gastrectomy, chronic hepatitis, congenital bone metabolism disorders, and other endocrine diseases.

[0040] S102. Statistically analyze the basic data. For continuous variables that conform to the normal distribution, they are expressed as mean ± standard deviation, and the independent sample t-test is used to compare the differences between the two groups. For continuous variables that do not conform to the normal distribution, the median (M) and quartiles (Q1, Q3) are used for description.

[0041] S103. Screen the influencing characteristics of fracture occurrence in osteoporosis patients from the basic data through univariate and multivariate analyses.

[0042] S104. Use Akaike information for backward stepwise selection to select the predictive variables.

[0043] S105. Based on the predictive variables, construct a prediction model for the risk of fracture occurrence in osteoporosis patients.

[0044] In some embodiments, the prediction model is in the form of a nomogram.

[0045] Figure 2 It is a schematic flowchart of a method for predicting the risk of fracture occurrence in osteoporosis patients provided by an embodiment of the present invention. Specifically, the method includes the following steps:

[0046] S201. Obtain predictive factors including age, menopause age, diabetes duration, PINP, CTX, and FF score for the subjects; the subjects are T2DM osteoporosis patients.

[0047] In some embodiments, the term "subject" or "test subject" or "test sample" used herein refers to any animal (e.g., mammal), including but not limited to humans, non-human primates, rodents, etc., which will be the recipient of a specific treatment. Generally, the terms "subject" and "patient" can be used interchangeably herein when referring to human subjects. Preferably, the subject is a human.

[0048] S202. Calculate the risk score based on the predictive factors, and output an auxiliary prediction result of the fracture risk probability of the subject according to the risk score; when the risk score is greater than the first threshold, output an auxiliary prediction result with a high fracture risk probability for the subject; when the risk score is less than the first threshold, output an auxiliary prediction result with a low fracture risk probability for the subject.

[0049] In some embodiments, the method further includes: for patients with age > 60 years, early menopause age < 47 years, diabetes duration > 10 years, CTX > 500 pg / mL and FF > 75%, recommend bone density combined with TBS detection every 6 months for treatment advice, and initiate bisphosphonate combined with SGLT2 inhibitor treatment if necessary.

[0050] In some embodiments, the auxiliary prediction results include, but are not limited to, paper or electronic reports. These results are only obtained by the intelligent machine through analyzing the relevant data of the subjects and are only for reference by medical staff, not as the final diagnosis result of the subjects.

[0051] In some embodiments, the first threshold is obtained by training with a training set of samples. It can be a specific threshold or an interval range, and the specific form is not specifically limited in this embodiment.

[0052] Figure 3 is a schematic diagram of a computer device provided by an embodiment of the present invention. As Figure 3 shown, the device 2000 may include: one or more processors 2010, and one or more memories 2020; wherein, computer-readable code is stored in the memory, and when the computer-readable code is run by the one or more processors, the methods described above can be executed.

[0053] The processor in this embodiment may be an integrated circuit chip with signal processing capabilities. The above-mentioned processor may be a general-purpose processor, a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components. It can implement or execute the various methods, operations and logic block diagrams disclosed in the embodiments of the present disclosure. The general-purpose processor may be a microprocessor or the processor may also be any conventional processor, etc., and may be of the X86 architecture or the ARM architecture.

[0054] Generally speaking, the various exemplary embodiments of the present disclosure can be implemented in hardware or special circuits, software, firmware, logic, or any combination thereof. Some aspects can be implemented in hardware, while other aspects can be implemented in firmware or software that can be executed by a controller, a microprocessor or other computing devices. When the aspects of the embodiments of the present disclosure are illustrated or described as block diagrams, flowcharts or using some other graphical representations, it will be understood that the blocks, devices, systems, technologies or methods described herein can be implemented as non-limiting examples in hardware, software, firmware, special circuits or logic, general hardware or a controller or other computing devices, or some combination thereof.

[0055] For example, the method or device according to the embodiments of the present disclosure can also be implemented by means of Figure 4 the architecture of the computing device 3000 shown. As Figure 4As shown, the computing device 3000 may include a bus 3010, one or more CPUs 3020, a read-only memory (ROM) 3030, a random access memory (RAM) 3040, a communication port 3050 connected to a network, an input / output component 3060, a hard disk 3070, etc. Storage devices in the computing device 3000, such as the ROM 3030 or the hard disk 3070, may store various data or files used for the processing and / or communication of the methods provided by this disclosure, as well as program instructions executed by the CPU. The computing device 3000 may also include a user interface 3080. Of course, Figure 4 the architecture shown is only exemplary, and when implementing different devices, one or more components in the computing device shown may be omitted according to actual needs. Figure 4

[0056] An embodiment of the present invention also provides a computer-readable storage medium, such as Figure 5 As shown, it is a schematic diagram of the storage medium 4000 provided by an embodiment of the present invention. Computer-readable instructions 4010 are stored on the computer storage medium 4020. When the computer-readable instructions 4010 are run by a processor, the methods according to the embodiments of this disclosure described with reference to the above figures can be executed. The computer-readable storage medium in the embodiments of this disclosure may be a volatile memory or a non-volatile memory, or may include both volatile and non-volatile memories. The non-volatile memory may be a read-only memory (ROM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), an electrically erasable programmable read-only memory (EEPROM), or a flash memory. The volatile memory may be a random access memory (RAM), which is used as an external cache. By way of example but not limitation, many forms of RAM are available, such as static random access memory (SRAM), dynamic random access memory (DRAM), synchronous dynamic random access memory (SDRAM), double data rate synchronous dynamic random access memory (DDR SDRAM), enhanced synchronous dynamic random access memory (ESDRAM), synchronous link dynamic random access memory (SLDRAM), and direct memory bus random access memory (DR RAM). It should be noted that the memories for the methods described herein are intended to include but not be limited to these and any other suitable types of memories. It should be noted that the memories for the methods described herein are intended to include but not be limited to these and any other suitable types of memories.

[0057] An embodiment of the present disclosure also provides a computer program product or system, including a computer program, which implements the steps of the above method when executed by a processor.

[0058] In some embodiments, this embodiment also discloses a system for constructing a fracture risk prediction model, the system includes:

[0059] A training set data acquisition module, which is used or configured to obtain the basic data of T2DM osteoporosis patients as training set samples; the training set samples are divided into a fracture group and a non-fracture group;

[0060] A data processing module, which is used or configured to perform statistical analysis on the basic data. For continuous variables that conform to the normal distribution, they are expressed as mean ± standard deviation, and the independent sample t-test is used to compare the differences between the two groups; for continuous variables that do not conform to the normal distribution, the median (M) and quartiles (Q1, Q3) are used to describe;

[0061] A first feature screening module, which is used or configured to screen the influencing features of osteoporosis patients with fractures from the basic data through univariate and multivariate analyses;

[0062] A second feature screening module, which is used or configured to perform backward stepwise selection using the Akaike information to select the predictor variables;

[0063] A model construction module, which is used or configured to construct a prediction model for the risk of fractures in osteoporosis patients based on the predictor variables.

[0064] In some embodiments, the present embodiment also discloses a prediction system for the risk of fractures in osteoporosis patients, and the system includes:

[0065] A predictor acquisition module, which is used or configured to acquire predictors including age, menopause age, diabetes duration, PINP, CTX, and FF score; the subjects are T2DM osteoporosis patients;

[0066] An auxiliary prediction result output module, which is used or configured to calculate a risk score based on the predictors and output an auxiliary prediction result of the risk probability of the subject having a fracture according to the risk score; when the risk score is greater than the first threshold, an auxiliary prediction result with a high risk probability of the subject having a fracture is output; when the risk score is less than the first threshold, an auxiliary prediction result with a low risk probability of the subject having a fracture is output. Specific embodiments

[0067] 2 Subjects and methods

[0068] 2.1 Calculation of samples

[0069] The 10 EPV (10 events per variable) rule is an empirical rule that is currently widely adopted. In the development of models based on binary classification or survival events, to ensure the accuracy of the model sample size, it is usually required that each predictor (i.e., the regression coefficient β value) corresponds to at least 10 outcome events. In this study, the occurrence of fractures in postmenopausal women with diabetic osteoporosis was used as the observation outcome. If we planned to include 10 variables as predictors, then at least 10 * 10 = 100 individuals would need to have the outcome event (fracture occur). In this study, 158 patients were selected as the research subjects, among which 102 had outcome events, within the range of the sample estimation results. The included research subjects were the same as those in the first part. The flow chart of case screening and model establishment is shown in Figure 6 .

[0070] 2.2 Data collection

[0071] 2.2.2 Inclusion and exclusion criteria: A total of 272 postmenopausal T2DM patients who visited the Second Affiliated Hospital of Shandong First Medical University from November 2022 to April 2024 were collected as the research subjects. This study was reviewed and approved by the Ethics Committee of the Second Affiliated Hospital of Shandong First Medical University.

[0072] Inclusion criteria: ① Women of any age who have been naturally menopausal for more than 1 year. ② Meeting the diagnostic criteria for type 2 diabetes (T2DM). According to the Guidelines for the Prevention and Treatment of Type 2 Diabetes in China (2020 Edition), the diagnostic criteria for T2DM are as follows: Patients have typical diabetes-related symptoms (polydipsia, polyuria, polyphagia, weight loss). If patients do not have typical diabetes symptoms, retesting is required. Based on the presence of typical symptoms in patients or on the basis of retesting, any one of the following four indicators can be used for a definite diagnosis. Fasting blood glucose (FPG) ≥ 7.0 mmol / L (126 mg / dL), where fasting means not eating for at least 8 hours. Postprandial blood glucose at 2 hours (OGTT2h) ≥ 11.1 mmol / L (200 mg / dL), which is measured by the oral glucose tolerance test (OGTT). Random blood glucose ≥ 11.1 mmol / L (200 mg / dL), and random blood glucose refers to the blood glucose at any time of the patient without considering the time of the last meal. Glycated hemoglobin (HbA1c) ≥ 6.5% (48 mmol / mol). Note: The blood glucose in the diagnostic criteria is venous plasma glucose, not capillary blood glucose. ③ Body mass index (BMI) range: 18.5 - 24 (the impact of body weight on osteoporosis is still controversial, so the BMI range is restricted). ④ All study subjects underwent magnetic resonance scanning to obtain IDEAL-IQ images and measured the lumbar fat fraction (FF). ⑤ All study subjects had complete clinical data and underwent tests for procollagen type I N-terminal propeptide (PINP) and C-terminal cross-linked telopeptide of type I collagen (CTX). ⑥ The diagnosis of osteoporosis (OP) conforms to the relevant diagnosis and treatment guidelines. Dual energy X-ray absorptiometry (DXA) measurement of bone density in the spine or hip joint is the gold standard for diagnosing OP. Detection site: lumbar spine (L1-4). The results are expressed as T values. An osteoporosis diagnosis is made when the T value is less than -2.5; osteopenia is diagnosed when -1 ≤ T value ≤ -2.5; and normal bone mass is diagnosed when the T value is greater than -1.

[0073] Exclusion criteria: ① Patients who are not naturally postmenopausal, such as those after hysterectomy; ② Patients with diseases affecting bone metabolism, such as Cushing's syndrome, hyperthyroidism, primary hyperparathyroidism, rheumatoid arthritis, leukemia, multiple myeloma, subtotal gastrectomy, chronic hepatitis, congenital bone metabolism disorders and other endocrine diseases; ③ Patients taking drugs affecting bone metabolism, including glucocorticoids, estrogen, calcitonin, bisphosphonates, antiepileptic drugs, thyroxine, diuretics, anticoagulants, etc.; ④ Type 1 diabetes, special types of diabetes; ⑤ Hepatic and renal insufficiency or other severe chronic diseases; ⑥ Cancer patients; ⑦ Patients who do not cooperate with the examination due to personal or other factors.

[0074] 2.2.3 Research process

[0075] 2.2.3.1 Data collection

[0076] (1) Collect the clinical data of the patients: including age, menopause time, diabetes duration, height, weight, body mass index, daily exercise duration, smoking history, drinking history, comorbidities (hypertension, coronary heart disease, cerebral infarction). Improve the test indicators such as biochemistry, blood lipid, glycated hemoglobin, fasting insulin, oral glucose tolerance test, calculate the insulin resistance index (Insulinresistance, IR) (fasting insulin * fasting blood glucose / 22.5), biochemistry and urinary microalbumin / creatinine, urinary renal function panel, 24-hour urinary protein quantification, etc.

[0077] (2) DXA scan: Scanning instrument: Dual-energy X-ray absorptiometer; Manufacturer: HOLLOGIC (USA); Measurement site: Lumbar spine (L1-L4 vertebral bodies); Scanning parameters: Scanning direction: Anterior-posterior view of the spine, Current range: 2.5 - 3.0 mA, Voltage range: 100 - 140 kV, Scanning width: 11.4 cm, Scanning length: 20.4 m; Measurement steps: Calibrate the instrument: The instrument needs to be calibrated according to the manufacturer's standards before each measurement. Perform the scan: Perform a BMD scan of the L1 to L4 vertebral bodies in the anterior-posterior view. According to the diagnostic criteria: 272 patients were divided into three groups: osteopenia group (n = 57), normal bone mass group (n = 57), osteoporosis group (n = 158)

[0078] (3) MRI scan: Instrument: GE 3.0T Discovery MR 750 magnetic resonance imaging system.

[0079] Lumbar spine routine scan: Pre-scan preparation: Ensure the patient has emptied their bladder before the scan to minimize image quality issues. Explain the scan procedure to the patient and ensure they understand and are relaxed. Remove any metal objects, jewelry, or other items that may obstruct imaging. Scan parameter settings: T1-weighted sequence: TR (repetition time): 400-600 ms, TE (echo time): 10-20 ms, matrix: 256x256 or 512x512, FOV (field of view): 20-30 cm, slice thickness: 3-5 mm. T2-weighted sequence: TR: 3000-5000 ms, TE: 80-120 ms, matrix: 256x256 or 512x512, FOV: 20-30 cm, slice thickness: 3-5 mm. Longer TR and TE values are used to enhance fluid signal. Scanning procedures: Patient positioning: Lay the patient flat on the scanning table, ensuring the lumbar spine is centered in the MRI machine. Positioning Scan: Perform a positioning scan to determine the correct scanning position and ensure that the scan covers the lumbar spine area from L1 to L5. Scan Execution: Perform T1- and T2-weighted scans according to the set parameters. Monitor the scanning process to ensure good image quality. Post-processing: After the scan is completed, perform image reconstruction and post-processing, including possible image comparison and analysis.

[0080] Lumbar spine IDEAL-IQ sequence scan: After the routine scan, perform the lumbar spine IDEAL-IQ sequence scan. Select the positioning plane: Use the midline plane of lumbar vertebra L3 as the scan center, ensuring that the scan includes L3 and its upper and lower vertebrae (L2 and L4). Scan parameter settings: Slice thickness 3 mm, TE (echo time): Select an appropriate TE value. The TE you mentioned may be between 1.2 ms and 7.38 ms. Typically, a shorter TE is selected to increase the strength of the water signal. TR (repetition time): 9 ms, inversion angle: 15°, number of excitations: 1, scan time: approximately 15 seconds.

[0081] Lumbar spine sagittal and transverse images obtained from the IDEAL-IQ sequence are processed, and regions of interest (ROIs) are delineated. Vertebral ROI delineation method: In the midsagittal plane of the vertebral fat fraction map (i.e., the plane with the largest vertebral display), the ROI for L1 to L5 vertebrae should be delineated within the cancellous bone region at the center of the vertebral body, avoiding surrounding cortical bone and areas of bone hyperplasia. The size and position of the ROI should be adjusted based on the size and shape of the vertebral body. The signal intensity of fat and water within the ROI is calculated automatically or manually using the software to determine the FF value. The FF value for the paraspinal muscles is the average of the bilateral FF values from two measurements.

[0082] Quality control: Regularly review the reconstructed images to ensure no obvious artifacts and non-uniformity. Draw regions of interest (ROIs) at the same location and size to ensure data consistency. Regularly review the selection criteria of ROIs to ensure consistency across different scans. The FF value was measured three times by three senior physicians, and the average value was taken.

[0083] (3)PINP and CTX detection: Instruments: Reagents: PINP detection kit, CTX detection kit;

[0084] Detect serum-related indicators using ELISA kits. After centrifuging the blood samples, check PINP and CTX according to the kit instructions. Equilibrate the reagents, dilute with concentrated washing solution, and set aside. Add samples, seal the plate and incubate, wash, and pat dry. Add enzyme, incubate and wash. Add chromogenic reagent for color development, add stop solution to terminate the reaction, and read the OD value at a wavelength of 450 nm.

[0085] 2.2.3.2 Research methods:

[0086] 272 patients were divided into three groups according to the diagnostic criteria for osteoporosis: osteopenia group (n = 57), normal bone mass group (n = 57), and osteoporosis group (n = 158). Compare the clinical data and imaging data of the three groups of patients. Analyze the correlation between the different clinical data and imaging data and FF. The Receiver Operating Characteristic (ROC) curve was used to judge the diagnostic value of FF for type 2 diabetic osteoporosis in postmenopausal women.

[0087] 2.2.4 Statistical methods:

[0088] SPSS 28.0 statistical software was used. Measurement data were expressed as mean ± standard deviation (±S), and count data were expressed as %. One-way analysis of variance was used for measurement data. When the variances were homogeneous, pairwise comparisons were performed using the least significant difference method. When the variances were heterogeneous, pairwise comparisons were performed using the Bonfemroni method. Count data were analyzed using the χ2 test and the exact probability calculation method. Pearson correlation analysis was used to evaluate the correlation between various indicators, and partial correlation analysis was used after adjusting for confounding factors. The ROC curve was used to judge the risk assessment value and threshold definition of FF for type 2 diabetic osteoporosis in postmenopausal women. P < 0.05 was considered statistically significant.

[0089] 2.3 Model establishment

[0090] According to whether the patients had fractures, they were divided into two groups: the fracture group and the non-fracture group. The indicators statistically analyzed in the first part were used as variables for statistical analysis. For continuous variables that conform to the normal distribution, we used the mean ± standard deviation (x±s) to represent them, and compared the differences between the two groups through an independent samples t-test; for continuous variables that do not conform to the normal distribution, the median (M) and quartiles (Q1, Q3) were used for description, and the Mann-Whitney U test was used for comparison.

[0091] In addition, we used univariate and multivariate Logistic regression models to analyze the influencing factors of fractures in postmenopausal women with diabetic osteoporosis. To optimize the fitting effect of the multivariate Logistic regression model, we used the Akaike Information Criterion (AIC) for backward stepwise selection and included the selected variables in the nomogram to predict the probability of fractures in postmenopausal women with diabetic osteoporosis.

[0092] 2.4 Model performance verification

[0093] We visualized and quantified the discrimination ability of the model by plotting the Receiver Operating Characteristic Curve (ROC) and calculating the Area under the Curve (AUC). The verification method used the Bootstrap technique with 2000 resamplings to plot the calibration curve, evaluate the consistency of the model, and perform Decision Curve Analysis (DCA) to evaluate its clinical applicability. The calibration curve shows the relationship between the observed outcome frequency and the predicted probability in the study group. In a well-calibrated model, the predicted results should fall on the 45-degree diagonal line. We also calculated the mean absolute error, and the smaller the value, the better the model prediction effect. The statistical significance level was P<0.05.

[0094] 3 Results

[0095] 3.1 Comparison of the general conditions of the study subjects

[0096] A total of 158 postmenopausal women with diabetic osteoporosis were included in this study. The cases were randomly split according to a 7:3 ratio: 110 patients in the training set and 48 patients in the validation set. Among them, 125 patients in the training set were used for the analysis of the influencing factors of fractures in postmenopausal women with diabetic osteoporosis to establish a Nomogram model; 42 patients in the validation set were used to verify the model. When comparing the general conditions and clinical data of the patients in the training set and the validation set, no significant statistical differences were found (P>0.05).

[0097] 3.2 Univariate analysis of influencing factors for fractures in postmenopausal women with diabetic osteoporosis

[0098] A total of 158 people were included in this study, among whom 56 had no fractures, accounting for 35.44%, and 102 had fractures, accounting for 64.56%. There were significant differences in age, BMI, T, PINP, and CTX between the two groups. Those in the fracture group (65.22 ± 8.65, 24.61 ± 3.03, -2.65 ± 0.93, 64.58 ± 13.57, 703.32 ± 144.74, 76.82 ± 10.17) were significantly higher than those in the non-fracture group, with statistical significance (t=-2.078, -5.738, -4.322, -4.919, -4.984, P<0.005). FF in the fracture group was significantly lower than that in the non-fracture group, with statistical significance (t=4.490, P<0.005).

[0099] Table 1 Results of univariate logistic regression

[0100]

[0101] 3.3 Results of multivariate logistic regression for fractures in postmenopausal women with diabetic osteoporosis

[0102] Further multivariate analysis revealed that T, PINP, FF, and CTX were independent risk factors for fractures in patients. The odds ratio (OR) and 95% confidence interval were as follows: T: 17.86 (2.55 - 124.78), PINP: 1.15 (1.03 - 1.29), FF: 0.62 (0.44 - 0.88), CTX: 1.01 (1.01 - 1.02). See Table 2.

[0103] Table 2 Results of multivariate logistic regression

[0104]

[0105] 3.4 Multicollinearity analysis

[0106] For the influencing factors such as T, PINP, FF, and CTX shown by multivariate analysis, the VIF was <10 and close to 1, indicating the independence among variables. See Table 3.

[0107] Table 3 Collinearity analysis

[0108]

[0109] A nomogram was established using age, menopause age, diabetes duration, PINP, CTX, and FF score factors (as shown in Figure 7 and Table 4). Through the nomogram model, we can visually observe the scores associated with each independent risk factor. In the nomogram, the scale values of each predictive indicator correspond to the scores on the scoring scale. The sum of all predictive variables gives a total score, which reflects the probability of a poor prognosis for the patient. For example, for a postmenopausal female diabetic osteoporosis patient, the examination results show: age 60 years (38 points), menopause age 46 years (55 points), diabetes duration 10 years (32 points), PINP 80 (35 points), CTX 650 (35 points), FF 85 (25 points). The total score of this patient is 38 + 55 + 32 + 35 + 35 + 25 = 220, and the risk of fracture is 0.8.

[0110] Table 4

[0111]

[0112] 3.6 Evaluation of the discrimination ability of the prediction model

[0113] After modeling, the area under the curve (AUC) and C-index of the training set were 0.972 (95% CI: 0.947 - 0.997), while the AUC and C-index of the validation set were 0.906 (95% CI: 0.923 - 1.000), indicating that the model has good discrimination ability, as shown in the figure. The specific results of the ROC analysis are shown in (Table 5), with an AUC of 0.972 (95% CI: 0.947 - 0.997), a sensitivity of 0.900, a specificity of 0.907, and an optimal cut-off value of 0.308. After verification, the prediction effect of the model is good.

[0114] Table 5

[0115]

[0116] 3.7 Calibration curve of the prediction model

[0117] The P value of the Hosmer-Lemeshow test of the model was 1.000, indicating a good match between the predicted probability and the observed probability in the two calibration curves, suggesting that the model has excellent calibration ability. In addition, the DCA curve also shows that the model curve is significantly higher than the all-intervention line and the all-no-intervention line, indicating the clinical utility of the model. The model shows significant clinical net benefits in both the modeling group and the validation group.

[0118] It should be noted that the flowcharts and block diagrams in the accompanying drawings illustrate the possible architectures, functions, and operations of systems, methods, and computer program products according to various embodiments of the present disclosure. In this regard, each block in the flowchart or block diagram may represent a module, a program segment, or a part of code that contains one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions marked in the blocks may occur in a different order than that marked in the accompanying drawings. For example, two consecutive blocks shown may actually be executed substantially in parallel, and they may sometimes be executed in the reverse order, depending on the functions involved. It should also be noted that each block in the block diagram and / or flowchart, as well as the combination of blocks in the block diagram and / or flowchart, may be implemented by a dedicated hardware-based system for performing the specified functions or operations, or may be implemented by a combination of dedicated hardware and computer instructions.

[0119] In general, the various example embodiments of the present disclosure may be implemented in hardware or dedicated circuits, software, firmware, logic, or any combination thereof. Some aspects may be implemented in hardware, while other aspects may be implemented in firmware or software that can be executed by a controller, a microprocessor, or other computing devices. When aspects of the embodiments of the present disclosure are illustrated or described as block diagrams, flowcharts, or using some other graphical representation, it will be understood that the blocks, devices, systems, techniques, or methods described herein may be implemented as non-limiting examples in hardware, software, firmware, dedicated circuits or logic, general hardware or controllers or other computing devices, or some combination thereof.

[0120] Those skilled in the art can clearly understand that for the convenience and brevity of description, the specific working processes of the systems, devices, and units described above can refer to the corresponding processes in the foregoing method embodiments and will not be elaborated herein.

[0121] In the several embodiments provided in this application, it should be understood that the disclosed systems, devices, and methods may be implemented in other ways. For example, the device embodiments described above are merely illustrative. For example, the division of the units is only a logical function division, and there may be other division methods in actual implementation. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Another point is that the displayed or discussed couplings or direct couplings or communication connections to each other may be indirect couplings or communication connections through some interfaces, devices, or units, and may be in electrical, mechanical, or other forms.

[0122] The unit described as a separation component may or may not be physically separated. The component shown as a unit may or may not be a physical unit, that is, it may be located in one place or may be distributed over multiple network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0123] In addition, in each embodiment of the present invention, each functional unit may be integrated in a processing unit, may exist separately as individual physical units, or two or more units may be integrated in one unit. The above integrated units may be implemented in the form of hardware or in the form of software functional units.

[0124] The exemplary embodiments of the present disclosure described in detail above are merely illustrative and not restrictive. Those skilled in the art should understand that various modifications and combinations can be made to these embodiments or their features without departing from the principles and spirit of the present disclosure, and such modifications should fall within the scope of the present disclosure.

Claims

1. A method for constructing a fracture risk prediction model, characterized in that, The method includes: S101. Obtain the basic data of T2DM osteoporosis patients as training set samples; the training set samples are divided into a fracture group and a non-fracture group; S102. Conduct statistical analysis on the basic data. For continuous variables that conform to the normal distribution, they are expressed as mean ± standard deviation, and the independent sample t-test is used to compare the differences between the two groups; for continuous variables that do not conform to the normal distribution, the median and quartiles are used for description; S103. Screen the influencing characteristics of fractures in osteoporosis patients from the basic data through univariate and multivariate analyses; S104. Use the Akaike information for backward stepwise selection to select the predictive variables; S105. Based on the predictive variables, construct a prediction model for the risk of fractures in osteoporosis patients.

2. The method for constructing a fracture risk prediction model according to claim 1, wherein The prediction model is in the form of a nomogram.

3. The method for constructing a fracture risk prediction model according to claim 1, wherein, The inclusion criteria for the training set samples include: women of any age who have been naturally menopausal for more than 1 year; meeting the diagnostic criteria for type 2 diabetes; with a BMI between 18.5 and 24.

4. The method for constructing a fracture risk prediction model according to claim 1, wherein The exclusion criteria for the training set samples include: non-naturally menopausal women including those after hysterectomy; patients with diseases affecting bone metabolism; patients taking drugs affecting bone metabolism; type 1 diabetes, special type diabetes.

5. The method for constructing a fracture risk prediction model according to claim 1, wherein Diseases affecting bone metabolism include any one or more of the following: Cushing's syndrome, hyperthyroidism, primary hyperparathyroidism, rheumatoid arthritis, leukemia, multiple myeloma, subtotal gastrectomy, chronic hepatitis, congenital bone metabolism disorders, and other endocrine diseases.

6. A method for predicting the risk of fracture in osteoporosis patients, characterized in that, The method includes: S201. Obtain the predictive factors of the subjects including age, menopause age, diabetes duration, PINP, CTX, FF score; the subjects are T2DM osteoporosis patients; S202. Calculate the risk score based on the predictive factors, and output the auxiliary prediction result of the probability of fracture occurrence of the subjects according to the risk score; when the risk score is greater than the first threshold, output the auxiliary prediction result with a high probability of fracture occurrence of the subjects; when the risk score is less than the first threshold, output the auxiliary prediction result with a low probability of fracture occurrence of the subjects.

7. The method for constructing a fracture risk prediction model according to claim 6, wherein, The method further includes: for patients with an age greater than 60 years, a menopause age < 47 years, a diabetes duration > 10 years, CTX > 500 pg / mL and FF > 75%, recommend a treatment suggestion of bone density combined with TBS detection every 6 months.

8. A computer device, characterized in that, The device includes: a memory and a processor; the memory is used to store a computer program; the processor executes the computer program to implement the steps of the method according to any one of claims 1-7.

9. A computer-readable storage medium, characterized in that, A computer program is stored thereon, and when the computer program is executed by the processor, it implements the steps of the method according to any one of claims 1-7.

10. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by the processor, it implements the steps of the method according to any one of claims 1-7.

Images