System and method for predicting tumor bone metastasis based on bone metastasis-specific bone markers

Abstract

The application relates to a tumor bone metastasis prediction system and method based on bone metastasis specific bone markers. The system comprises a calculation module, which calculates the sensitive weight of each common marker for the same cancer according to the metabolic speed of each common marker of a plurality of patients and the number of the plurality of patients; a compensation module, which determines the compensation coefficient of each specific marker for the same cancer based on the difference between the detection value of each specific marker of the same patient and historical detection data; a determination module, which determines the target coupling coefficient of each specific marker according to the correlation between each specific marker and each common marker of a plurality of patients; and an analysis module, which performs risk analysis according to the sensitive weight, the compensation coefficient, the target coupling coefficient and the target growth speed of each common marker to obtain an analysis result. The application can quantify the synchronism of the changes of the common markers and the specific markers, improve the early bone metastasis detection capability and reduce the misjudgment rate.

Description

Technical Field

[0001] This application relates to the field of medical data processing technology, specifically to a tumor bone metastasis prediction system and method based on bone metastasis-specific bone markers. Background Technology

[0002] Bone metastasis refers to the spread of cancer cells from the primary tumor to the bone via the bloodstream or lymphatic system, where they continue to grow and form new lesions. It is a common manifestation of advanced cancer, potentially leading to complications such as bone pain, fractures, and hypercalcemia, impacting patients' quality of life and survival. Common biomarkers of bone metabolism are easily affected by non-cancerous orthopedic diseases, leading to false-positive increases. Traditional models relying on fixed thresholds cannot distinguish between the gradual metabolic acceleration specific to early bone metastasis and fluctuations in underlying diseases. Specific biomarkers of the primary cancer often show delayed increases or false negatives in early bone metastasis. Simply discarding negative biomarkers results in the loss of risk signals, while retaining fixed weights introduces noise and misjudgments. Existing technologies have rigid processing of the synergistic mechanisms of these two types of biomarkers; when specific biomarkers fail, the associated common biomarker proxy function cannot be activated, easily leading to missed diagnoses of high-risk patients. Summary of the Invention

[0003] To address the aforementioned technical problems, the purpose of this application is to provide a tumor bone metastasis prediction system and method based on bone metastasis-specific bone markers. The specific technical solution adopted is as follows:

[0004] Firstly, a tumor bone metastasis prediction system based on bone metastasis-specific bone markers is provided, the system comprising:

[0005] The calculation module is used to calculate the sensitivity weight of each common biomarker for the same cancer based on the metabolic rate of each common biomarker in multiple patients and the number of multiple patients; the metabolic rate is determined based on the difference of the same common biomarker in the same patient between adjacent time points;

[0006] The compensation module is used to determine the compensation coefficient for each specific biomarker for the same cancer based on the difference between the detection value of each specific biomarker and historical detection data for the same patient; the compensation coefficient is used to adjust the weight of the specific biomarker.

[0007] The determination module is used to determine the target coupling coefficient of each specific biomarker based on the correlation between each specific biomarker and each common biomarker in multiple patients;

[0008] The analysis module is used to perform risk analysis based on the sensitivity weight, the compensation coefficient, the target coupling coefficient, and the target growth rate of each common biomarker to obtain the analysis results; the target growth rate is determined based on the difference between the same common biomarker of the same patient at the first time point and the second time point.

[0009] Optionally, the calculation module is also used for:

[0010] The growth rate of the common biomarker in the same patient at adjacent time points is calculated based on the difference in detection values ​​of the same common biomarker at adjacent time points; the difference in detection values ​​is obtained by the difference between the detection values ​​of the same common biomarker at the third time point and the detection values ​​of the same common biomarker at the fourth time point, where the fourth time point is the time point preceding the third time point.

[0011] The metabolic rate of the same common biomarker in the same patient can be determined by the rate of increase of the same common biomarker at all adjacent time points.

[0012] For each common biomarker, the sensitivity coefficient of each common biomarker for the same cancer is determined based on the ratio of the number of sensitive patients to the total number of patients. The metabolic mean is then weighted according to the sensitivity coefficient to obtain the sensitivity weight of each common biomarker for the same cancer. The number of sensitive patients is the number of patients among the multiple patients whose metabolic rate for the corresponding common biomarker is positive. The metabolic mean indicates the average metabolic rate of the corresponding common biomarker among the multiple patients.

[0013] Optionally, the compensation module is also used for:

[0014] Based on the detection values ​​of the same specific biomarker of the same patient within a preset time window and the detection values ​​of the same specific biomarker of the same patient within all time windows in historical detection data, the latent response coefficient of the same specific biomarker of the same patient for the same cancer within the preset time window is determined, so as to obtain the latent response coefficient of the same specific biomarker of the same patient for the same cancer within all time windows; the all time windows include the preset time window.

[0015] Based on the latent response coefficients of the same specific biomarker for the same cancer in multiple patients across all time windows, the compensation coefficient of the same specific biomarker for the same cancer is determined to obtain the compensation coefficient of each specific biomarker for the same cancer.

[0016] Optionally, the compensation module is also used for:

[0017] The abnormal pulse amplitude of the same specific biomarker for the same cancer within a preset time window is determined by the ratio of the range of detection values ​​of the same specific biomarker in the same patient within a preset time window to the mean of detection values ​​of the same specific biomarker in the same patient across all time windows in historical detection data.

[0018] The abnormal pulse frequency of the same specific biomarker for the same cancer within the preset time window is determined based on the ratio of the number of abnormalities to the duration of the preset time window; the number of abnormalities indicates the number of times the detection value of the same specific biomarker of the same patient exceeds the preset detection value within the preset time window; the preset detection value is obtained by calculating the standard deviation and / or mean value of the detection values ​​of the same specific biomarker of the same patient in all time windows in historical detection data.

[0019] Based on the product of the abnormal pulse amplitude and the abnormal pulse frequency, the latent response coefficient of the same specific biomarker for the same cancer in the same patient within a preset time window is determined.

[0020] Based on the latent response coefficient of the same specific biomarker for the same cancer in the same patient within a preset time window, the compensation coefficient of the same specific biomarker for the same cancer within the preset time window is determined, so as to obtain the compensation coefficient of each specific biomarker for the same cancer in all time windows for the same patient.

[0021] Optionally, the compensation module is also used for:

[0022] The latency response threshold is determined based on the maximum latency response coefficient for each patient; the maximum latency response coefficient is obtained by screening the maximum value from the latency response coefficients of the same specific biomarker for the same cancer in all time windows for each patient.

[0023] The compensation coefficient for the same specific biomarker for the same cancer is determined based on the ratio of the maximum latent response coefficient to the latent response threshold for each patient, thus obtaining the compensation coefficient for each specific biomarker for the same cancer.

[0024] Optionally, the compensation module is also used for:

[0025] The maximum latent response coefficient of each patient in multiple patients is sorted according to a preset rule to obtain the latent response sequence;

[0026] The maximum latent response coefficient at a preset position in the latent response sequence is determined as the latent response threshold.

[0027] Optionally, the determining module is also used for:

[0028] Correlation analysis was performed on each specific biomarker and each common biomarker for each patient to obtain multiple correlation scores for each patient. The multiple correlation scores of the same patient corresponded one-to-one with multiple biomarker combinations. Each biomarker combination included one specific biomarker and one common biomarker. Any two different biomarker combinations included different specific biomarkers and / or different common biomarkers.

[0029] In multiple biomarker combinations, the correlations corresponding to each biomarker combination are analyzed, the coupling coefficient corresponding to each biomarker combination is determined, and the multiple correlations corresponding to each biomarker combination correspond one-to-one with multiple patients;

[0030] From the multiple coupling coefficients corresponding to each specific biomarker, the largest coupling coefficient is selected to determine the target coupling coefficient for each specific biomarker, so as to obtain multiple target coupling coefficients corresponding to multiple specific biomarkers.

[0031] Optionally, the analysis module is also used for:

[0032] The difference between the detection value of the same specific biomarker of the same patient at the third time point and the mean detection value of the same specific biomarker of the same patient in all time windows in the historical detection data is calculated. Based on the ratio of the difference to the mean detection value of the same specific biomarker of the same patient in all time windows in the historical detection data, the variation coefficient of the same specific biomarker is determined to obtain the variation coefficient of each specific biomarker.

[0033] Risk analysis is performed based on the change coefficient, the sensitivity weight, the compensation coefficient, the target coupling coefficient, and the target growth rate of each common marker to obtain the analysis results.

[0034] Optionally, the analysis module is also used for:

[0035] The compensation integral of each specific biomarker is determined based on the coefficient of change of each specific biomarker, the target growth rate of each common biomarker, the target coupling coefficient of each specific biomarker, and the compensation coefficient of each specific biomarker for the same cancer.

[0036] The compensation integral of the common biomarker is determined based on the sensitivity weight of each common biomarker for the same cancer and the target growth rate of each common biomarker.

[0037] Risk analysis is performed based on the sum of the compensation integrals of the specific markers and the compensation integrals of the common markers to obtain the analysis results.

[0038] Secondly, a method for predicting tumor bone metastasis based on bone metastasis-specific bone markers is provided, the method comprising:

[0039] The sensitivity weight of each common biomarker for the same cancer is calculated based on the metabolic rate of each common biomarker in multiple patients and the number of patients; the metabolic rate is determined based on the difference of the same common biomarker in the same patient between adjacent time points;

[0040] Based on the differences between the detection values ​​of each specific biomarker for the same patient and historical detection data, a compensation coefficient for each specific biomarker for the same cancer is determined; the compensation coefficient is used to adjust the weight of the specific biomarker.

[0041] The target coupling coefficient for each specific biomarker is determined based on the correlation between each specific biomarker and each common biomarker in multiple patients.

[0042] Risk analysis is performed based on the sensitivity weight, the compensation coefficient, the target coupling coefficient, and the target growth rate of each common biomarker to obtain the analysis results; the target growth rate is determined based on the difference between the same common biomarker of the same patient at the first time point and the second time point.

[0043] Based on common knowledge in the field, the above-mentioned preferred conditions can be combined arbitrarily to obtain various preferred embodiments of the present invention.

[0044] This application offers the following advantages: It introduces a target coupling coefficient to dynamically bind the strongest pathologically correlated biomarker pairs, avoiding interference from multiple biomarker noise through a one-to-one proxy mechanism, and accurately transferring to highly correlated common biomarkers when specific biomarkers fail. A dual-channel integral model is constructed, combining sensitive weights and compensation coefficients to correct for individual differences, outputting the normalized final risk probability. The synchronicity of changes in common and specific biomarkers is quantified; when a specific biomarker fails, the most correlated common biomarker is selected to proxy its function. The changing trends of common and specific biomarkers are dynamically bound through the target coupling coefficient. When the target coupling coefficient is large, the common biomarker can proxy the function of the specific biomarker, significantly improving the early detection capability of bone metastases and greatly reducing the misjudgment rate. Attached Figure Description

[0045] To more clearly illustrate the technical solutions and advantages in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0046] Figure 1 A schematic diagram of a tumor bone metastasis prediction system based on bone metastasis-specific bone markers provided in one embodiment of this application;

[0047] Figure 2 A flowchart illustrating a tumor bone metastasis prediction method based on bone metastasis-specific bone markers, provided as an embodiment of this application;

[0048] Figure 3This is a schematic diagram of the structure of an electronic device provided in one embodiment of this application. Detailed Implementation

[0049] To further illustrate the technical means and effects adopted by this application to achieve the intended purpose of the invention, the following, in conjunction with the accompanying drawings and preferred embodiments, details the specific implementation, structure, features, and effects of a tumor bone metastasis prediction system and method based on bone metastasis-specific bone markers proposed in this application. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. Furthermore, specific features, structures, or characteristics in one or more embodiments can be combined in any suitable form.

[0050] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application pertains.

[0051] The following, with reference to the accompanying drawings, details the specific scheme of a tumor bone metastasis prediction system based on bone metastasis-specific bone markers provided in this application, such as... Figure 1 As shown, the system includes:

[0052] The calculation module 11 is used to calculate the sensitivity weight of each common biomarker for the same cancer based on the metabolic rate of each common biomarker in multiple patients and the number of multiple patients.

[0053] The metabolic rate was determined based on the difference in the same common biomarker between adjacent time points for the same patient.

[0054] Common biomarkers (such as common biomarkers of bone metabolism) are biological indicators that can reflect the pathophysiological processes (such as inflammation, tissue damage, and metabolic disorders) common to many diseases. They do not have disease specificity. For example, NTX (type I collagen cross-linked amino-terminal peptide) exists in many tissues such as bone, cartilage, and skin. It can reflect osteoclast activity, and elevated serum or urine levels indicate accelerated bone destruction.

[0055] Primary cancer-specific biomarkers are biological indicators highly correlated with a single disease or specific pathological state. For example, neuron-specific enolase reflects markers of bone metastasis in lung cancer, while carbohydrate antigen 15-3 and carcinoembryonic antigen reflect markers of bone metastasis in breast cancer. Therefore, when predicting the probability of tumor bone metastasis, the levels of primary cancer-specific biomarkers can be combined with the levels of common biomarkers of bone metabolism for synergistic analysis, and finally, imaging techniques (computed tomography or magnetic resonance imaging), biopsy, etc., can be used to determine the location of tumor bone metastases throughout the skeleton.

[0056] The patients mentioned in this application can be those with bone metastases from tumors, typically those diagnosed with a certain type of cancer, or those with a history of cancer. Since patients with a history of cancer still have a higher risk of cancer recurrence, this application embodiment measures the levels of common biomarkers and specific biomarkers corresponding to all types of tumors at preset time intervals, and records the time interval between adjacent testing processes to form periodic biomarker testing, thereby obtaining the patient's historical testing data. The methods for measuring the levels of common and specific biomarkers can include venous blood sampling, body fluid collection, urine collection, etc., and the preset time interval can be set according to the actual situation.

[0057] Common biomarkers can be used to differentiate between different types of diseases. Bone collagen breakdown products are one of the commonly used bone resorption biomarkers and are widely used to monitor patients with breast cancer, prostate cancer, lung cancer, and other diseases prone to osteolytic metastases. However, common biomarkers reflect the common pathological process of increased osteoclast activity, and are not specific to tumors. Non-tumor diseases such as rheumatoid arthritis and hyperthyroidism may also lead to elevated bone collagen breakdown products due to inflammatory factors stimulating osteoclasts. Therefore, analysis must be combined with the patient's underlying diseases. If the patient also has chronic inflammation or metabolic bone disease, the false positive probability of elevated common biomarkers will increase significantly. A comprehensive judgment based on imaging, medical history, and dynamic trends is necessary to avoid misdiagnosis as tumor bone metastasis.

[0058] Compared to other basic orthopedic diseases, some common biomarkers caused by bone metastasis show a gradual increasing trend, while some common biomarkers caused by other basic orthopedic diseases may fluctuate slightly within a certain range. Early bone metastasis manifests as changes in metabolic rate rather than an absolute value exceeding the standard. Therefore, when the content of common biomarkers increases at an accelerated rate over time, the sensitivity weight for the probability of bone metastasis in the current type of primary cancer is higher.

[0059] Preliminary clustering was performed on the common and specific biomarkers of each patient in the historical testing data. The historical testing data of patients with the same primary cancer were grouped into one class. In other words, the historical testing data of the same cancer were grouped into one class to obtain multiple classes of different cancers. Each class included the testing data of multiple common biomarkers and multiple specific biomarkers of multiple patients with the same cancer.

[0060] For the historical detection data of the i-th class, i.e., the i-th cancer, we can obtain detection data for multiple common biomarkers and multiple specific biomarkers for the j-th patient. For the x-th common biomarker of the j-th patient, based on the difference in detection values ​​of the common biomarker during the y-th detection process and its adjacent y-1-th detection process, we obtain the metabolic rate of the x-th common biomarker for the j-th patient, and thus the sensitivity weight of the x-th common biomarker for the i-th cancer. By sequentially calculating the sensitivity weights of multiple common biomarkers for multiple cancers, we can obtain the sensitivity weight of each common biomarker for each cancer. Understandably, the y-th and y-1-th detection processes are performed at different times, and the detection times of the y-th and y-1-th detection processes are adjacent.

[0061] In one embodiment, the computing module is further configured to:

[0062] The growth rate of the common biomarker in the same patient at adjacent time points is calculated based on the difference in detection values ​​of the same common biomarker at adjacent time points; the difference in detection values ​​is obtained by the difference between the detection values ​​of the same common biomarker at the third time point and the detection values ​​of the same common biomarker at the fourth time point, where the fourth time point is the time point preceding the third time point.

[0063] The metabolic rate of the same common biomarker in the same patient can be determined by the rate of increase of the same common biomarker at all adjacent time points.

[0064] For each common biomarker, the sensitivity coefficient of each common biomarker for the same cancer is determined based on the ratio of the number of sensitive patients to the total number of patients. The metabolic mean is then weighted according to the sensitivity coefficient to obtain the sensitivity weight of each common biomarker for the same cancer. The number of sensitive patients is the number of patients among the multiple patients whose metabolic rate for the corresponding common biomarker is positive. The metabolic mean indicates the average metabolic rate of the corresponding common biomarker among the multiple patients.

[0065] In the historical detection data of the i-th cancer, multiple common biomarkers and multiple specific biomarkers of the j-th patient can be obtained. For the x-th common biomarker of the j-th patient, the detection values ​​of the common biomarker are sequentially obtained in the y-th detection process and its adjacent y-1-th detection process. The y-th detection process is performed at time 3, and the y-1-th detection process is performed at time 4. The difference between the detection values ​​of the common biomarker at time 3 and time 4 is subtracted from the detection value at time 4 to obtain the difference in detection values ​​of the x-th common biomarker between adjacent time points. The time interval between time 3 and time 4 is calculated. Based on the ratio of the difference in detection values ​​to the time interval, the growth rate of the x-th common biomarker at adjacent time points is calculated. The growth rate of the j-th patient in the historical detection data is calculated at all adjacent time points, and the difference between each adjacent growth rate is calculated. The mean of all differences in growth rates is calculated to obtain the metabolic rate of the x-th common biomarker for the j-th patient with the i-th cancer. This allows us to calculate the metabolic rate of each common biomarker for the j-th patient with the i-th cancer. Metabolic rate A positive number and a larger value indicate a slower metabolic rate for the x-th common biomarker, a higher risk of bone metastasis, and thus a greater sensitivity weight for the x-th common biomarker.

[0066] For primary cancers, if the metabolic rate of the same common biomarker shows a significant slowing trend across different patients, it indicates that the current type of common biomarker has a high reference weight in assessing the risk of bone metastasis. However, evaluating the reference weight of a current type of common biomarker solely based on the numerical value of its metabolic rate can easily lead to a decrease in the actual sensitivity weight, affecting the final accuracy. Therefore, the reference weight can be determined by considering the number of patients with positive metabolic rates for the x-th common biomarker among multiple patients with the i-th cancer, and the total number of patients with the i-th cancer. The ratio, i.e., the number of sensitive patients. The number of multiple patients with the i-th cancer The ratio of the values ​​is used to determine the sensitivity coefficient of the x-th common biomarker for the i-th cancer. Sensitivity coefficient The calculation formula is: ;in, Let x be the sensitivity coefficient of the x-th common biomarker for the i-th cancer. Let i be the number of multiple patients suffering from the i-th cancer. This represents the number of patients among multiple patients with the i-th cancer whose metabolic rate for the x-th common biomarker is positive. Sensitivity coefficient. The larger the value, the higher the sensitivity of the xth common biomarker to the i-th cancer.

[0067] Obtain the mean metabolic rate of the x-th common biomarker from multiple patients with the i-th cancer, and then calculate the metabolic mean based on the sensitivity coefficient. The metabolic mean is weighted to obtain the product of the sensitivity coefficient and the metabolic mean. The product of the sensitivity coefficient and the metabolic mean is then normalized using the norm function to obtain the normalized sensitivity weight of the x-th common biomarker for the i-th cancer. Sensitive weights The reference weight of the x-th common biomarker is used to characterize the risk of bone metastasis for the i-th cancer. Understandably, different common biomarkers have different sensitivity weights for different cancers.

[0068] The compensation module 12 is used to determine the compensation coefficient for each specific biomarker for the same cancer based on the difference between the detection value of each specific biomarker for the same patient and historical detection data.

[0069] The compensation coefficient is used to adjust the weight of specific biomarkers.

[0070] In this embodiment, the levels of common biomarkers and specific biomarkers corresponding to all types of tumors in patients are detected at preset intervals, and the time intervals between adjacent detection processes are recorded to form periodic biomarker detection, thereby obtaining the patient's historical detection data. The preset intervals can be set according to actual conditions, such as 15 days or 30 days. Determining the difference between the detection value of each specific biomarker for the same patient and the historical detection data can be done by comparing the detection value of each specific biomarker for the same patient with the corresponding specific biomarker detection data for the same patient in the historical detection data.

[0071] Different types of primary cancers require different specific biomarkers. Once a patient's history of a specific primary cancer is confirmed, targeted screening based on the corresponding specific biomarkers can be performed. However, some bone metastasis-specific biomarkers may not be elevated in early bone metastases (delayed elevation), such as the lack of elevation in prostate-specific antigen (PSA) in some prostate cancer bone metastases, leading to false negatives. This can easily cause misdiagnosis when using common and specific biomarkers for synergistic judgment. Analysis shows that changes in specific biomarkers in early bone metastases often exhibit transient, pulse-like fluctuations. Analyzing at fixed intervals can capture these transient, pulse-like fluctuations.

[0072] Therefore, in one embodiment, the compensation module is further configured to:

[0073] Based on the detection values ​​of the same specific biomarker of the same patient within a preset time window and the detection values ​​of the same specific biomarker of the same patient within all time windows in historical detection data, the latent response coefficient of the same specific biomarker of the same patient for the same cancer within the preset time window is determined, so as to obtain the latent response coefficient of the same specific biomarker of the same patient for the same cancer within all time windows; all time windows include the preset time window.

[0074] Based on the latent response coefficients of the same specific biomarker to the same cancer in multiple patients across all time windows, the compensation coefficient of the same specific biomarker for the same cancer is determined, so as to obtain the compensation coefficient of each specific biomarker for the same cancer.

[0075] In historical testing data, the time series of specific biomarkers for each patient with the same cancer is divided into time windows of fixed duration, such as 90-day time windows. Each time window is tested ≥3 times, and each test corresponds to one test value, that is, each time window has ≥3 test values ​​of specific biomarkers.

[0076] Multiple detection values ​​of the r-th specific biomarker can be obtained within a preset time window, which can be any time window among all time windows in historical detection data. Based on the detection value of the r-th specific biomarker of the same patient within the preset time window and the detection values ​​of the r-th specific biomarker of the same patient in all time windows in historical detection data, the latent response coefficient of the r-th specific biomarker of the same patient for the same cancer within the preset time window can be determined. Then, based on the latent response coefficients of the same specific biomarker of multiple patients for the same cancer in all time windows, the compensation coefficient of the same specific biomarker for the same cancer can be determined to obtain the compensation coefficient of each specific biomarker for the same cancer.

[0077] In one embodiment, the compensation module is further configured to:

[0078] The abnormal pulse amplitude of the same specific biomarker for the same cancer within a preset time window is determined by the ratio of the range of detection values ​​of the same specific biomarker in the same patient within a preset time window to the mean of detection values ​​of the same specific biomarker in the same patient across all time windows in historical detection data.

[0079] The abnormal pulse frequency of the same specific biomarker for the same cancer within the preset time window is determined by the ratio of the number of abnormalities to the duration of the preset time window. The number of abnormalities indicates the number of times the detection value of the same specific biomarker of the same patient exceeds the preset detection value within the preset time window. The preset detection value is obtained by calculating the standard deviation and / or mean of the detection values ​​of the same specific biomarker of the same patient in all time windows in historical detection data.

[0080] Based on the product of the abnormal pulse amplitude and the abnormal pulse frequency, the latent response coefficient of the same specific biomarker for the same cancer in the same patient within a preset time window is determined.

[0081] Based on the latent response coefficient of the same specific biomarker for the same cancer in the same patient within a preset time window, the compensation coefficient of the same specific biomarker for the same cancer within the preset time window is determined, so as to obtain the compensation coefficient of each specific biomarker for the same cancer in all time windows for the same patient.

[0082] Multiple detection values ​​of the r-th specific biomarker for the same patient within a preset time window can be obtained. The preset time window can be any time window from all historical detection data. Then, the maximum and minimum values ​​among the multiple detection values ​​within the preset time window are determined, and the difference between the maximum and minimum values ​​is calculated, which is also the range of the multiple detection values ​​within the preset time window. Then, the mean of the detection values ​​of the r-th specific biomarker for the same patient across all time windows in historical detection data is obtained. The ratio of the range of the multiple detection values ​​to the mean of the detection values ​​across all time windows is then calculated to obtain the abnormal pulse amplitude of the r-th specific biomarker for the same cancer within the preset time window. The larger the abnormal pulse amplitude, the more obvious the pulse manifestation.

[0083] The number of times the detection value of the r-th specific biomarker of the same patient exceeds a preset detection value within a preset time window is determined as the number of abnormalities. The preset detection value can be twice the mean and / or standard deviation of the detection values ​​of the r-th specific biomarker of the same patient across all time windows in historical testing data. If the detection value of the r-th specific biomarker of the same patient within the preset time window exceeds twice the mean and / or standard deviation of the detection values ​​of the r-th specific biomarker of the same patient across all time windows, it is counted as an abnormality. Based on the ratio of the number of abnormalities to the duration of the preset time window, the abnormal pulse frequency of the r-th specific biomarker for the same cancer within the preset time window is determined.

[0084] For the same patient, the latent response coefficient of the i-th specific biomarker to the i-th cancer within the preset time window can be determined by multiplying the abnormal pulse amplitude of the r-th specific biomarker to the i-th cancer within the preset time window with the abnormal pulse frequency of the r-th specific biomarker to the i-th cancer within the preset time window. Latent response coefficient The larger the value, the more pronounced the abnormal pulse changes of the r-th specific biomarker for the i-th cancer during the historical detection process.

[0085] Then, based on the latent response coefficient of the r-th specific biomarker for the same cancer within a preset time window for the same patient, the compensation coefficient of the r-th specific biomarker for the i-th cancer within a preset time window is determined, so as to obtain the compensation coefficient of each specific biomarker for the i-th cancer for the same patient within all time windows.

[0086] In one embodiment, the compensation module is further configured to:

[0087] The latency response threshold is determined based on the maximum latency response coefficient for each patient; the maximum latency response coefficient is obtained by screening the maximum value from the latency response coefficients of the same specific biomarker for the same cancer across all time windows for each patient.

[0088] The compensation coefficient for the same specific biomarker for the same cancer is determined based on the ratio of the maximum latent response coefficient to the latent response threshold for each patient, thus obtaining the compensation coefficient for each specific biomarker for the same cancer.

[0089] In early risk assessment of bone metastasis, specific biomarkers face a dual dilemma: simply assigning zero weight to negative specific biomarkers results in the loss of their potential information, while applying fixed weights introduces noise. Therefore, a solution is to obtain testing data from multiple patients with the i-th cancer from historical testing data, extract multiple latent response coefficients of the r-th specific biomarker for each patient across all time windows for the same cancer, and then select the largest latent response coefficient from these coefficients to obtain the maximum latent response coefficient for the r-th specific biomarker for each patient. The latent response threshold is determined based on the maximum latent response coefficient. Then, the compensation coefficient for the same specific biomarker for the same cancer is determined based on the ratio of the maximum latent response coefficient to the latent response threshold, so as to obtain the compensation coefficient for each specific biomarker for the same cancer.

[0090] In one embodiment, the compensation module is further configured to:

[0091] The maximum latent response coefficient of each patient in multiple patients is sorted according to a preset rule to obtain the latent response sequence;

[0092] The maximum latent response coefficient at a preset position in the latent response sequence is determined as the latent response threshold.

[0093] The maximum latent response coefficient of multiple patients The latent response sequence is obtained by sorting the responses according to a preset rule, such as ascending order. Then, the maximum latent response coefficient at a preset position in the sequence, such as the maximum latent response coefficient at the 90th quantile, is determined as the latent response threshold of the r-th specific marker. If the latent response coefficient of the r-th specific biomarker exceeds the latent response threshold, then the r-th specific biomarker is determined to produce a significant abnormal pulse in the early stage of bone metastasis.

[0094] Calculate the ratio of the maximum latent response coefficient to the latent response threshold for the r-th specific biomarker for each patient. If the ratio is <1, it indicates a weak response from the specific biomarker, requiring compensation for the patient's abnormality in practice, i.e., complete compensation for the absence of the specific biomarker. If the ratio is ≥1, it indicates a strong response from the specific biomarker, and the ratio of the maximum latent response coefficient to the latent response threshold can be retained. Then, normalize the ratio to [0,1] using the norm function to obtain... , This is the ratio of the maximum latent response coefficient to the latent response threshold of the r-th specific biomarker for each patient after normalization, i.e. Let be the compensation coefficient for the r-th specific biomarker. When the maximum latent response coefficient of the r-th specific biomarker... When the value is close to 0, When the value approaches 0, analysis relies on the content of common biomarkers. The maximum latent response coefficient of the r-th specific biomarker... Value greater than the latent response threshold hour, Approaching 1, the original contribution of the specific biomarker is preserved.

[0095] Module 13 is used to determine the target coupling coefficient of each specific biomarker based on the correlation between each specific biomarker and each common biomarker among multiple patients.

[0096] This involves performing correlation analysis on the same specific biomarker and different common biomarkers for each patient across all testing processes to obtain multiple correlation scores, and then performing correlation analysis on the same common biomarker and different specific biomarkers for each patient across all testing processes to obtain multiple correlation scores.

[0097] In one embodiment, the determining module is further configured to:

[0098] Correlation analysis was performed on each specific biomarker and each common biomarker for each patient to obtain multiple correlation scores for each patient. The multiple correlation scores of the same patient correspond one-to-one with multiple biomarker combinations. Each biomarker combination includes one specific biomarker and one common biomarker. Any two different biomarker combinations may include different specific biomarkers and / or different common biomarkers.

[0099] In multiple biomarker combinations, the correlations corresponding to each biomarker combination are analyzed, the coupling coefficient corresponding to each biomarker combination is determined, and the multiple correlations corresponding to each biomarker combination correspond one-to-one with multiple patients;

[0100] From the multiple coupling coefficients corresponding to each specific biomarker, the largest coupling coefficient is selected to determine the target coupling coefficient for each specific biomarker, so as to obtain multiple target coupling coefficients corresponding to multiple specific biomarkers.

[0101] Correlation analysis was performed on the same specific biomarker and different common biomarkers for each patient across all testing processes, yielding multiple correlation scores. Then, correlation analysis was performed on the same common biomarker and different specific biomarkers for each patient across all testing processes, yielding multiple correlation scores. Each patient's multiple correlation scores correspond one-to-one with multiple biomarker combinations. Each biomarker combination includes one specific biomarker and one common biomarker; any two different biomarker combinations include different specific biomarkers and / or different common biomarkers. Within these multiple biomarker combinations, the mean of the correlation scores for each common biomarker corresponding to each specific biomarker across all testing processes was calculated. The absolute value was then normalized to the [0,1] interval to obtain the coupling coefficient for each biomarker combination. Each biomarker combination's multiple correlation scores correspond one-to-one with multiple patients. A larger coupling coefficient indicates a stronger association between the common and specific biomarkers, meaning the common biomarker can substitute for the specific biomarker in calculations; conversely, a smaller coupling coefficient indicates the common biomarker cannot substitute for the specific biomarker in calculations. Pearson correlation coefficients can be used for correlation analysis.

[0102] To avoid decision ambiguity caused by simultaneous compensation from multiple common biomarkers, the largest coupling coefficient among all common biomarkers can be selected as the target coupling coefficient for the r-th specific biomarker. This yields multiple target coupling coefficients corresponding to multiple specific markers.

[0103] Analysis module 14 is used to perform risk analysis based on sensitivity weights, compensation coefficients, target coupling coefficients, and the target growth rate of each common marker, and obtain analysis results.

[0104] The target growth rate is determined based on the difference between the same common biomarker of the same patient at a first time and a second time. The first time can be the time when the same patient underwent the first testing process in historical testing data, and the first testing process is the latest testing process. The second time can be the time when the same patient underwent the second testing process in historical testing data, and the second testing process is the testing process preceding the first testing process.

[0105] Understandably, the difference in detection value of the xth common biomarker between the first and second times is obtained by subtracting its detection value from the detection value at the first time point. The time interval between the first and second times is then calculated. Based on the ratio of the detection value difference to the time interval, the target growth rate of the xth common biomarker is calculated. .

[0106] In one embodiment, the analysis module is also used for:

[0107] The difference between the detection value of the same specific biomarker of the same patient at the third time point and the mean detection value of the same specific biomarker of the same patient in all time windows in the historical detection data is calculated. Based on the ratio of the difference to the mean detection value of the same specific biomarker of the same patient in all time windows in the historical detection data, the change coefficient of the same specific biomarker is determined to obtain the change coefficient of each specific biomarker.

[0108] Risk analysis is performed based on the change coefficient, the sensitivity weight, the compensation coefficient, the target coupling coefficient, and the target growth rate of each common marker to obtain the analysis results.

[0109] In historical testing data, the detection value of the r-th specific biomarker for any patient with the i-th type of cancer at time 3 (the third time being the moment of the patient's most recent testing) is obtained. The mean detection value of the r-th specific biomarker for the same patient across all time windows in the historical testing data is calculated. Then, the difference between the detection value of the specific biomarker at time 3 and the mean detection value is calculated. By calculating the ratio of this difference to the mean detection value of the r-th specific biomarker across all time windows, the coefficient of variation of the r-th specific biomarker is determined. The variation coefficient of each specific biomarker was obtained. Risk analysis was then performed based on the variation coefficient, sensitivity weight, compensation coefficient, target coupling coefficient, and target growth rate of each common biomarker to obtain the analysis results.

[0110] When assessing the risk of patients, the combined effects of common and specific biomarkers should be considered. However, in the early stages of bone metastasis, there is often a contradictory phenomenon where specific biomarkers do not increase but bone destruction accelerates, rendering a single assessment dimension ineffective. Therefore, a two-way information channel can be established to enable the two types of biomarkers to form a dynamic balance in the assessment. When one fails, the other automatically fills the gap, and when both are effective, they work synergistically.

[0111] Therefore, in one embodiment, the analysis module is further used for:

[0112] The compensation integral of each specific biomarker is determined based on the coefficient of change of each specific biomarker, the target growth rate of each common biomarker, the target coupling coefficient of each specific biomarker, and the compensation coefficient of each specific biomarker for the same cancer.

[0113] The compensation integral of the common biomarker is determined based on the sensitivity weight of each common biomarker for the same cancer and the target growth rate of each common biomarker.

[0114] Risk analysis is performed based on the sum of the compensation integrals of the specific markers and the compensation integrals of the common markers to obtain the analysis results.

[0115] Based on the compensation coefficient obtained in the above steps It can be used to correct non-specific fluctuations in common biomarkers when specific biomarkers are strongly positive, and to adjust the reliability of specific biomarkers. The compensation integral of a specific biomarker can be determined based on the variation coefficient of each specific biomarker, the target growth rate of each common biomarker, the target coupling coefficient of each specific biomarker, and the compensation coefficient of each specific biomarker for the same cancer. The formula for the compensation integral of a specific biomarker is as follows:

[0116] ;

[0117] in, Let M be the compensation integral for the specific biomarker of the i-th cancer, M represent the total number of specific biomarkers for the i-th cancer, and r represent the r-th specific biomarker. The compensation coefficient characterizing the r-th specific biomarker for the i-th cancer. The target coupling coefficient characterizing the r-th specific marker, The target growth rate characterizing the xth common marker. The coefficient of variation characterizing the r-th specific biomarker . The larger the value, the greater the reference weight of the change in the xth common marker. The larger the value, the greater the reference weight of the change in the r-th specific biomarker, and the greater the compensation coefficient of the r-th specific biomarker for the i-th cancer. Weighting is performed to obtain the compensation integral for the specific biomarker of the i-th cancer. The compensation score for the specific biomarker of the i-th cancer The larger the value, the greater the probability that the current patient has bone metastasis of the i-th cancer.

[0118] Simultaneously, based on the sensitivity weight of each common biomarker for the same cancer and the target growth rate of each common biomarker, the compensation score for the common biomarker is determined. Compensation integral of common markers The formula is as follows:

[0119] ;

[0120] in, The compensation integral represents the common biomarkers of the i-th cancer, where G represents the total number of common biomarkers and x represents the x-th common biomarker. The sensitivity weight of the x-th common biomarker for the i-th cancer is represented. The target growth rate characterizing the x-th common biomarker. The compensation integral of the common biomarker for the i-th cancer. The larger the value, the greater the probability that the current patient has bone metastasis of the i-th cancer.

[0121] Risk analysis is performed by summing the compensation integrals of specific and common biomarkers to obtain the current patient's risk probability of bone metastasis. This probability is then normalized using a norm function to obtain the final normalized risk probability. When the final risk probability is higher than a risk threshold, further investigation is required, including imaging and biopsy. The final risk probability is output to assist physicians in assessing the patient's condition based on the final risk probability and the results of further examinations. When the final risk probability is lower than the risk threshold, it helps physicians determine the patient's condition. The final risk probability can be obtained for different types of cancer. Data from each patient's biomarker examinations is stored in a computer and updated regularly to improve the accuracy of subsequent testing.

[0122] This application's embodiments design a latent response coefficient to quantify the abnormal impulse characteristics of specific biomarkers. Combined with population statistical thresholds, a compensation coefficient is dynamically generated to achieve intelligent weight decay and automatic switching to common biomarkers as proxies in false-negative scenarios, while retaining specific decision-making in strong-positive scenarios, thus breaking the assessment deadlock. A target coupling coefficient is introduced to dynamically bind the biomarker pair with the strongest pathological association. A one-to-one proxy mechanism avoids interference from multiple biomarker noise, accurately transferring to highly relevant common biomarkers when specific biomarkers fail. A dual-channel integral model is constructed, combining sensitive weights and compensation coefficients to correct for individual differences, outputting the normalized final risk probability. This significantly improves the early detection capability of bone metastases, greatly reduces the false positive rate, reduces unnecessary imaging examinations, provides early warning, and continuously optimizes the system through closed-loop data updates, providing highly reliable dynamic decision support for clinical practice. The synchronicity of changes in common and specific biomarkers is quantified. When a specific biomarker fails, the most relevant common biomarker is selected to proxy its function. The changing trends of common and specific biomarkers are dynamically bound through the target coupling coefficient. When the target coupling coefficient is large, the common biomarker can proxy the function of the specific biomarker. When dealing with scenarios with high compensation coefficients, specific biomarkers are used as the primary indicator. In scenarios with low compensation coefficients, the problem of missed detection can be solved by shifting the weight to highly relevant common biomarkers.

[0123] This application also provides a method for predicting tumor bone metastasis based on bone metastasis-specific bone markers, such as... Figure 2 As shown, the method includes:

[0124] S21. Calculate the sensitivity weight of each common biomarker for the same cancer based on the metabolic rate of each common biomarker in multiple patients and the number of patients; the metabolic rate is determined based on the difference between adjacent time points for the same common biomarker in the same patient.

[0125] S22. Based on the difference between the detection value of each specific biomarker for the same patient and historical detection data, determine the compensation coefficient for each specific biomarker for the same cancer; the compensation coefficient is used to adjust the weight of the specific biomarker;

[0126] S23. Determine the target coupling coefficient for each specific biomarker based on the correlation between each specific biomarker and each common biomarker in multiple patients;

[0127] S24. Perform risk analysis based on the sensitivity weight, the compensation coefficient, the target coupling coefficient, and the target growth rate of each common biomarker to obtain the analysis results; the target growth rate is determined based on the difference between the same common biomarker of the same patient at the first time point and the second time point.

[0128] For the method embodiments, since they are basically corresponding to the system embodiments, the relevant parts can be referred to in the description of the system embodiments.

[0129] It should be understood that, although Figure 2 The steps in the flowchart are shown sequentially as indicated by the arrows, but these steps are not necessarily executed in the order indicated by the arrows. Unless otherwise specified herein, there is no strict order in which these steps are executed, and they can be performed in other orders. Figure 2 At least some of the steps in the process may include multiple sub-steps or multiple stages. These sub-steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these sub-steps or stages is not necessarily sequential, but can be executed in turn or alternately with other steps or at least some of the sub-steps or stages of other steps.

[0130] Figure 3 This is a schematic diagram of the structure of an electronic device according to an example embodiment of this application. The electronic device includes a memory, a processor, and a computer program stored in the memory and used to run on the processor. When the processor executes the computer program, it implements the method described in any of the above embodiments. Figure 3 The electronic device 30 shown is merely an example and should not impose any limitations on the functionality and scope of use of the embodiments of this application.

[0131] like Figure 3 As shown, the electronic device 30 can be manifested as a general-purpose computing device, such as a server device. The components of the electronic device 30 may include, but are not limited to: at least one processor 31, at least one memory 32, and a bus 33 connecting different system components (including memory 32 and processor 31).

[0132] Bus 33 includes a data bus, an address bus, and a control bus.

[0133] The memory 32 may include volatile memory, such as random access memory (RAM) 321 and / or cache memory 322, and may further include read-only memory (ROM) 323.

[0134] The memory 32 may also include a program tool 325 (or utility) having a set (at least one) program module 324, such program module 324 including but not limited to: an operating system, one or more application programs, other program modules and program data, each or some combination of these examples may include an implementation of a network environment.

[0135] The processor 31 executes various functional applications and data processing, such as the methods provided in any of the above embodiments, by running computer programs stored in the memory 32.

[0136] Electronic device 30 can also communicate with one or more external devices 34 (e.g., keyboard, pointing device, etc.). This communication can be performed via input / output (I / O) interface 35. Furthermore, electronic device 30 can also communicate with one or more networks (e.g., local area network (LAN), wide area network (WAN), and / or public network, such as the Internet) via network adapter 36. As shown, network adapter 36 communicates with other modules of electronic device 30 via bus 33. It should be understood that, although not shown in the figure, other hardware and / or software modules can be used in conjunction with electronic device 30, including but not limited to: microcode, device drivers, redundant processors, external disk drive arrays, RAID (disk array) systems, tape drives, and data backup storage systems.

[0137] It should be noted that although several units / modules or sub-units / modules of the electronic device have been mentioned in the detailed description above, this division is merely exemplary and not mandatory. In fact, according to the embodiments of this application, the features and functions of two or more units / modules described above can be embodied in one unit / module. Conversely, the features and functions of one unit / module described above can be further divided and embodied by multiple units / modules.

[0138] This application also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the method provided in any of the above embodiments.

[0139] The readable storage medium may be more specifically adopted, including but not limited to: portable disk, hard disk, random access memory, read-only memory, erasable programmable read-only memory, optical storage device, magnetic storage device, or any suitable combination thereof.

[0140] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods. Any references to memory, storage, databases, or other media used in the embodiments provided in this application can include non-volatile and / or volatile memory. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory can include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in various forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), dual data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), Rambus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), etc.

[0141] This application also provides a computer program product, including a computer program that, when executed by a processor, implements the method described in any of the above embodiments.

[0142] The program code for executing the computer program product of this application can be written in any combination of one or more programming languages. The program code can be executed entirely on the user device, partially on the user device, as a standalone software package, partially on the user device and partially on a remote device, or entirely on a remote device.

[0143] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0144] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of the application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these modifications and improvements all fall within the protection scope of this application.

[0145] The various embodiments in this specification are described in a progressive manner. The same or similar parts between the various embodiments can be referred to each other. Each embodiment focuses on describing the differences from other embodiments.

Claims

1. A tumor bone metastasis prediction system based on bone metastasis-specific bone markers, characterized in that, The system includes: The calculation module is used to calculate the sensitivity weight of each common biomarker for the same cancer based on the metabolic rate of each common biomarker in multiple patients and the number of multiple patients; the metabolic rate is determined based on the difference of the same common biomarker in the same patient between adjacent time points; The compensation module is used to determine the compensation coefficient for each specific biomarker for the same cancer based on the difference between the detection value of each specific biomarker and historical detection data for the same patient; the compensation coefficient is used to adjust the weight of the specific biomarker. The determination module is used to determine the target coupling coefficient of each specific biomarker based on the correlation between each specific biomarker and each common biomarker in multiple patients; The analysis module is used to perform risk analysis based on the sensitivity weight, the compensation coefficient, the target coupling coefficient, and the target growth rate of each common biomarker to obtain analysis results; the target growth rate is determined based on the difference between the same common biomarker of the same patient at a first time point and a second time point; The computing module is also used for: The growth rate of the common biomarker in the same patient at adjacent time points is calculated based on the difference in detection values ​​of the same common biomarker at adjacent time points; the difference in detection values ​​is obtained by the difference between the detection values ​​of the same common biomarker at the third time point and the detection values ​​of the same common biomarker at the fourth time point, where the fourth time point is the time point preceding the third time point. The metabolic rate of the same common biomarker in the same patient can be determined by the rate of increase of the same common biomarker at all adjacent time points. For each common biomarker, the sensitivity coefficient of each common biomarker for the same cancer is determined based on the ratio of the number of sensitive patients to the total number of patients. The metabolic mean is then weighted according to the sensitivity coefficient to obtain the sensitivity weight of each common biomarker for the same cancer. The number of sensitive patients is the number of patients among the multiple patients whose metabolic rate for the corresponding common biomarker is positive. The metabolic mean indicates the average metabolic rate of the corresponding common biomarker among the multiple patients. The compensation module is also used for: Based on the detection values ​​of the same specific biomarker of the same patient within a preset time window and the detection values ​​of the same specific biomarker of the same patient within all time windows in historical detection data, the latent response coefficient of the same specific biomarker of the same patient for the same cancer within the preset time window is determined, so as to obtain the latent response coefficient of the same specific biomarker of the same patient for the same cancer within all time windows; the all time windows include the preset time window. Based on the latent response coefficients of the same specific biomarker for the same cancer in multiple patients across all time windows, the compensation coefficient of the same specific biomarker for the same cancer is determined to obtain the compensation coefficient of each specific biomarker for the same cancer. The determining module is also used for: Correlation analysis was performed on each specific biomarker and each common biomarker for each patient to obtain multiple correlation scores for each patient. The multiple correlation scores of the same patient corresponded one-to-one with multiple biomarker combinations. Each biomarker combination included one specific biomarker and one common biomarker. Any two different biomarker combinations included different specific biomarkers and / or different common biomarkers. In multiple biomarker combinations, the correlations corresponding to each biomarker combination are analyzed, the coupling coefficient corresponding to each biomarker combination is determined, and the multiple correlations corresponding to each biomarker combination correspond one-to-one with multiple patients; From the multiple coupling coefficients corresponding to each specific biomarker, the largest coupling coefficient is selected to determine the target coupling coefficient for each specific biomarker, so as to obtain multiple target coupling coefficients corresponding to multiple specific biomarkers.

2. The tumor bone metastasis prediction system based on bone metastasis-specific bone markers as described in claim 1, characterized in that, The compensation module is also used for: The abnormal pulse amplitude of the same specific biomarker for the same cancer within a preset time window is determined by the ratio of the range of detection values ​​of the same specific biomarker in the same patient within a preset time window to the mean of detection values ​​of the same specific biomarker in the same patient across all time windows in historical detection data. Based on the ratio of the number of abnormal events to the duration of a preset time window, the abnormal pulse frequency of the same specific biomarker for the same cancer within the preset time window is determined. The abnormal number indicates the number of times the detection value of the same specific biomarker of the same patient exceeds the preset detection value within a preset time window. The preset detection value is obtained by calculating the standard deviation and / or mean value of the detection values ​​of the same specific biomarker of the same patient in all time windows in historical detection data. Based on the product of the abnormal pulse amplitude and the abnormal pulse frequency, the latent response coefficient of the same specific biomarker for the same cancer in the same patient within a preset time window is determined. Based on the latent response coefficient of the same specific biomarker for the same cancer in the same patient within a preset time window, the compensation coefficient of the same specific biomarker for the same cancer within the preset time window is determined, so as to obtain the compensation coefficient of each specific biomarker for the same cancer in all time windows for the same patient.

3. The tumor bone metastasis prediction system based on bone metastasis-specific bone markers as described in claim 1, characterized in that, The compensation module is also used for: The latency response threshold is determined based on the maximum latency response coefficient for each patient; the maximum latency response coefficient is obtained by screening the maximum value from the latency response coefficients of the same specific biomarker for the same cancer in all time windows for each patient. The compensation coefficient for the same specific biomarker for the same cancer is determined based on the ratio of the maximum latent response coefficient to the latent response threshold for each patient, thus obtaining the compensation coefficient for each specific biomarker for the same cancer.

4. The tumor bone metastasis prediction system based on bone metastasis-specific bone markers as described in claim 3, characterized in that, The compensation module is also used for: The maximum latent response coefficient of each patient in multiple patients is sorted according to a preset rule to obtain the latent response sequence; The maximum latent response coefficient at a preset position in the latent response sequence is determined as the latent response threshold.

5. The tumor bone metastasis prediction system based on bone metastasis-specific bone markers as described in claim 1, characterized in that, The analysis module is also used for: The difference between the detection value of the same specific biomarker of the same patient at the third time point and the mean detection value of the same specific biomarker of the same patient in all time windows in the historical detection data is calculated. Based on the ratio of the difference to the mean detection value of the same specific biomarker of the same patient in all time windows in the historical detection data, the variation coefficient of the same specific biomarker is determined to obtain the variation coefficient of each specific biomarker. Risk analysis is performed based on the change coefficient, the sensitivity weight, the compensation coefficient, the target coupling coefficient, and the target growth rate of each common marker to obtain the analysis results.

6. The tumor bone metastasis prediction system based on bone metastasis-specific bone markers as described in claim 5, characterized in that, The analysis module is also used for: The compensation integral of each specific biomarker is determined based on the coefficient of change of each specific biomarker, the target growth rate of each common biomarker, the target coupling coefficient of each specific biomarker, and the compensation coefficient of each specific biomarker for the same cancer. The compensation integral of the common biomarker is determined based on the sensitivity weight of each common biomarker for the same cancer and the target growth rate of each common biomarker. Risk analysis is performed based on the sum of the compensation integrals of the specific markers and the compensation integrals of the common markers to obtain the analysis results.

7. A method for predicting tumor bone metastasis based on bone metastasis-specific bone markers, characterized in that, The method is applied to the tumor bone metastasis prediction system based on bone metastasis-specific bone markers according to any one of claims 1-6, and the method includes: The sensitivity weight of each common biomarker for the same cancer is calculated based on the metabolic rate of each common biomarker in multiple patients and the number of patients; the metabolic rate is determined based on the difference of the same common biomarker in the same patient between adjacent time points; Based on the differences between the detection values ​​of each specific biomarker for the same patient and historical detection data, a compensation coefficient for each specific biomarker for the same cancer is determined; the compensation coefficient is used to adjust the weight of the specific biomarker. The target coupling coefficient for each specific biomarker is determined based on the correlation between each specific biomarker and each common biomarker in multiple patients. Risk analysis is performed based on the sensitivity weight, the compensation coefficient, the target coupling coefficient, and the target growth rate of each common biomarker to obtain the analysis results; the target growth rate is determined based on the difference between the same common biomarker of the same patient at the first time point and the second time point.

Images