Medical examination index reduction method based on rough set positive domain and information entropy synthesis

Abstract

The invention discloses a medical examination index reduction method based on rough set positive domain and information entropy synthesis, is suitable for high-dimensional data attribute reduction and knowledge discovery in scenes such as medical data analysis, and provides a new reduction strategy by fusing two perspectives of a rough set theory and an information theory method. The comprehensive importance degree of the attributes is obtained through a weighted average mode, and the weight is determined by the ratio of the positive domain to the boundary domain. Iterative reduction is carried out according to the comprehensive importance, deterministic and non-deterministic elements are reasonably considered, the problem of feasibility of attribute reduction of a large-scale fuzzy information decision system is solved, and a more accurate reduction set is obtained. Key diagnosis indexes are conveniently focused, the diagnosis efficiency and accuracy are improved, the clinical interpretability is enhanced, unnecessary detection items are reduced, and the effects of optimizing data, improving the diagnosis efficiency and saving the medical cost are achieved.

Description

Technical Field

[0001] The present invention relates to a method for reducing medical test indicators based on the comprehensive consideration of the positive region of rough sets and information entropy, and belongs to the technical field of medical test indicator data processing. Background Art

[0002] Under the background of the explosive growth of current medical data, a single examination often generates hundreds of dimensional feature parameters, forming a high-dimensional fuzzy information decision system. Attribute reduction plays a role in optimizing data, improving diagnostic efficiency, and saving medical costs by extracting key features.

[0003] Taking the detection of lung CT nodules as an example, among the 52 significant attributes screened in the early stage, using the first 8 supervised attributes, the accuracy, sensitivity, and specificity all reach 100%. Moreover, experiments show that increasing the number of attributes does not significantly improve the performance. In the unsupervised case, increasing the number of attributes may even reduce the classification effect. This indicates that the number of features is not necessarily the more the better, and redundant attributes may introduce noise and reduce the generalization ability of the model. This reduction of data dimensions enables doctors to focus on key diagnostic features such as nodule morphology and density gradient, significantly improving the efficiency of clinical decision-making.

[0004] In an information decision system, in order to solve the problems of the complexity and accuracy of high-dimensional data calculation and eliminate the influence of redundant and irrelevant attributes on the calculation process and the final result, attribute reduction is a necessary step. The commonly used existing attribute reduction methods are as follows: 1. The discernibility matrix method based on rough set theory, which determines the core attributes and gradually deletes redundant attributes. This method has a huge space complexity and is not suitable for the processing of high-dimensional data; 2. The positive region-dependent attribute reduction method combining fuzzy mathematics and rough set theory, which screens attributes by calculating the importance of attributes to the lower approximation of decision classes. This method relies on deterministic elements and ignores non-deterministic elements; 3. The information entropy information gain attribute reduction method based on information theory methods, which screens attributes by calculating the information amount of attributes based on entropy reduction. This method mainly focuses on the information entropy brought by uncertain elements and ignores the deterministic elements in decision-making. Summary of the Invention

[0005] The technical problem to be solved by the present invention is to provide a method for reducing medical test indicators based on the comprehensive consideration of the positive region of rough sets and information entropy, reasonably taking into account both deterministic and non-deterministic elements, solving the feasibility problem of attribute reduction in a large-scale fuzzy information decision system, and obtaining a more accurate reduction set.

[0006] The present invention adopts the following technical solutions to solve the above technical problems: A medical test index reduction method based on the comprehensive consideration of the positive region of rough sets and information entropy is used to reduce all medical test indexes for a certain disease, including the following steps: Step 1: Obtain a medical test index data set. Each sample in the medical test index data set includes a patient, all test index data of the patient for a certain disease, and the diagnosis conclusion of the patient for a certain disease; convert the medical test index data set into a fuzzy information decision system , where represents the set of all patients, represents the th patient, , is the number of all patients, represents the set of all medical test indexes for a certain disease, represents the th index, , is the number of all medical test indexes, represents dividing into disjoint diagnosis conclusion categories, represents the th diagnosis conclusion category, , is the number of diagnosis conclusion categories; Step 2: Calculate the positive region of the set of all medical test indexes by the rough set positive region method, calculate the weighted coefficient and the dependence degree of the diagnosis conclusion category set on ; ; Step 3: Convert the fuzzy information decision system into a deterministic information decision system by the clustering method, and calculate the information entropy of and the conditional information entropy under the full attribute ; Step 4: Set the reduction set as , and initialize the reduction set as an empty set; Step 5: For each index in the set that has not been selected into , combine each index with the current reduction set to obtain the union corresponding to each index, calculate the importance based on the rough set positive region and the importance based on the information entropy of the union corresponding to each index, and then calculate the comprehensive importance of the union corresponding to each index; Step 6: Add the indicator corresponding to the maximum comprehensive importance to the current reduction set to obtain the set , which serves as the new reduction set ; Step 7: Calculate the dependency degree of the diagnostic conclusion category set for the new reduction set obtained in Step 6 using the same method as in Step 2, and calculate the conditional entropy under the new reduction set obtained in Step 6 using the same method as in Step 3 ; ; Step 8: Determine whether it satisfies and . If it satisfies, output the new reduction set obtained in Step 6 as the final indicator reduction set; otherwise, return to Step 5 and continue to select the indicator corresponding to the maximum comprehensive importance from the indicators that have not been selected into and add it to the new reduction set obtained in Step 6 to generate a new reduction set again until it satisfies and and .

[0007] Compared with the prior art, the present invention adopts the above technical solution and has the following technical effects: 1. The present invention combines two perspectives of rough set theory and information theory, proposes a new reduction strategy, obtains the comprehensive importance of detection indicators through weighted average, and its weight is determined by the ratio of the positive domain to the boundary domain. Iterative reduction is performed according to the comprehensive importance, reasonably taking into account both deterministic and non-deterministic elements, solving the feasibility problem of attribute reduction in a large-scale fuzzy information decision-making system, obtaining a more accurate reduction set, and ensuring the accuracy rate of medical detection in the reduction set for the diagnostic conclusion.

[0008] 2. The method proposed by the present invention is convenient for focusing on key diagnostic indicators, improving the diagnostic efficiency and accuracy, enhancing clinical interpretability, reducing unnecessary detection items, and reducing medical costs. BRIEF DESCRIPTION OF THE DRAWINGS

[0009] Figure 1 is the flow chart of the reduction of medical test indicators of the present invention; Figure 2 is the flow chart of determining the attribute importance based on the positive domain method of rough sets of the present invention; Figure 3 is the flow chart of determining the attribute importance by the information entropy method of the present invention; Figure 4It is the flowchart of the iterative reduction of the present invention. Detailed implementation manners

[0010] The following details the implementation manners of the present invention, and examples of the implementation manners are shown in the accompanying drawings. The implementation manners described below with reference to the accompanying drawings are exemplary and are only used to explain the present invention, and should not be construed as limiting the present invention.

[0011] As Figure 1 shown, the present invention proposes a medical test index reduction method based on the comprehensive rough set positive region and information entropy. On the one hand, a distance-based fuzzy equivalence relation is defined in the fuzzy decision system, a lower approximation operator and a fuzzy positive region are defined, and further, the importance degree of the conditional attribute based on the positive region is defined. On the other hand, the fuzzy decision system is binarized and converted into a deterministic information decision system, and the importance degree of the conditional attribute based on the information entropy is defined. The two importance degrees are weighted and averaged to obtain a comprehensive importance degree. Heuristic reduction is performed according to the comprehensive importance degree, and finally an attribute reduction set is obtained.

[0012] Data preparation: Preprocess the medical test index data and diagnosis results and convert them into a fuzzy information decision system. is a fuzzy information decision system, where is the universe of discourse, representing the patients undergoing medical examinations; is a set of finite non-empty conditional attributes, representing all medical test indexes for a certain disease. For each attribute , there is a mapping , which assigns the attribute value (fuzzy value) to the individuals in , representing the test values of relevant medical tests; represents the symbolic decision attribute, representing the diagnosis conclusion. This attribute divides into disjoint decision classes . The membership degree of the individual for each diagnosis conclusion is defined as: .

[0013] Step 1, determine the importance degree of each attribute with respect to the deterministic elements through the rough set positive region method Step 1.1 In the fuzzy decision information system , define a fuzzy equivalence relation measured by distance Step 1.1.1 For each conditional attribute , define the distance function: represents the degree of difference in medical test data between different individuals under the attribute

[0014] Step 1.1.2 For the conditional attribute subset , define the composite distance function: , which means the comprehensive difference degree of medical test data under the combined action of multiple attributes among different individuals, and is determined by the maximum value of the data differences under all its single attributes.

[0015] Step 1.2 Define the decision class Based on the metric-based fuzzy approximation operator Step 1.2.1 Define the lower approximation operator: , The lower approximation operator measures the certainty degree of giving a certain diagnosis conclusion for the individual and is quantified as the minimum difference between the individual and the individuals with non-this diagnosis conclusion.

[0016] Step 1.2.2 Define the upper approximation operator: , The upper approximation operator evaluates the possibility of the individual getting a certain diagnosis conclusion, which is defined by its maximum similarity with the individuals getting the same diagnosis conclusion.

[0017] Step 1.3 Define the fuzzy positive region of the attribute subset : , The fuzzy positive region reflects the diagnostic certainty under part of the medical tests . The larger the positive region, the higher the certainty membership degree of the individual, indicating that the diagnostic certainty of the attribute subset is stronger.

[0018] Step 1.4 Calculate the positive region of the full attribute set .

[0019] Step 1.5 Define the decision variable The dependence degree of the attribute subset : , The dependence degree is the standardization of the fuzzy positive region, and this value is within the interval .

[0020] Step 1.6 Calculate the dependence degree of the decision attribute on the full conditional attribute set .

[0021] Step 1.7 Define the importance degree of the attribute based on the positive region: , Measure the impact of medical tests on the certainty of the diagnosis result.

[0022] Step 2, determine the importance of each attribute with respect to the uncertainty element through the information entropy method Step 2.1 Binarize the fuzzy decision system Step 2.1.1 For each conditional attribute , use the fuzzy c-means (FCM) algorithm to cluster the samples into clusters: , where each individual For each cluster the membership degree is .

[0023] Step 2.1.2 According to the principle of maximum membership degree, convert the fuzzy clustering into a deterministic classification, and each individual is assigned to the classification with the maximum membership degree, that is, for each medical test index , according to the test value of the individual, the individuals are divided into classes.

[0024] Step 2.1.3 Generate a deterministic information decision system : Here, the fuzzy c-means (FCM) algorithm is selected, which can retain the fuzzy membership attributes and facilitate docking with fuzzy-related algorithms. If the fuzziness does not need to be retained, the k-means algorithm can be selected for clustering to directly generate a deterministic information decision system .

[0025] Step 2.2 Define the decision attribute information entropy in : , The information entropy measures the degree of uncertainty of the diagnostic conclusion in a mathematical form.

[0026] Step 2.3 Calculate the information entropy ; Step 2.4 In , define the joint partition of the sample set induced by the attribute subset using the way of partition refinement: , where represents the classification of the sample set under the attribute .

[0027] Step 2.5 Define the attribute subset in Conditional entropy : , Conditional entropy measures the degree of uncertainty of the diagnosis conclusion in medical tests .

[0028] Step 2.6 Calculate the information entropy under the full attributes ; ; Step 2.7 Define the importance degree of attributes based on information entropy: , which measures how much the medical test affects the uncertainty of the diagnosis.

[0029] Step 3, Define the weighted coefficient through the proportional relationship between the positive domain and the boundary domain : , In a given fuzzy decision system, since each individual has a clear diagnosis conclusion , the boundary region quantifies the remaining uncertainty after considering all the clearly classifiable objects in the positive region and is complementary to the positive domain value.

[0030] Step 4, Define the comprehensive importance degree of conditional attributes : , which comprehensively considers deterministic and uncertain decisions and measures the impact of the medical test on the diagnosis.

[0031] Step 5, Perform iterative reduction to obtain the attribute reduction set Step 5.1 Initialize the reduction set ; Step 5.2 For each unselected attribute , calculate the comprehensive importance degree; Step 5.2.1 Calculate the fuzzy positive domain of the conditional attribute set ; Step 5.2.2 Calculate the dependence degree of the decision on the attribute subset ; Step 5.2.3 Calculate the importance degree of the attribute based on the positive domain: , Step 5.2.4 Calculate the partition induced by the conditional attribute set : , Step 5.2.5 Calculate the conditional entropy ; Step 5.2.6 Calculate the importance degree of the attribute based on information entropy: , Step 5.2.7 Calculate the comprehensive importance degree of the conditional attribute : , Step 5.3 Select the current optimal attribute: ; Step 5.4 Update the reduction set ; Step 5.5 Verify the termination condition , and , if satisfied, output the reduction set , otherwise return to Step 5.2.

[0032] Embodiment 1 is a fuzzy information decision table based on the lung nodule discrimination scenario, as shown in Table 1, where: the individual sample set represents 9 patients; the conditional attribute set represents 6 medical test indexes, and their values are all preprocessed data, represents the staining depth rating, represents the nuclear-cytoplasmic ratio, represents index, that is, the proportion of positive cells (%), represents the degree of cell arrangement disorder, represents the nodule edge morphology (lobulation index), represents the nodule growth rate; the decision attribute set , respectively represent two categories of lung nodule diagnosis, .

[0033] Table 1

[0034] Step 1, determine the importance degree of each attribute with respect to the deterministic elements through the rough set positive region method 1.1 In the fuzzy decision information system , for each conditional attribute , calculate the distance function: , Here only list the distance function of the conditional attribute : , 1.2 Calculating all attributes The distance function , , 1.3 Calculate the total attributes of each sample The lower approximation of : , Calculated: , , , , , , , , , 1.4 Calculating decision variables For all attribute subsets Dependencies: .

[0035] Step 2: Determine the importance of each attribute with respect to the non-deterministic element using the information entropy method 2.1 Binarization of Fuzzy Decision System For each conditional attribute , using the fuzzy clustering algorithm FCM to cluster the samples into Clusters (in this embodiment, ), , where each individual The degree of membership in each cluster is According to the maximum membership principle, the fuzzy clustering is transformed into a deterministic classification, and each individual Belong to the category with the maximum membership, and generate a deterministic information decision system , as shown in Table 2.

[0036] Table 2

[0037] 2.2 In Calculate the information entropy of decision attributes in , 2.3 Calculating the full attribute subset Induced sample partitioning: , 2.4 Calculating a subset of all attributes The conditional entropy of: .

[0038] Step 3, calculate the weighted coefficient : 。

[0039] Step 4, initialize the reduction set , and perform iterative reduction to obtain the attribute reduction set 4.1 The first round of iteration 4.1.1 For each unselected attribute , calculate the attribute dependency degree based on the positive region: , , , , , , 4.1.2 Calculate the importance of attribute based on the positive region: Since , so , and the result is shown in Table 3.

[0040] Table 3

[0041] 4.1.3 Calculate the partition induced by the conditional attribute set in :

[0042] 4.1.4 Calculate the conditional entropy of the attribute : , , , , , , 4.1.5 Calculate the importance of attribute based on the information entropy: , and the result is shown in Table 4.

[0043] Table 4

[0044] 4.1.6 Calculate the comprehensive importance of the conditional attribute : , and the result is shown in Table 5.

[0045] Table 5

[0046] 4.1.7 Select the current optimal attribute: ; 4.1.8 Update the reduction set ; 4.1.9 At this time , , the termination condition is not reached, and the second round of iteration is performed.

[0047] 4.2 Second round of iteration 4.2.1 For each unselected attribute , calculate the importance of the attribute based on the positive region, and the results are shown in Table 6; ,

[0048] Table 6

[0049] 4.2.2 Calculate the importance of the attribute based on information entropy, and the results are shown in Table 7: , In , calculate the partition induced by the conditional attribute set :

[0050] Table 7

[0051] 4.2.3 Calculate the comprehensive importance of the conditional attribute : , and the results are shown in Table 8.

[0052] Table 8

[0053] 4.2.4 Select the current optimal attribute: ; 4.2.5 Update the reduction set ; 4.2.6 At this time , , the termination condition is not reached, and the third round of iteration is performed.

[0054] 4.3 Third round of iteration 4.3.1 For each unselected attribute , calculate the importance of the attribute based on the positive region, and the results are shown in Table 9: ,

[0055] Table 9

[0056] 4.3.2 Computed Properties Importance based on information entropy: Since we have obtained ,so ,property The importance based on information entropy is ; 4.3.3 Attributes at this time The comprehensive importance of Decide, maximum; 4.3.4 Select the current optimal attribute: ; 4.3.5 Updating the Reduced Set ; 4.3.6 At this time , , the termination condition is not reached, and the fourth round of iteration is performed.

[0057] 4.4 Fourth Iteration 4.4.1 For each unselected attribute , calculated properties Based on the importance of the positive domain, the results are shown in Table 10: ,

[0058] Table 10

[0059] 4.4.2 Select the current optimal attribute: ; 4.4.3 Updating the Reduced Set ; 4.4.4 At this time , , the termination condition is not reached, and the fifth round of iteration is performed.

[0060] 4.5 Fifth Iteration 4.5.1 For each unselected attribute , computed properties Based on the importance of the positive domain, the results are shown in Table 11: ,

[0061] Table 11

[0062] 4.5.2 Select the current optimal attribute: ; 4.5.3 Update the reduction set ; 4.5.4 At this time , , the termination condition is reached, the reduction process ends, and the attribute reduction set is obtained .

[0063] It shows that in the decision-making diagnosis of lung nodules, among the 6 medical test indicators, the "nuclear-cytoplasmic ratio" has no meaning for the test conclusion and is a redundant test, which can be cancelled.

[0064] Example 2 It is a fuzzy information decision table based on the diagnosis scenario of diabetic nephropathy, as shown in Table 12. Among them, the individual sample set represents 10 diabetic patients; the conditional attribute set represents 4 medical test indicators, and their values are all preprocessed data. Among them represents fasting blood glucose, represents glycated hemoglobin, represents the ratio of urinary microalbumin to creatinine, represents systolic blood pressure; the decision attribute set , respectively represent whether the patient is diagnosed with diabetic nephropathy.

[0065] Table 12

[0066] According to the method of the present invention, the reduction set obtained is , indicating that "fasting blood glucose" and "the ratio of urinary microalbumin to creatinine" are the core attributes for diagnosing diabetic nephropathy, and this reduction result is consistent with the pathological mechanism of diabetic nephropathy "hyperglycemia → renal microvascular injury".

[0067] Based on the same inventive concept, an embodiment of the present application provides a computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the steps of the foregoing medical test index reduction method based on the combination of the positive region and information entropy of rough sets are implemented.

[0068] Based on the same inventive concept, an embodiment of the present application provides a computer-readable storage medium, which stores a computer program. When the computer program is executed by a processor, the steps of the foregoing medical test index reduction method based on the combination of the positive region and information entropy of rough sets are implemented.

[0069] Those skilled in the art should understand that the embodiments of the present invention can be provided as a method, a system, or a computer program product. Therefore, the present invention can take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present invention can take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) that contain computer-usable program code.

[0070] The present invention is described with reference to the flowcharts and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the present invention. It should be understood that each flow and / or block in the flowchart and / or block diagram, and the combination of flows and / or blocks in the flowchart and / or block diagram, can be realized by computer program instructions. These computer program instructions can be provided to the processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing devices to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing devices generate means for realizing the functions specified in Figure 1 one flow or multiple flows and / or blocks Figure 1 one block or multiple blocks.

[0071] These computer program instructions can also be stored in a computer-readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer-readable memory generate a manufactured article including instruction means, and the instruction means realizes the functions specified in Figure 1 one flow or multiple flows and / or blocks Figure 1 one block or multiple blocks.

[0072] These computer program instructions can also be loaded onto a computer or other programmable data processing device, so that a series of operation steps are executed on the computer or other programmable device to generate a computer-implemented process, and thus the instructions executed on the computer or other programmable device provide steps for realizing the functions specified in Figure 1 one flow or multiple flows and / or blocks Figure 1 one block or multiple blocks.

[0073] The above embodiments are only used to illustrate the technical idea of the present invention, and the protection scope of the present invention cannot be limited thereby. Any modification made on the basis of the technical solution according to the technical idea proposed by the present invention falls within the protection scope of the present invention.

Claims

1. A medical test index reduction method based on the comprehensive of positive region of rough set and information entropy, which is used to reduce all medical test indexes for a certain disease, and is characterized in that It includes the following steps: Step 1, obtain a medical test index dataset, where each sample in the medical test index dataset includes a patient, all the test index data of the patient for a certain disease, and the diagnosis conclusion of the patient for a certain disease; convert the medical test index dataset into a fuzzy information decision system , where represents the set of all patients, represents the th patient, , is the number of all patients, represents the set of all medical test indexes for a certain disease, represents the th index, , is the number of all medical test indexes, represents dividing into non - overlapping diagnosis conclusion categories, represents the th diagnosis conclusion category, , is the number of diagnosis conclusion categories; Step 2: Calculate the set of all medical test indicators using the rough set positive domain method The positive domain of Calculate the weighted coefficient in the positive domain And the diagnostic conclusion category set right Dependence ; Step 3: Convert the fuzzy information decision system into a deterministic information decision system through a clustering method , and calculate in the deterministic information decision system information entropy and the conditional information entropy under the full attributes ; ; Step 4, set the reduction set as , and initialize the reduction set as an empty set; Step 5, for the set in which the ones not selected for each index, combine each index with the current reduction set to obtain the union set corresponding to each index, calculate the importance degree based on the positive region of the rough set and the importance degree based on the information entropy of the union sets corresponding to each index, and then calculate the comprehensive importance degree of the union sets corresponding to each index; Step 6: Add the index corresponding to the maximum comprehensive importance to the current reduction set to obtain a set which serves as the new reduction set ; Step 7: Calculate the diagnostic conclusion category set according to the same method as in Step 2 For the new reduct set obtained in Step 6 Dependency , and calculate the conditional entropy of the new reduct set obtained in Step 6 according to the same method as in Step 3 Below ; Step 8, determine whether it satisfies and , if satisfied, output the new reduction set obtained in Step 6 as the final index reduction set; otherwise, return to Step 5, and continue to select the index corresponding to the maximum comprehensive importance from the indices that have not been selected and add it to the new reduction set obtained in Step 6, and generate a new reduction set again, until it satisfies and .

2. The medical test index reduction method based on the comprehensive positive region of rough sets and information entropy according to claim 1, characterized in that The specific process of the said step 2 is as follows: Step 2.1, in the fuzzy information decision-making system for each medical test index calculate the distance between any two patients to obtain the distance matrix corresponding to each medical test index ; Among them, for medical test indicators , the distance calculation formula between any two patients is as follows: ， Among them, respectively represent the patient for medical test indicators values; Step 2.2, according to the distance matrices corresponding to all medical test indicators respectively, obtain the distance matrix of the set of ; The elements in are expressed as: ， Step 2.3, according to the distance matrix calculate the lower approximation of each patient with respect to the set as follows: ， Step 2.4, based on Step 2.3, calculate the positive region of the set as follows: , and the formula is as follows: ， Step 2.5, according to the set of the positive region calculate the weighted coefficient , the formula is as follows: ， Among them, represents the number of all patients, that is ; Step 2.6, according to the set positive region calculate the diagnostic conclusion category set for dependency , the formula is as follows: 。 3. The medical test index reduction method based on the comprehensive positive region and information entropy of rough sets according to claim 2, characterized in that The specific process of the said step 3 is as follows: Step 3.1, for each medical test index , use the fuzzy clustering algorithm to cluster all patients into clusters. According to the principle of maximum membership degree, transform the fuzzy clustering into a deterministic classification, and each patient belongs to the classification corresponding to the maximum membership degree to generate a deterministic information decision system ; Step 3.2, calculate in the information entropy of : ， Among them, represents the probability that a patient is classified into the th diagnostic conclusion category , and represents the number of patients classified into the th diagnostic conclusion category; Step 3.3, calculate the full attributes The combined partition of all patients induced by: , and then calculate the full attributes The conditional information entropy under .

4. The medical test index reduction method based on the comprehensive positive region of rough sets and information entropy according to claim 3, characterized in that The specific process of the said step 5 is as follows: Step 5.1, for each unselected index , combine with the reduction set obtained in the previous iteration to obtain . In the first iteration, ; Calculate each positive region : ， Among them, represents the lower approximation of each patient with respect to the set ; Calculate each Degree of dependence based on the positive region : ， Step 5.2, calculate each importance based on the positive region of rough sets: ， Among them, is the dependency degree of the current reduction set . During the first iteration, , ; Step 5.3, in calculate each induced partition ; the partition in the first iteration is the clustering obtained by using the fuzzy clustering algorithm. Step 5.4, calculate each conditional entropy : ， Among them, represents the number of patients in the set ; represents the number of patients diagnosed as in the set ; represents the number of classes into which the patients are divided under the condition set ; represents the total number of all patients, that is ; Step 5.5, calculate each importance based on information entropy : ， Among them, is the reduction set obtained in the previous iteration of the conditional entropy. In the first iteration, ; Step 5.6, calculate the comprehensive importance degree of each and : 。 5. A computer device, comprising a memory, a processor, and a computer program stored in the memory and capable of running on the processor, characterized in that, When the processor executes the said computer program, it implements the steps of the method for reducing medical test indexes based on the comprehensive positive region and information entropy of rough sets as described in any one of claims 1 to 4.

6. A computer-readable storage medium storing a computer program, characterized in that, When the said computer program is executed by the processor, it implements the steps of the method for reducing medical test indexes based on the comprehensive positive region and information entropy of rough sets as described in any one of claims 1 to 4.

Images