Orthopedic trauma postoperative infection risk intelligent early warning system

Abstract

The invention relates to the technical field of health risk assessment, in particular to an intelligent early warning system for postoperative infection risk of orthopedic trauma. The system comprises a data acquisition module for acquiring physiological feature data and biomarker data; the key monitoring and screening module screens key monitoring patients according to the continuous abnormal representation growth degree of the initial physiological signs; the infection risk assessment module obtains a marker level index by using the initial deviation level and the growth amplitude of the biomarker, and obtains a wound infection risk coefficient in combination with the prediction condition of historical biomarker data; and the early warning report generation module is used for dividing patient early warning levels according to the physiological parameter abnormality and the wound infection risk coefficient to generate a risk report. According to the method, the evaluation accuracy of postoperative infection is improved through the physiological sign abnormity continuity and the biomarker sudden increase and prediction trend, the subsequent early warning level and risk report generation are more reliable, and a more personalized and accurate treatment scheme can be provided for a patient.

Description

Technical Field

[0001] The present invention relates to the technical field of health risk assessment, and particularly to an intelligent early warning system for the risk of postoperative infection in orthopedic trauma. Background Art

[0002] Postoperative infection in orthopedic trauma is one of the common complications after orthopedic surgery, which seriously affects the rehabilitation process of patients, may lead to an extended hospital stay, increased treatment costs, and even cause serious complications. In order to minimize losses as much as possible, it is necessary to develop an intelligent early warning system for the risk of postoperative infection in orthopedic trauma, which can monitor the physiological conditions of patients in real time, and combine artificial intelligence and data analysis technologies to achieve early identification and warning. Thus, the prediction ability of postoperative infection can be significantly improved, and more personalized and accurate treatment plans can be provided for patients.

[0003] Currently, the detection of postoperative infection in orthopedic patients is generally based on the real-time physiological data of patients and the concentration changes of corresponding biomarkers of infection to predict whether there is an infection risk in patients. However, some inflammatory reactions may occur in postoperative patients, resulting in the same change effect of biomarkers. Infectious inflammation is caused by the invasion of pathogens into the wound and requires antibiotic or debridement treatment. Non-infectious inflammation is caused by physiological processes such as surgical trauma and tissue repair and does not require anti-infection treatment. Since the biomarkers caused by inflammatory reactions have the same trend as infections, it leads to inaccurate analysis and judgment of infections, interferes with the monitoring of the infection risk at the patient's wound, and affects the subsequent care of patients. Summary of the Invention

[0004] In order to solve the technical problem in the prior art that the biomarkers caused by inflammatory reactions have the same trend as infections, resulting in inaccurate analysis and judgment of infections, the purpose of the present invention is to provide an intelligent early warning system for the risk of postoperative infection in orthopedic trauma. The specific technical solutions adopted are as follows: An embodiment of the present invention provides an intelligent early warning system for the risk of postoperative infection in orthopedic trauma, and the system includes: A data acquisition module, which is used to obtain the physiological sign data at each sampling moment in the initial sampling period and the biomarker data at each key sampling moment after the initial sampling period for each orthopedic trauma patient after surgery; the physiological signs include body temperature, heart rate, and blood oxygen saturation, and the biomarker data is the concentration of C-reactive protein; A key monitoring and screening module, which is used to obtain the physiological parameter abnormality degree of each patient according to the growth degree of the continuous abnormal characterization of the physiological sign data of each patient in the initial sampling period in time series; determine the key monitoring patients based on the magnitude of the physiological parameter abnormality degree; An infection risk assessment module, which is used to obtain a biomarker level index for each key monitored patient through the deviation level of biomarker data at the initial key sampling moment and the growth level at the report generation moment; after predicting the biomarker data based on the change trend of consecutive biomarker data before the report generation moment, analyze the growth change trend of all biomarker data in time series, and combine the biomarker level index to obtain the wound infection risk coefficient of each key monitored patient at the report generation moment; An early warning report generation module, which is used to combine the wound infection risk coefficient and the physiological parameter abnormality degree of each key monitored patient to obtain the early warning level of each key monitored patient; obtain the risk report of the key monitored patient at the report generation moment.

[0005] Furthermore, the method for obtaining the physiological parameter abnormality degree includes: For each patient, obtain the physiological characteristic factor at each sampling moment by taking the ratio of the product of the body temperature data and the heart rate data to the blood oxygen saturation data at each sampling moment; Arrange the physiological characteristic factors in chronological order to obtain a physiological characteristic sequence; based on the continuous growth distribution of the physiological characteristic factors in the physiological characteristic sequence, screen out the abnormally growing sequence; Take the ratio of the total number of physiological characteristic factors in the abnormally growing sequence to the total number of physiological characteristic factors in the physiological characteristic sequence as the abnormal proportion; perform a negative correlation mapping on the interval period between the sampling moment corresponding to the first physiological characteristic factor in the abnormally growing sequence and the first sampling moment in the initial sampling period to obtain the time difference abnormality degree; Combine the abnormal proportion and the time difference abnormality degree to obtain the physiological parameter abnormality degree of each patient.

[0006] Furthermore, the method for obtaining the abnormally growing sequence includes: Obtain the difference sequence of the physiological characteristic sequence; record the difference values greater than zero in the difference sequence as growth difference values; Take each consecutive adjacent growth difference value as a positive number sequence; take the positive number sequence with the most growth difference values as the target sequence; Take the sequence composed of the physiological characteristic factors in the physiological characteristic sequence corresponding to each growth difference value in the target sequence as the abnormally growing sequence.

[0007] Furthermore, determining the key monitored patients based on the magnitude of the physiological parameter abnormality degree includes: When the value after normalizing the physiological parameter abnormality degree is greater than or equal to the preset attention threshold, mark the corresponding patient as a key monitored patient.

[0008] Furthermore, the acquisition level of the biomarker level index includes: For each key monitored patient, the difference between the biomarker data of the patient at the first key sampling moment and the normal biomarker data is used as the initial deviation index; the difference between the biomarker data of the patient between the report generation moment and the first key sampling moment is used as the growth rate index; Combining the initial deviation index and the growth rate index of the patient, the biomarker level index of the patient at the report generation moment is obtained.

[0009] Furthermore, the method for obtaining the wound infection risk coefficient includes: According to the growth change degree of the continuous biomarker data before the report generation moment, each unsampled key sampling moment after the report generation moment is predicted to obtain the predicted biomarker data; Based on all the biomarker data and the predicted biomarker data in time series, according to the data growth deviation between each adjacent key sampling moment, the growth trend index is obtained; Combining the growth trend index and the biomarker level index, the wound infection risk coefficient of each key monitored patient at the report generation moment is obtained.

[0010] Furthermore, the method for obtaining the predicted biomarker data includes: Taking the report generation moment and all unsampled key sampling moments after the moment as the analysis moments; traversing each analysis moment in chronological order, before the analysis moment, after calculating the difference between the biomarker data of the latter key sampling moment and the former key sampling moment in each adjacent two key sampling moments, the average value of the differences between all adjacent key sampling moments is obtained as the average difference of the analysis moment; Taking the sum of the biomarker data of the analysis moment and the average difference as the predicted biomarker data of the next analysis moment.

[0011] Furthermore, the method for obtaining the growth trend index includes: Recording the biomarker data and the predicted biomarker data of all key sampling moments with corresponding data as the target data; After calculating the difference between the biomarker data of the latter key sampling moment and the former key sampling moment in each adjacent key sampling moment, the average value of the differences between the target data of all adjacent key sampling moments is obtained as the growth trend index.

[0012] Furthermore, the method for obtaining the warning level includes: Normalizing the product of the wound infection risk coefficient and the physiological parameter abnormality degree of each key monitored patient to obtain the classification index of each key monitored patient; When the classification index is less than the preset low threshold, the warning level of the corresponding key monitored patient is recorded as the low warning level; When the classification index is greater than or equal to the preset low threshold and less than the preset medium threshold, the warning level of the corresponding key monitored patient is recorded as the medium warning level; When the classification index is greater than or equal to the preset medium threshold, the warning level of the corresponding key monitored patient is recorded as the high warning level.

[0013] Furthermore, the risk report includes: patient basic information, warning level, historical physiological characteristic data, and the risk coefficient of wound infection.

[0014] The present invention has the following beneficial effects: Based on the growth degree of the continuous abnormal manifestations of the initial physiological signs of orthopedic trauma patients after surgery, the present invention screens key monitored objects, accurately identifies high-risk patient groups, and reduces the consumption of unnecessary high-cost biomarker detection resources. By using the deviation level of biomarkers at the initial key sampling moment and the growth degree at the report generation moment, a biomarker level index is constructed. Combining the prediction situation of historical biomarker data, the possible overall dynamic change characteristics in time series are analyzed to obtain the risk coefficient of wound infection. By integrating the sudden increase characteristics and continuous growth amplitude of biomarker levels, the normal inflammatory response and infectious inflammation after surgery are distinguished, significantly improving the specificity of infection determination. Combining the abnormal degree of physiological parameters and the risk coefficient of wound infection to divide the warning level of patients, and generating a customized risk report based on the individual characteristics of patients, providing a quantitative basis for clinical intervention. Through the collaborative analysis of the continuous abnormal physiological signs and the sudden increase and prediction trend of biomarkers, the present invention improves the accuracy of the assessment of postoperative infection in orthopedic trauma, makes the subsequent warning level division and risk report generation more reliable, and helps to provide more personalized and accurate treatment plans for patients. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] In order to more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for the description of the embodiments or the prior art. Obviously, the following drawings are only some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0016] Figure 1 It is a structural diagram of an intelligent early warning system for postoperative infection risk in orthopedic trauma provided by an embodiment of the present invention; Figure 2 It is a schematic diagram of the wound surface of bone wound infection provided by an embodiment of the present invention; Figure 3 It is a schematic diagram of bone wound surface image provided by an embodiment of the present invention; Figure 4 It is a schematic diagram of the change of biomarker level provided by an embodiment of the present invention; Figure 5 Flowchart of a method for obtaining the risk coefficient of wound infection provided by an embodiment of the present invention. Detailed implementation manners

[0017] In order to further elaborate on the technical means and effects adopted by the present invention to achieve the predetermined invention purpose, the following combines the accompanying drawings and preferred embodiments to elaborate in detail on a smart early warning system for the risk of postoperative infection in orthopedic trauma proposed according to the present invention, including its specific implementation manners, structures, features and effects. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. In addition, the specific features, structures or characteristics in one or more embodiments can be combined in any suitable form.

[0018] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by those skilled in the technical field to which the present invention belongs.

[0019] The following specifically describes the specific solution of a smart early warning system for the risk of postoperative infection in orthopedic trauma provided by the present invention with reference to the accompanying drawings.

[0020] Since the wound surface after orthopedic surgery needs to be tightly bandaged and fixed during the postoperative recovery process, in order to determine the real-time wound infection situation of the patient and give an early warning, it is less feasible to directly take real-time pictures of the bone wound surface image, and the postoperative infection of the patient does not necessarily occur. In order to reduce costs and improve feasibility, this solution uses the method of initial sampling - verification - resampling for early warning. During the postoperative recovery process of orthopedic trauma patients, if the wound care is improper or the fracture fixation method is improper, it may lead to wound infection. Please refer to Figure 2 , which shows a schematic diagram of the wound surface of bone wound infection provided by an embodiment of the present invention. Please refer to Figure 3 , which shows a schematic diagram of the bone wound surface image provided by an embodiment of the present invention.

[0021] To improve, please refer to Figure 1 , which shows a structural diagram of a smart early warning system for the risk of postoperative infection in orthopedic trauma provided by an embodiment of the present invention. The system includes: a data acquisition module 101, a key monitoring and screening module 102, an infection risk assessment module 103, and an early warning report generation module 104.

[0022] The data acquisition module 101 is used to obtain the physiological sign data at each sampling moment during the initial sampling period and the biomarker data at each key sampling moment after the initial sampling period for each orthopedic trauma patient after surgery; the physiological signs include body temperature, heart rate, and blood oxygen saturation, and the biomarker data is the C-reactive protein concentration.

[0023] In the initial stage of general wound infection, it is accompanied by changes in the patient's own physiological parameters, such as a slight increase in body temperature, an increase in heart rate, and a decrease in blood oxygen saturation. In the embodiments of the present invention, after the operation, physiological sign data can be collected through clinical wearable monitoring devices, and the real-time changes in the signs can be analyzed to verify whether the patient needs to be monitored intensively. The initial sampling period is set to 2 days, and the sampling interval at each sampling moment is 5 seconds. The implementer can adjust the specific sampling settings according to the specific implementation scenario.

[0024] If a patient needs to be monitored intensively, it indicates that the patient may have a relatively high risk of infection. Therefore, after the initial sampling period, the concentration of the patient's biomarker is further collected. Biomarkers will show significant fluctuations during the occurrence of infection. These monitors can help evaluate the risk of postoperative infection. C-reactive protein is an acute-phase reactant, and its level rapidly increases when there is inflammation or infection in the body. Using the blood analysis method, in the embodiments of the present invention, the key sampling moment of the biomarker is set to once a day. In other embodiments of the present invention, the sampling frequency can be determined according to the time interval of the patient's dressing change, which is not limited here. Thus, the level of the inflammatory response in the patient's body can be quickly obtained to help with early infection warning diagnosis.

[0025] The key monitoring and screening module 102 is used to obtain the physiological parameter abnormality degree of each patient based on the growth degree of the continuous abnormal representation of the physiological sign data of each patient in the time series during the initial sampling period; and determine the patients to be monitored intensively based on the magnitude of the physiological parameter abnormality degree.

[0026] Since the changes in the patient's physiological indicators are relatively more obvious and easier to sample in the initial stage of wound infection, during the initial sampling stage, the recovery status of the patient is determined by comparing the real-time physiological parameters. If there are no obvious changes in the patient's physiological parameters, it indicates a good recovery status, and monitoring continues. If the fluctuations in the physiological parameters are similar to the trend of the infection situation, it indicates that the patient may have a wound infection, and it is necessary to focus on monitoring and analyzing in combination with the resampling data.

[0027] Preferably, in the embodiments of the present invention, the method for obtaining the physiological parameter abnormality degree includes: Since in the initial stage of infection, it is accompanied by a slight increase in body temperature, an increase in heart rate, and a decrease in blood oxygen saturation of physiological signs, for each patient, the ratio of the product of the body temperature data and the heart rate data at each sampling moment to the blood oxygen saturation data is obtained to obtain the physiological characteristic factor at each sampling moment. As an example, the expression of the physiological characteristic factor is: ; in the formula, represents the physiological characteristic factor at the th sampling moment, represents the body temperature data at the th sampling moment, Represented as the heart rate data at the th sampling moment, Represented as the blood oxygen saturation data at the th sampling moment, where the body temperature, heart rate are directly proportional to the physiological factor, and the blood oxygen saturation is inversely proportional to the physiological factor. That is, the more significant the patient's infection symptoms, the larger the corresponding physiological factor.

[0028] Arrange the physiological characteristic factors in chronological order to obtain a physiological characteristic sequence. In actual situations, it is not that the physiological characteristic factor at a certain moment that more conforms to the infection characteristics can indicate that the patient may have an infection. Generally, an infection is accompanied by gradually obvious infection symptoms, that is, a continuous increase in the physiological characteristic factor. Therefore, based on the continuous growth distribution of the physiological characteristic factors in the physiological characteristic sequence, an abnormal growth sequence is screened out.

[0029] In the embodiment of the present invention, obtain the difference sequence of the physiological characteristic sequence, and record the difference values greater than zero in the difference sequence as growth difference values. When the difference value is greater than zero, it represents the situation that the subsequent physiological characteristic factor in the time series is greater than the previous physiological characteristic factor. Take each continuously adjacent growth difference value as a positive number sequence. That is, traverse the difference sequence. Whenever a growth difference value is traversed, the growth difference values are sequentially placed into another sequence until the next difference value is not a growth difference value to complete the acquisition of a sequence as the positive number sequence, and continue to traverse to the next growth difference value to obtain a new positive number sequence.

[0030] Among multiple positive number sequences, take the positive number sequence containing the most growth difference values as the target sequence, which reflects the situation of the most abnormal continuous growth of the physical sign factors. According to the values corresponding to the target sequence in the physiological characteristic sequence, form an abnormal growth sequence of the growth characteristic factors, and take the sequence composed of the physiological characteristic factors corresponding to each growth difference value in the target sequence as the abnormal growth sequence.

[0031] Further, take the ratio of the total number of physiological characteristic factors in the abnormal growth sequence to the total number of physiological characteristic factors in the physiological characteristic sequence as the abnormal ratio. The abnormal ratio represents the proportion of the sampling moments that conform to the infection characteristics in the initial sampling period. The larger its value, the more likely it is to have infection symptoms and the greater the degree of abnormality.

[0032] Perform a negative correlation mapping on the time interval between the sampling moment corresponding to the first physiological feature factor in the abnormal growth sequence and the first sampling moment in the initial sampling period to obtain the time difference abnormality degree. The time difference abnormality degree characterizes the time difference between the sampling moment conforming to the infection feature and the initial sampling moment. Since the infection risk is relatively greater just after the operation, the smaller the time difference abnormality degree, the closer the appearance of the infection feature is to the operation completion moment, and the greater the corresponding abnormality degree. It should be noted that the smaller the time difference abnormality degree, the greater the physiological parameter abnormality degree. The time difference abnormality degree is inversely proportional to the physiological parameter abnormality degree. The negative correlation mapping can be performed in the form of an inverse proportional or negative exponential function. The negative correlation mapping method is a well-known technical means for those skilled in the art and will not be elaborated and limited herein.

[0033] Therefore, finally, by combining the abnormal proportion and the time difference abnormality degree, the physiological parameter abnormality degree of each patient is obtained. In the embodiment of the present invention, the product of the abnormal proportion and the time difference abnormality degree is used as the physiological parameter abnormality degree of each patient. Through the physiological parameter abnormality degree, patients with higher abnormalities can be screened into the subsequent key monitoring stage.

[0034] Therefore, in the embodiment of the present invention, when the value after the normalization processing of the physiological parameter abnormality degree is greater than or equal to the preset attention threshold, the patient is a high-risk patient with an infection situation, and the corresponding patient is recorded as a key monitoring patient. The preset attention threshold is set to 0.8, and the specific value can be adjusted by the implementer. It should be noted that normalization is a well-known technical means for those skilled in the art. The choice of normalization can be linear normalization or standard normalization, etc. The specific normalization method is not limited herein.

[0035] The infection risk assessment module 103 is used to obtain the biomarker level index for each key monitoring patient through the deviation level of the biomarker data at the initial key sampling moment and the growth level at the report generation moment; after predicting the biomarker data based on the change trend of the continuous biomarker data before the report generation moment, analyze the growth change trend of all biomarker data in time series, and combine the biomarker level index to obtain the wound infection risk coefficient of each key monitoring patient at the report generation moment.

[0036] In the key monitoring stage, the patient has a greater possibility of infection. In order to further evaluate and judge the infection risk of the patient, the data acquisition module collects the changes of biomarkers related to infection in the patient's body, which is the change of C-reactive protein concentration in the embodiment of the present invention. Predict according to the real-time trend of the biomarker level to accurately determine the infection situation.

[0037] Although C-reactive protein increases rapidly when there is inflammation or infection in the body, an increase in the biomarker in a patient does not necessarily indicate that the patient has a wound infection. Biomarkers are also sensitive to the body's inflammatory response, which is the body's immune response to any type of injury, irritation, or harmful factor. It is a protective response of tissues in the body to injury, such as physical trauma, chemical irritation, or abnormal immune responses. Therefore, there may be a persistent inflammatory response after surgery that affects the level of biomarkers. Please refer to Figure 4 , which shows a schematic diagram of the change in the level of a biomarker provided by an embodiment of the present invention. The x position represents the time of report generation, and the z position represents the time of final key sampling.

[0038] Generally, if a patient's wound is infected, the initial level value of the corresponding biomarker data is relatively high and shows a trend of significant increase. In contrast, the biomarker corresponding to the inflammatory response gradually increases based on the standard level. Therefore, it is possible to determine whether a patient is at risk of wound infection based on the trend of the biomarker.

[0039] At the time of report generation, which is also the current time and can be the time when a key monitored patient needs to have a report evaluated. Considering that the biomarker level of wound infection in the initial postoperative period is more intense compared to the inflammatory response, the greater the level value, the biomarker level index of the infection symptom may be obtained by analyzing the deviation level at the initial key sampling time and the growth level at the time of report generation.

[0040] Preferably, in the embodiment of the present invention, the method for obtaining the biomarker level index includes: For each key monitored patient, the difference between the biomarker data of the patient at the first key sampling time and the normal biomarker data is used as the initial deviation index. The larger the initial deviation index, the higher the initial sudden increase level, and the greater the possibility that the corresponding patient has an infection symptom. It should be noted that the normal biomarker data is set according to the specific implementation scenario. For example, in this embodiment, the normal biomarker data for the C-reactive protein concentration is set to 10 mg / L.

[0041] The difference between the biomarker data of the patient between the time of report generation and the first key sampling time is used as the growth amplitude index. The larger the growth amplitude index, the stronger the growth trend of the biomarker, and the higher the possibility that the corresponding patient has an infection symptom.

[0042] Finally, by combining the initial deviation index and the growth rate index of the patient, the biomarker level index of the patient at the report generation time is obtained. In the embodiment of the present invention, the product of the initial deviation index and the growth rate index of the patient is used as the biomarker level index of the patient at the report generation time. The larger the biomarker level index, the more significant the infection symptoms. As an example, the method for obtaining the biomarker level index includes: ; where is denoted as the biomarker level index, is denoted as the biomarker data at the first key sampling time, is denoted as the normal biomarker data, is denoted as the report generation time of the biomarker data. is denoted as the initial deviation index, is denoted as the growth rate index.

[0043] During the stage of key monitoring of patients, the change of biomarker data at multiple consecutive key sampling times can reflect the progress of infection. However, in order to give an early warning of the patient's infection situation, by predicting the data change trend and combining the biomarker level index, the pre-evaluation and analysis of the wound infection risk are realized.

[0044] Preferably, in the embodiment of the present invention, for the method of obtaining the wound infection risk coefficient, please refer to Figure 5 , which shows a flowchart of a method for obtaining the wound infection risk coefficient provided by an embodiment of the present invention. The method includes the following steps: S301: According to the degree of growth change of the consecutive biomarker data before the report generation time, predict each unsampled key sampling time after the report generation time to obtain the predicted biomarker data.

[0045] By the average change trend of the historical key sampling times, predict and analyze each future sampling time. In the embodiment of the present invention, the report generation time and all unsampled key sampling times after the time are used as the analysis times. Since the sampling frequency of the key sampling times is determined, data prediction is performed on all unsampled times between the report generation time and the final key sampling time.

[0046] Traverse each analysis time according to the time sequence, and predict the biomarker data in sequence according to the time sequence. Before the analysis time, after calculating the difference between the biomarker data of the latter key sampling time and the former key sampling time in each adjacent two key sampling times, find the average value of the differences between all adjacent key sampling times as the average difference at the analysis time, which characterizes the change trend of the real-time monitored biomarker.

[0047] Furthermore, the biomarker data at the analysis moment and the sum value of the average differences are used as the predicted biomarker data at the next analysis moment. For the analysis moments analyzed in sequence, the predicted biomarker data is the biomarker data at the analysis moment. Each analysis moment is predicted and adjusted based on the previous change level to make the predicted data change trend more accurate.

[0048] S302: Based on all the biomarker data and the predicted biomarker data in time series, obtain the growth trend index according to the data growth deviation between every two adjacent key sampling moments.

[0049] Since there are differences in the growth amplitudes of biomarker levels under infection symptoms and inflammatory responses, the overall growth amplitude of its monitoring process can be estimated by combining the historical and predicted values of biomarker levels, reflecting the significance of the appearance of infection symptoms.

[0050] In the embodiment of the present invention, the biomarker data and the predicted biomarker data at all key sampling moments with corresponding data are recorded as target data. The actually collected and predicted biomarker data before the final key sampling moment in the entire monitoring period required by the patient are recorded as target data, which is convenient for subsequent unified analysis.

[0051] Furthermore, after calculating the difference between the target data of the latter key sampling moment and the former key sampling moment between every two adjacent key sampling moments, the average value of the differences between the target data between all adjacent key sampling moments is obtained as the growth trend index, which characterizes the growth amplitude of the estimated biomarker level. The larger its value, the greater the growth trend of the patient's biomarker level, and the greater the risk of the patient's wound infection.

[0052] S303: Combine the growth trend index and the biomarker level index to obtain the wound infection risk coefficient of each key monitored patient at the report generation moment.

[0053] At the same time, the larger the biomarker level index of the patient, the greater the risk of wound infection. Therefore, in the embodiment of the present invention, the product of the growth trend index and the biomarker level index is used as the wound infection risk coefficient of each key monitored patient at the report generation moment. The larger the wound infection risk coefficient, the more significant the appearance of infection symptoms.

[0054] The early warning report generation module 104 is used to combine the wound infection risk coefficient and the physiological parameter abnormality degree of each key monitored patient to obtain the early warning level of each key monitored patient; obtain the risk report of the key monitored patient at the report generation moment.

[0055] Due to the different infection situations and physical qualities of different patients, it is necessary to determine the corresponding infection warning levels for different patients, so as to intervene in the treatment time of patients with higher levels as early as possible. Based on the data characteristics of the patient's biomarkers, the corresponding risk degree of wound infection and the initial infection risk degree of the abnormal physiological parameter characterization after the patient's operation are obtained, and the warning levels of the patients are divided.

[0056] Preferably, in the embodiment of the present invention, the method for obtaining the warning level includes: First, normalize the product of the wound infection risk coefficient and the degree of abnormal physiological parameters of each key monitored patient to obtain the classification index of each key monitored patient. Both reflect the patient's infection risk in different aspects, so the warning level of this patient is determined by combining the two aspects.

[0057] In the embodiment of the present invention, the preset low threshold is set to 0.4, and the preset medium threshold is set to 0.8. The specific values can be adjusted by the implementer. When the classification index is less than the preset low threshold, it indicates that the severity of the infection risk is relatively low, and the warning level of the corresponding key monitored patient is recorded as the low warning level. When the classification index is greater than or equal to the preset low threshold and less than the preset medium threshold, it indicates that there is a certain degree of infection risk, and the warning level of the corresponding key monitored patient is recorded as the medium warning level. When the classification index is greater than or equal to the preset medium threshold, it indicates that the infection risk is relatively serious, and the warning level of the corresponding key monitored patient is recorded as the high warning level.

[0058] When the warning level is higher, the intervention should be carried out earlier. Based on the warning levels of different patients and combined with basic physiological indicators, a personalized risk report for the patient is generated. In the embodiment of the present invention, to ensure the integrity of the important information in the patient risk report, the risk report at least includes the patient's basic information, warning level, historical physiological characteristic data, and wound infection risk coefficient, etc., providing more detailed infection risk report information for medical staff. For example, the report of a certain patient may include: Patient's basic information: name, gender, age, hospital admission number or medical record number, report generation time, etc.; General situation of infection risk: warning level of infection risk, wound infection risk coefficient, etc.; Details of physiological assessment: historical physiological characteristic data, predicted physiological characteristic values, etc.

[0059] Based on the data analysis basis provided by the patient's real-time risk report, the nursing intervention suggestions of medical staff are made more reliable, realizing the transformation from extensive monitoring to precise management, and ultimately achieving the core goals of reducing the infection missed diagnosis rate, optimizing the allocation of medical resources, and improving the postoperative rehabilitation effect.

[0060] It should be noted that, for the convenience of calculation, all the index data involved in the operations in the embodiments of the present invention have undergone data preprocessing, thereby eliminating the influence of dimensions. The specific means of eliminating the influence of dimensions are well-known technical means to those skilled in the art and will not be limited herein.

[0061] In summary, the present invention screens key monitoring objects based on the growth degree of the continuous abnormal manifestations of the initial physiological signs of orthopedic trauma patients, accurately identifies high-risk patient groups, and reduces the consumption of unnecessary high-cost biomarker detection resources. By using the deviation level of the biomarker at the initial key sampling moment and the growth degree at the report generation moment, a biomarker level index is constructed. Combining the prediction situation of historical biomarker data, the possible overall dynamic change characteristics in time series are analyzed to obtain the wound infection risk coefficient. By fusing the sudden increase characteristics of the biomarker level and the continuous growth amplitude, the normal inflammatory reaction after surgery and the infectious inflammation are distinguished, significantly improving the specificity of infection determination. The patient warning level is divided by combining the physiological parameter abnormality degree and the wound infection risk coefficient, and a customized risk report is generated based on the individual characteristics of the patient, providing a quantitative basis for clinical intervention. Through the collaborative analysis of the continuous abnormal physiological signs, the sudden increase of biomarkers and the prediction trend, the present invention improves the accuracy of the assessment of orthopedic trauma postoperative infection, makes the subsequent warning level division and risk report generation more reliable, and helps to provide a more personalized and accurate treatment plan for patients.

[0062] It should be noted that the above-mentioned sequence of the embodiments of the present invention is only for description and does not represent the advantages and disadvantages of the embodiments. The processes depicted in the drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0063] Each embodiment in this specification is described in a progressive manner. The same or similar parts among the embodiments can be referred to each other, and the differences between each embodiment and other embodiments are emphasized.

Claims

1. An intelligent early warning system for the risk of postoperative infection in orthopedic trauma, characterized in that, The system includes: A data acquisition module, which is used to obtain physiological sign data at each sampling moment during the initial sampling period and biomarker data at each key sampling moment after the initial sampling period for each orthopedic trauma patient after surgery; the physiological signs include body temperature, heart rate, and blood oxygen saturation, and the biomarker data is the C-reactive protein concentration; A key monitoring and screening module, which is used to obtain the physiological parameter abnormality degree of each patient according to the continuous abnormal characterization growth degree of the physiological sign data of each patient in the initial sampling period in time series; determine the key monitoring patients based on the magnitude of the physiological parameter abnormality degree; An infection risk assessment module, which is used for each key monitoring patient to obtain a biomarker level index through the deviation level of the biomarker data at the initial key sampling moment and the growth level at the report generation moment; after predicting the biomarker data based on the change trend of the continuous biomarker data before the report generation moment, analyze the growth change trend of all biomarker data in time series, and combine the biomarker level index to obtain the wound infection risk coefficient of each key monitoring patient at the report generation moment; An early warning report generation module, which is used to combine the wound infection risk coefficient and the physiological parameter abnormality degree of each key monitoring patient to obtain the early warning level of each key monitoring patient; obtain the risk report of the key monitoring patients at the report generation moment.

2. The intelligent early warning system for the risk of postoperative infection in orthopedic trauma according to claim 1, wherein The method for obtaining the physiological parameter abnormality degree includes: For each patient, obtain the physiological characteristic factor at each sampling moment by taking the ratio of the product of the body temperature data and the heart rate data to the blood oxygen saturation data at each sampling moment; Arrange the physiological characteristic factors in chronological order to obtain a physiological characteristic sequence; based on the continuous growth distribution of the physiological characteristic factors in the physiological characteristic sequence, screen out the abnormal growth sequence; Take the ratio of the total number of physiological characteristic factors in the abnormal growth sequence to the total number of physiological characteristic factors in the physiological characteristic sequence as the abnormal proportion; perform a negative correlation mapping on the interval period between the sampling moment corresponding to the first physiological characteristic factor in the abnormal growth sequence and the first sampling moment in the initial sampling period to obtain the time difference abnormality degree; Combine the abnormal proportion and the time difference abnormality degree to obtain the physiological parameter abnormality degree of each patient.

3. The intelligent early warning system for the risk of orthopedic trauma postoperative infection according to claim 2, wherein, The method for obtaining the abnormal growth sequence includes: Obtain the difference sequence of the physiological characteristic sequence; record the difference values greater than zero in the difference sequence as growth difference values; Take each continuously adjacent growth difference value as a positive number sequence; take the positive number sequence containing the most growth difference values as the target sequence; Take the sequence composed of the physiological characteristic factors in the physiological characteristic sequence corresponding to each growth difference value in the target sequence as the abnormal growth sequence.

4. The intelligent early warning system for the risk of postoperative infection in orthopedic trauma according to claim 1, wherein Determining the key monitoring patients based on the magnitude of the physiological parameter abnormality degree includes: When the value after normalizing the physiological parameter abnormality degree is greater than or equal to the preset attention threshold, mark the corresponding patient as a key monitoring patient.

5. The intelligent early warning system for the risk of orthopedic trauma postoperative infection according to claim 1, wherein The acquisition level of the biomarker level index includes: For each key monitored patient, the difference between the biomarker data of the patient at the first key sampling moment and the normal biomarker data is used as the initial deviation index; the difference between the biomarker data of the patient between the report generation moment and the first key sampling moment is used as the growth rate index; Combining the initial deviation index and the growth rate index of the patient, the biomarker level index of the patient at the report generation moment is obtained.

6. The intelligent early warning system for the risk of postoperative infection in orthopedic trauma according to claim 1, characterized in that, The method for obtaining the wound infection risk coefficient includes: According to the degree of growth change of consecutive biomarker data before the report generation moment, each unsampled key sampling moment after the report generation moment is predicted to obtain predicted biomarker data; Based on all biomarker data and predicted biomarker data in time series, according to the data growth deviation between each adjacent key sampling moment, a growth trend index is obtained; Combining the growth trend index and the biomarker level index, the wound infection risk coefficient of each key monitored patient at the report generation moment is obtained.

7. The intelligent early warning system for the risk of postoperative infection in orthopedic trauma according to claim 6, wherein The method for obtaining the predicted biomarker data includes: Taking the report generation moment and all unsampled key sampling moments after the moment as analysis moments; traversing each analysis moment in chronological order, before the analysis moment, calculate the difference between the biomarker data of the latter key sampling moment and the former key sampling moment among every two adjacent key sampling moments, and then calculate the average value of the differences between all adjacent key sampling moments as the average difference of the analysis moment; Taking the sum of the biomarker data of the analysis moment and the average difference as the predicted biomarker data of the next analysis moment.

8. An intelligent early warning system for the risk of postoperative infection in orthopedic trauma according to claim 6, characterized in that, The method for obtaining the growth trend index includes: Recording the biomarker data and predicted biomarker data of all key sampling moments with corresponding data as target data; After calculating the difference between the biomarker data of the latter key sampling moment and the former key sampling moment among every two adjacent key sampling moments in the target data, calculate the average value of the differences between the target data among all adjacent key sampling moments as the growth trend index.

9. The intelligent early warning system for the risk of postoperative infection in orthopedic trauma according to claim 1, characterized in that, The method for obtaining the warning level includes: Normalizing the product of the wound infection risk coefficient and the physiological parameter abnormality degree of each key monitored patient to obtain the classification index of each key monitored patient; When the classification index is less than the preset low threshold, the warning level of the corresponding key monitored patient is recorded as the low warning level; When the classification index is greater than or equal to the preset low threshold and less than the preset medium threshold, the warning level of the corresponding key monitored patient is recorded as the medium warning level; When the classification index is greater than or equal to the preset medium threshold, the warning level of the corresponding key monitored patient is recorded as the high warning level.

10. The intelligent early warning system for the risk of postoperative infection in orthopedic trauma according to claim 1, wherein The risk report includes: patient basic information, warning level, historical physiological characteristic data, and wound infection risk coefficient.

Images