Fracture fixation effect evaluation method and system

Abstract

The invention discloses a fracture fixation effect evaluation method and system, and relates to the technical field of biomedicine, and the method comprises the steps: obtaining the fracture state parameters of a target patient, carrying out the fracture type evaluation, and building a to-be-selected fixation mode set of the fracture type of the target patient; based on the physiological parameters and occupational function requirements of the target patient as decision influence factors, carrying out secondary screening on the to-be-selected fixing mode set of the fracture type of the target patient to obtain an optional fixing mode set of the fracture type of the target patient; and verifying the infection recovery risk probability of each selectable fixing mode in the selectable fixing mode set of the fracture type of the target patient, and taking the selectable fixing mode with the minimum infection recovery risk probability in the selectable fixing mode set of the fracture type of the target patient as a fixing mode scheme of the fracture type of the target patient. The method has the advantages that a personalized fracture treatment scheme is realized, the treatment effect is improved, and the risk of complications is reduced.

Description

Technical Field

[0001] The present invention relates to the field of biomedical technology, and in particular to a method and system for evaluating fracture fixation effects. Background Art

[0002] Fracture fixation effectiveness evaluation refers to the comprehensive judgment of fracture healing, the adaptability and effectiveness of fixation methods after fracture treatment through multiple indicators such as imaging examinations, functional recovery, complication monitoring, and patients' subjective feelings, to ensure that the patient's fracture is effectively treated and complications are minimized to promote functional recovery.

[0003] Existing fracture fixation effectiveness assessments often rely on universal standards and methods, without fully considering the individual differences and specific needs of patients. Therefore, they may not be able to accurately assess the actual recovery of different patients, resulting in poor functional recovery of the fracture site in the target patients, which in turn affects the effectiveness of functional recovery and increases the risk of infection, causing some patients to fail to obtain the best treatment effect. Summary of the Invention

[0004] In order to solve the above technical problems, a method and system for evaluating the effect of fracture fixation are provided. This technical solution solves the problem that the existing fracture fixation effect evaluation often relies on general standards and methods, and does not fully consider the individual differences and specific needs of patients. Therefore, it may not be possible to accurately evaluate the actual recovery of different patients, resulting in poor functional recovery of the fracture position of the target patient, which in turn affects the effect of functional recovery and increases the risk of infection, causing some patients to fail to obtain the best treatment effect.

[0005] In order to achieve the above objects, the technical solution adopted by the present invention is: A method for evaluating fracture fixation effect, comprising: Based on multi-source sensors, the fracture status parameters of the target patient are obtained to assess the fracture type, and the fixation method that meets the target patient's fracture type is selected to form a set of candidate fixation methods for the target patient's fracture type. Based on the target patient's physiological parameters and occupational functional requirements as decision-making factors, the target patient's fracture fixation adaptability preference is evaluated, and a secondary screening is performed on the set of candidate fixation methods for the target patient's fracture type to obtain a set of optional fixation methods for the target patient's fracture type; Verify the recovery infection risk probability of each optional fixation method in the set of optional fixation methods for the target patient's fracture type, and use the optional fixation method that minimizes the recovery infection risk probability in the set of optional fixation methods for the target patient's fracture type as the fixation method plan for the target patient's fracture type.

[0006] Furthermore, based on multi-source sensors, the fracture status parameters of the target patient are obtained; According to the fracture status parameters of the target patient, align the sampling rate for the fracture status parameters according to the data acquisition timestamp to obtain the fracture status time-aligned parameter data of the target patient; Preprocess the fracture status time-aligned parameter data of the target patient, extract the multi-dimensional feature quantization values of the fracture status of the target patient, and form the fracture status feature vector of the target patient; Use PCA principal component analysis to reduce the dimension of the data for the fracture status feature vector of the target patient, and form the fracture status reduced-dimension feature vector matrix of the target patient.

[0007] Furthermore, according to the fixed method database of known fracture types, establish a knowledge graph of optional fixed methods for fracture types; Based on the decision tree, establish classification decision trees for each fracture type. Using the knowledge graph of optional fixed methods for fracture types as the root node, each fracture type as a branch node, and the fixed method of each fracture type as a branch leaf node, establish the fracture status classification decision tree of the target patient and form a weighted random forest; According to the D-S evidence theory, calculate the confidence of each element in the fracture status reduced-dimension feature vector matrix of the target patient for the fracture status classification decision tree of the target patient, and assign confidence weights to each fracture status classification decision tree of the target patient; Based on the fracture status classification model of the target patient, take the fracture status reduced-dimension feature vector matrix and the weighted matrix of the fracture status reduced-dimension feature vector of the target patient as inputs, calculate the information gain of the fracture status reduced-dimension feature vector for each fracture type as a branch node, and output the set of alternative fixed methods for the fracture type of the target patient.

[0008] Furthermore, based on the physiological parameters and occupational function requirements of the target patient, use linear mapping to transform the characteristic parameters of the influencing factors for fracture fixation method selection to obtain the personalized preference feature vector of the target patient; According to the AHP analytic hierarchy process, establish a judgment matrix of the personalized preference feature vector of the target patient based on the personalized preference feature vector of the target patient; Perform normalization processing on the judgment matrix of the personalized preference feature vector of the target patient to obtain the normalized judgment matrix of the personalized preference feature vector of the target patient; According to the average value of each row in the normalized judgment matrix of the personalized preference feature vector of the target patient, assign weights to the corresponding personalized preference feature vectors to obtain the weights of the personalized preference feature vectors of the target patient.

[0009] Furthermore, based on the set of alternative fixed methods for the fracture type of the target patient, verify the treatment multi-dimensional vectors of each alternative fixed method to obtain the multi-dimensional treatment feature vectors of each alternative fixed method; According to the weighted cosine similarity formula, calculate the adaptability scores between the personalized preference feature vector of the target patient, the personalized preference feature weight vector of the target patient, and the multi-dimensional treatment feature vectors of each candidate fixation method.

[0010] According to the maximum value of the adaptability scores between the multi-dimensional treatment feature vectors of the candidate fixation methods, screen the candidate fixation methods that meet the personalized preferences of the target patient from the set of candidate fixation methods for the fracture type of the target patient, and obtain the set of optional fixation methods for the fracture type of the target patient.

[0011] Furthermore, obtain the clinical big data of the fixation methods for the fracture types of historical target patients, and obtain the set of infection risk data for the fixation methods of historical fracture types; Preprocess the set of infection risk data for the fixation methods of historical fracture types; Mark the factors affecting the infection risk in the set of infection risk data for the fixation methods of historical fracture types as infection risk feature data, and use the infection events corresponding to the infection risks as infection event label data; Based on the infection risk feature data and infection event label data in the set of infection risk data for the fixation methods of historical fracture types, establish an infection risk sample for the fixation methods of historical target patient fracture types.

[0012] Furthermore, based on the clinical big data of the fixation methods for the fracture types of historical target patients, quantify the basic infection risk of the historical target patient fracture types and the basic infection risk of the fixation methods; Based on the infection risk samples of the fixation methods for the fracture types of historical target patients, train the Logistic regression. Use the basic infection risk of the historical target patient fracture types as the inherent risk layer, the basic infection risk of the fixation methods for the historical target patient fracture types as the device risk layer, the non-linear synergistic effect between the basic infection risk of the historical target patient fracture types and the basic infection risk of the fixation methods as the interaction risk layer, and use minimizing the error function as the training end goal to obtain an infection risk assessment model for the fixation methods of fracture types; Based on the infection risk assessment model for the fixation methods of fracture types, use the set of optional fixation methods for the fracture type of the target patient as the input, and use the probability of recovery infection risk of each optional fixation method as the output.

[0013] Furthermore, a system for evaluating the fracture fixation effect includes: A fracture fixation candidate module, a fracture fixation optional module, and a fracture fixation plan generation module; The fracture fixation candidate module is used to obtain the fracture status parameters of the target patient based on multi-source sensors, evaluate the fracture type, screen the fixation methods that meet the fracture type of the target patient, and form a set of candidate fixation methods for the fracture type of the target patient; The fracture fixation optional module is electrically connected to the fracture fixation candidate module. The fracture fixation optional module is used to evaluate the fracture fixation adaptability preference of the target patient based on the physiological parameters and occupational function requirements of the target patient, and perform a secondary screening on the set of candidate fixation methods for the fracture type of the target patient to obtain a set of optional fixation methods for the fracture type of the target patient; The fracture fixation scheme generation module is electrically connected to the fracture fixation optional module. The fracture fixation scheme generation module is used to verify the probability of recovery and infection risk of each optional fixation method in the set of optional fixation methods for the fracture type of the target patient, and use the optional fixation method that minimizes the probability of recovery and infection risk in the set of optional fixation methods for the fracture type of the target patient as the fixation method scheme for the fracture type of the target patient.

[0014] Compared with the prior art, the beneficial effects of the present invention are as follows: The present invention proposes a method for evaluating the fracture fixation effect. The fracture status parameters of the target patient are obtained through multi-source sensors, and the adaptability of the fracture fixation method is evaluated in combination with the physiological characteristics and occupational needs of the patient. First, a preliminary set of fixation methods is screened based on the fracture type, and then a secondary screening is performed through the individual needs of the patient. Finally, a set of fixation methods suitable for the patient is generated. By verifying the probability of recovery and infection risk of each fixation method, the fixation method with the minimum infection risk is selected as the final scheme, so as to realize a personalized and accurate fracture treatment plan, improve the treatment effect and reduce the risk of complications. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] Figure 1 is a flowchart of a method for evaluating the fracture fixation effect; Figure 2 is a framework diagram of a system for evaluating the fracture fixation effect; DETAILED DESCRIPTION OF THE INVENTION

[0016] The following description is used to disclose the present invention so that those skilled in the art can implement the present invention. The preferred embodiments described below are only examples, and those skilled in the art can think of other obvious variations.

[0017] Refer to Figure 1 As shown, a method for evaluating the fracture fixation effect includes: Step 1: Based on multi-source sensors, obtain the fracture status parameters of the target patient to evaluate the fracture type, screen the fixation methods that meet the fracture type of the target patient, and form a set of candidate fixation methods for the fracture type of the target patient; The first step includes the following contents: Step 101: Based on multi-source sensors, obtain the fracture status parameters of the target patient; As a further content, the multi-source sensors include: mechanical sensors, bioimpedance sensors, and OTC sensors; and the corresponding fracture status parameters include: stress distribution / micro-motion displacement parameters of the fracture site, conductivity change parameters in the callus formation stage, and three-dimensional images of the microscopic structure of bone fractures; According to the fracture status parameters of the target patient, align the sampling rate for the fracture status parameters according to the data acquisition timestamp to obtain the fracture status time-aligned parameter data of the target patient; Preprocess the fracture status time-aligned parameter data of the target patient, extract the multi-dimensional feature quantization values of the fracture status of the target patient, and form the fracture status feature vector of the target patient; As a further content, extracting the multi-dimensional feature data of the fracture status of the target patient includes, but is not limited to, integrating the image gradient magnitude based on the three-dimensional image of the microscopic structure of the bone fracture to quantify the degree of crack extension, and obtaining the bone fracture length of the fracture area; calculating the dynamic stress entropy of the fracture area according to the stress distribution / micro-motion displacement parameters of the measured fracture site; Use PCA principal component analysis to reduce the dimension of the data for the fracture status feature vector of the target patient, and form the fracture status reduced-dimension feature vector matrix of the target patient , where is the i th fracture status reduced-dimension feature vector of the target patient, is the total number of fracture status reduced-dimension feature vectors; Step 102: According to the fixed method database of known fracture types, establish a knowledge graph of optional fixed methods for fracture types; Based on the decision tree, establish classification decision trees for each fracture type. Using the knowledge graph of optional fixed methods for fracture types as the root node, each fracture type as a branch node, and the fixed method of each fracture type as a branch leaf node, establish a fracture status classification decision tree for the target patient, and form a weighted random forest; According to the D-S evidence theory, calculate the confidence of each element in the fracture status reduced-dimension feature vector matrix of the target patient for the fracture status classification decision tree of the target patient, and assign confidence weights to each fracture status classification decision tree of the target patient, in the following way:

[0018] where is the confidence weight of the k th fracture status classification decision tree of the target patient, is thei The support degree value of the fracture status dimensionality reduction feature vector for the fracture status classification decision tree of the th target patient, is the i th fracture status dimensionality reduction feature vector of the target patient for the th target patient's fracture status classification decision tree's possibility degree value, is the belief function, is the possibility function; Based on the fracture status classification model of the target patient, taking the fracture status dimensionality reduction feature vector matrix and the fracture status dimensionality reduction feature vector weighted matrix of the target patient as inputs, calculate the information gain of the fracture status dimensionality reduction feature vector for each fracture type as a branch node, and take the set of candidate fixation methods for the target patient's fracture type as the output, as follows:

[0019] Among them, is the set of candidate fixation methods for the target patient's fracture type, is the candidate fixation method for classification to be determined, is the set of candidate fixation methods for classification to be determined, is the fracture classification of the fracture status classification decision tree of the kth target patient under the fracture status dimensionality reduction feature vector matrix of the given target patient , is the indicator function (takes 1 when matching the fracture type , otherwise 0), is the deviation degree of the candidate fixation method z for classification to be determined from the standard treatment path of the fracture classification , is the matching credibility between the candidate fixation method for classification to be determined and the cth fracture classification under the fracture status dimensionality reduction feature vector matrix of the given target patient; When in use, combine the content in 101 to 102: As further content, use multi-sensor data fusion and PCA dimensionality reduction to achieve optimized characterization of fracture features, combine knowledge graphs to construct a fracture type-fixation method association model, perform intelligent discrimination of fracture types with a weighted random forest classifier (dynamically adjust the confidence weights of decision trees through D-S evidence theory), and finally screen the optimal set of fixation methods based on information gain. By integrating the advantages of multi-modal data standardization processing, feature engineering noise reduction, knowledge-driven decision-making, and machine learning classification, it can effectively improve the accuracy of fracture typing, shorten the decision-making time of fixation plans, reduce the risk of clinical misjudgment, and achieve precise decision-making support for personalized orthopedic treatment.

[0020] Step 2: Based on the physiological parameters and occupational function requirements of the target patient as decision-making influencing factors, evaluate the fracture fixation adaptability preference of the target patient, and conduct a secondary screening on the set of alternative fixation methods for the fracture type of the target patient to obtain the set of optional fixation methods for the fracture type of the target patient; The second step includes the following contents: Step 201: Based on the physiological parameters and occupational function requirements of the target patient, use linear mapping to transform the characteristic parameters of the influencing factors for fracture fixation method selection to obtain the personalized preference feature vector of the target patient; As a further content, by obtaining the physiological parameters and occupational function requirements of the target patient, such as bone metabolism indicators, soft tissue status, occupational types (heavy physical labor, sedentary office work, etc.), establishing the personalized preference feature data of the target patient is to ensure the recovery speed of the target patient and at the same time ensure that the occupational function requirements of the target patient are met; However, the existing conventional fracture fixation methods mainly limit the fracture position by applying plaster to make it heal on its own, and cannot take into account the individual recovery attributes and occupational needs of the population. For example, the physiological parameters of the target patient can meet the self-healing recovery by applying plaster fixation, but the corresponding occupation is heavy physical labor, resulting in the subsequent functional recovery of the fracture position not reaching the original level, reducing the quality of life; According to the AHP (Analytic Hierarchy Process), based on the personalized preference feature vector of the target patient, establish the personalized preference feature vector judgment matrix of the target patient; Perform normalization processing on the personalized preference feature vector judgment matrix of the target patient to obtain the normalized judgment matrix of the personalized preference feature vector of the target patient; According to the average value of each row in the normalized judgment matrix of the personalized preference feature vector of the target patient, assign weights to the corresponding personalized preference feature vectors to obtain the weights of the personalized preference feature vectors of the target patient; Step 202: Based on the set of alternative fixation methods for the fracture type of the target patient, verify the treatment multi-dimensional vectors of each alternative fixation method to obtain the multi-dimensional treatment feature vectors of each alternative fixation method; According to the weighted cosine similarity formula, calculate the adaptability scores between the personalized preference feature vector of the target patient, the personalized preference feature weight vector of the target patient and the multi-dimensional treatment feature vectors of each alternative fixation method, as follows:

[0021] where, is the personalized preference feature vector of the target patient and the treatment feature vector of the alternative fixation method the adaptability score between them, is the The weight of a feature vector, is the value of the th feature vector of the th multi-dimensional treatment of the candidate fixation methods, and

[0022] where is the set of candidate fixation methods for the fracture type of the target patient; When in use, Step 3: Verify the probability of recovery infection risk for each candidate fixation method in the set of candidate fixation methods for the fracture type of the target patient, and use the candidate fixation method that minimizes the probability of recovery infection risk in the set of candidate fixation methods for the fracture type of the target patient as the fixation method scheme for the fracture type of the target patient; The Based on the infection risk samples of the fixation methods for the fracture types of historical target patients, train a Logistic regression. Use the basic infection risk of the fracture types of historical target patients as the inherent risk layer, the basic infection risk of the fixation methods for the fracture types of historical target patients as the device risk layer, the non-linear synergy between the basic infection risk of the fracture types of historical target patients and the basic infection risk of the fixation methods as the interaction risk layer, and minimize the error function as the training end goal to obtain an infection risk assessment model for the fixation methods of fracture types; Based on the infection risk assessment model for the fixation methods of fracture types, use the set of optional fixation methods for the fracture type of the target patient as the input, and use the probability of recovery infection risk for each optional fixation method as the output, as follows:

[0023] Among them, is the probability of recovery infection risk for each optional fixation method under the set of optional fixation methods for the fracture type of the given target patient, is the inherent risk layer of the basic infection risk of the fracture types of historical target patients, is the device risk layer of the basic infection risk of the fixation methods for the fracture types of historical target patients, is the interaction risk layer of the non-linear synergy between the basic infection risk of the fracture types of historical target patients and the basic infection risk of the fixation methods, is the intercept term, is the weight of the inherent risk layer, is the weight of the device risk, is the weight of the interaction term.

[0024] When in use, combine the content in 301 to 302: As further content, by integrating the clinical big data of historical fracture cases, first quantify the inherent basic infection risk of the patient's fracture type (reflecting the patient's physiological characteristics) and the device basic risk of the fixation method (reflecting the characteristics of the treatment method), and then construct a Logistic regression model containing three risk layers (inherent risk layer, device risk layer, interaction risk layer). The model dynamically evaluates the infection probability of each fixation method by capturing the non-linear synergy between the patient and the device (such as the risk amplification when a diabetic patient uses an external fixator).

[0025] The beneficial effects are as follows: 1. Objectively reflect the complex relationship between the fixation method and the infection risk through big data-driven modeling; 2. The hierarchical quantification mechanism can distinguish the independent / synergistic effects of patient individual differences and device characteristics on infection; 3. The model outputs specific probability values, providing a quantitative basis for clinical decision-making and helping to minimize the infection risk in the selection of fracture fixation schemes.

[0026] Refer to Figure 2 As shown, a system for evaluating the effect of fracture fixation includes: A fracture fixation candidate module, a fracture fixation optional module, and a fracture fixation scheme generation module; The fracture fixation candidate module is used to obtain the fracture state parameters of the target patient based on multi-source sensors for fracture type assessment, screen the fixation methods that meet the fracture type of the target patient, and form a set of candidate fixation methods for the fracture type of the target patient; The fracture fixation optional module is electrically connected to the fracture fixation candidate module. The fracture fixation optional module is used to evaluate the fracture fixation adaptability preference of the target patient based on the physiological parameters and occupational function requirements of the target patient as decision-making influencing factors, and perform a secondary screening on the set of candidate fixation methods for the fracture type of the target patient to obtain a set of optional fixation methods for the fracture type of the target patient; The fracture fixation scheme generation module is electrically connected to the fracture fixation optional module. The fracture fixation scheme generation module is used to verify the probability of recovery infection risk of each optional fixation method in the set of optional fixation methods for the fracture type of the target patient, and use the optional fixation method that minimizes the probability of recovery infection risk in the set of optional fixation methods for the fracture type of the target patient as the fixation method scheme for the fracture type of the target patient.

[0027] The above shows and describes the basic principles, main features and advantages of the present invention. Those skilled in the art of this industry should understand that the present invention is not limited by the above embodiments. What is described in the above embodiments and the specification is only the principle of the present invention. Without departing from the spirit and scope of the present invention, the present invention will have various changes and improvements, and these changes and improvements fall within the scope of the present invention claimed. The scope of protection claimed by the present invention is defined by the appended claims and their equivalents.

Claims

1. A method for evaluating the effect of fracture fixation, characterized in that, Including: Based on multi-source sensors, obtain the fracture status parameters of the target patient to evaluate the fracture type, screen the fixation methods that meet the fracture type of the target patient, and form a set of candidate fixation methods for the fracture type of the target patient; Based on the physiological parameters and occupational function requirements of the target patient as decision-making influencing factors, evaluate the fracture fixation adaptability preference of the target patient, and conduct a secondary screening for the set of candidate fixation methods for the fracture type of the target patient to obtain a set of optional fixation methods for the fracture type of the target patient; Verify the probability of recovery and infection risk of each optional fixation method in the set of optional fixation methods for the fracture type of the target patient, and use the optional fixation method that minimizes the probability of recovery and infection risk in the set of optional fixation methods for the fracture type of the target patient as the fixation method plan for the fracture type of the target patient.

2. The method for evaluating the fracture fixation effect according to claim 1, wherein: Based on multi-source sensors, obtain the fracture status parameters of the target patient; According to the fracture status parameters of the target patient, align the sampling rate for the fracture status parameters according to the data acquisition timestamp to obtain the fracture status time-aligned parameter data of the target patient; Perform preprocessing on the fracture status time-aligned parameter data of the target patient, extract the multi-dimensional feature quantization values of the fracture status of the target patient, and form the fracture status feature vector of the target patient; Use PCA principal component analysis to perform data dimensionality reduction on the fracture status feature vector of the target patient, and form the fracture status dimensionality reduction feature vector matrix of the target patient.

3. The method for evaluating the fracture fixation effect according to claim 2, wherein: According to the database of fixation methods for known fracture types, establish a knowledge graph of optional fixation methods for fracture types; Based on the decision tree, establish classification decision trees for each fracture type. Using the knowledge graph of optional fixation methods for fracture types as the root node, each fracture type as the branch node, and the fixation method for each fracture type as the branch leaf node, establish a classification decision tree for the fracture status of the target patient, and form a weighted random forest; According to the D-S evidence theory, calculate the confidence of each element in the fracture status dimensionality reduction feature vector matrix of the target patient for the classification decision tree of the fracture status of the target patient, and assign confidence weights to each classification decision tree of the fracture status of the target patient, as follows: ； Among them, is the confidence weight of the fracture status classification decision tree for the k-th target patient, is the support degree value of the i-th dimensionality-reduced feature vector of the fracture status of the target patient for the fracture status classification decision tree of the -th target patient, is the possibility degree value of the i-th dimensionality-reduced feature vector of the fracture status of the target patient for the fracture status classification decision tree of the -th target patient, is the belief function, is the possibility function; Based on the fracture status classification model of the target patient, use the fracture status dimensionality reduction feature vector matrix and the weighted matrix of the fracture status dimensionality reduction feature vector of the target patient as inputs, calculate the information gain of the fracture status dimensionality reduction feature vector for each fracture type as the branch node, and use the set of candidate fixation methods for the fracture type of the target patient as the output, as follows: ； Among them, is a set of candidate fixation methods for the fracture type of the target patient, is a candidate solution for the fixation method to be classified, is a set of candidate solution sets for the fixation method to be classified, is the fracture classification of the k-th fracture status classification decision tree of the target patient under the dimensionality-reduced feature vector matrix of the fracture status of the given target patient , is an indicator function (taking 1 when matching the fracture type and 0 otherwise), is the deviation degree of the candidate solution z of the fixation method to be classified from the standard treatment path of the fracture classification , is the matching credibility between the dimensionality-reduced feature vector matrix of the fracture status of the given target patient and the c-th fracture classification.

4. The method for evaluating the fracture fixation effect according to claim 3, wherein: Based on the physiological parameters and occupational function requirements of the target patient, use linear mapping to transform the characteristic parameters of the influencing factors for the choice of fracture fixation method to obtain the personalized preference feature vector of the target patient; According to the AHP (Analytic Hierarchy Process), based on the personalized preference eigenvector of the target patient, a judgment matrix of the personalized preference eigenvector of the target patient is established; The judgment matrix of the personalized preference eigenvector of the target patient is normalized to obtain the normalized judgment matrix of the personalized preference eigenvector of the target patient; According to the average value of each row in the normalized judgment matrix of the personalized preference eigenvector of the target patient, weights are assigned to the corresponding personalized preference eigenvectors to obtain the weights of the personalized preference eigenvectors of the target patient.

5. The method for evaluating the fracture fixation effect according to claim 4, characterized in that: Based on the set of alternative fixation methods for the fracture type of the target patient, the treatment multi-dimensional vectors of each alternative fixation method are verified to obtain the multi-dimensional treatment characteristic vectors of each alternative fixation method; According to the weighted cosine similarity formula, the fitness scores between the personalized preference eigenvector of the target patient, the personalized preference weight vector of the target patient and the multi-dimensional treatment characteristic vectors of each alternative fixation method are calculated; According to the maximum value of the fitness scores between the multi-dimensional treatment characteristic vectors of the alternative fixation methods, the alternative fixation methods that meet the personalized preference characteristics of the target patient are selected from the set of alternative fixation methods for the fracture type of the target patient to obtain the set of optional fixation methods for the fracture type of the target patient.

6. The method for evaluating the fracture fixation effect according to claim 5, characterized in that: The clinical big data of the fixation methods for the fracture type of the historical target patient is obtained to obtain the set of infection risk data of the fixation methods for the historical fracture type; Preprocessing is performed on the set of infection risk data of the fixation methods for the historical fracture type; The factors affecting the infection risk in the set of infection risk data of the fixation methods for the historical fracture type are marked as infection risk characteristic data, and the infection events corresponding to the infection risks are used as infection event label data; Based on the infection risk characteristic data and the infection event label data in the set of infection risk data of the fixation methods for the historical fracture type, an infection risk sample of the fixation methods for the historical fracture type of the target patient is established.

7. The method for evaluating the fracture fixation effect according to claim 6, characterized in that: Based on the clinical big data of the fixation methods for the historical fracture type of the target patient, the basic infection risk of the historical fracture type of the target patient and the basic infection risk of the fixation method are quantified; Based on the infection risk sample of the fixation methods for the historical fracture type of the target patient, Logistic regression is trained. The basic infection risk of the historical fracture type of the target patient is used as the inherent risk layer, the basic infection risk of the fixation methods for the historical fracture type of the target patient is used as the instrument risk layer, the non-linear synergistic effect between the basic infection risk of the historical fracture type of the target patient and the basic infection risk of the fixation method is used as the interaction risk layer, and minimizing the error function is used as the training end goal to obtain the infection risk assessment model of the fixation methods for the fracture type; Based on the infection risk assessment model of the fixation methods for the fracture type, the set of optional fixation methods for the fracture type of the target patient is used as the input, and the probability of recovery infection risk of each optional fixation method is used as the output.

8. A fracture fixation effect evaluation system, comprising: A fracture fixation candidate module, a fracture fixation optional module, and a fracture fixation plan generation module; The fracture fixation candidate module is configured to obtain fracture status parameters of a target patient based on multi-source sensors for fracture type assessment, screen fixation methods that meet the fracture type of the target patient, and form a set of candidate fixation methods for the fracture type of the target patient; [[ID=Z3]]The fracture fixation optional module is electrically connected to the fracture fixation candidate module. The fracture fixation optional module is configured to evaluate the fracture fixation adaptability preference of the target patient based on the physiological parameters and occupational function requirements of the target patient as decision-making influencing factors, and perform a secondary screening on the set of candidate fixation methods for the fracture type of the target patient to obtain a set of optional fixation methods for the fracture type of the target patient; The fracture fixation plan generation module is electrically connected to the fracture fixation optional module. The fracture fixation plan generation module is configured to verify the probability of recovery infection risk of each optional fixation method in the set of optional fixation methods for the fracture type of the target patient, and use the optional fixation method that minimizes the probability of recovery infection risk in the set of optional fixation methods for the fracture type of the target patient as the fixation method plan for the fracture type of the target patient.

Images