Method and device for assessing the type of organic intestinal disease based on a rule-based reasoning of confidence

Abstract

The application relates to a kind of organic intestinal disease type evaluation method and device based on confidence rule base reasoning.The input and output characteristics of BRB-ER model are determined according to the clinical data of organic intestinal disease;The collected clinical data of patients is coded and converted and given semantic value, an initial rule base is established, and the rule weight of confidence rule base and the confidence parameter of output result are set;The input feature vector is converted into corresponding matching degree calculation;Calculate the activation weight corresponding to the activated rule;The activated rule is inferred and fused through ER algorithm to obtain the confidence distribution result;According to the type evaluation result, the confidence rule base of colorectal cancer and ischemic intestinal disease is analyzed and optimized, and the optimal evaluation result is obtained.The application can establish the nonlinear complex relationship between the input characteristics and the evaluation result of organic intestinal disease, and provide active assistance to the doctor, effectively improve the accuracy of organic intestinal disease type evaluation.

Description

Technical Field

[0001] This invention relates to an assessment method and apparatus based on clinical data of organic bowel diseases and reasoning using a confidence rule base. Background Technology

[0002] Organic bowel diseases are a common clinical condition caused by lesions and space-occupying lesions of the intestinal mucosa. If left untreated, they can cause permanent damage to intestinal organs or related tissue systems, and even be life-threatening. Clinically, there are six types of organic bowel diseases: ulcerative colitis, Crohn's disease, intestinal polyps, intestinal adenomas, colorectal cancer, and ischemic bowel disease. Among these, colorectal cancer is one of the major malignant tumors threatening human health.

[0003] According to the latest global cancer burden data released by the WHO's International Agency for Research on Cancer in 2020, more than 1.93 million people worldwide were newly diagnosed with colorectal cancer, accounting for 9.7% of all new cancer diagnoses globally. Its incidence rate ranks third, after breast and lung cancer; the number of deaths from colorectal cancer reached 935,000, accounting for approximately 9.4% of all malignant tumor deaths, ranking second. With rapid economic development and an aging population, it has become the second most common and most prevalent malignant tumor of the digestive system. Most colorectal cancers develop slowly from precancerous lesions, generally taking 5-10 years. This means that colorectal cancer is one of the few diseases that can be effectively detected through population screening and preventative treatment can be implemented. Therefore, if early detection, diagnosis, and treatment can be achieved, patients have a high chance of prevention and cure, potentially significantly reducing its incidence and mortality rates.

[0004] In addition to the above, current clinical practice still primarily relies on pathological characteristics and behavioral manifestations to assess the type of organic bowel disease in patients. This traditional method of organic bowel disease classification first collects clinical data, and then doctors, based on their understanding of the pathological condition and clinical experience, arrive at a classification result. Due to the influence of subjective experience, the processing of multi-source information introduces errors and uncertainties. Furthermore, the classification process involves uncertain data measurement (such as data ambiguity, imprecise terminology, and incomplete data), and the varying levels of physician experience lead to inaccurate results. Therefore, to effectively prevent colorectal cancer and significantly reduce its incidence and mortality, an objective, accurate, and scientific organic bowel disease classification scheme based on an expert knowledge base and reasoning framework is needed. Summary of the Invention

[0005] To address the shortcomings of existing technologies, this invention proposes a method, apparatus, storage medium, and device for assessing organic intestinal disease types based on confidence rule base reasoning.

[0006] A first aspect of the present invention provides a method for assessing the type of organic bowel disease based on confidence rule base reasoning, comprising:

[0007] S1: Collect clinical data of patients with organic bowel disease and provide multiple characteristics to assess the type of organic bowel disease;

[0008] S2: Determine the input characteristics and output type of the organic bowel disease assessment model;

[0009] S3: Obtain the sample dataset, divide the sample dataset into training and test sets according to different proportions, and construct the BRB-ER model for assessing the type of organic bowel disease.

[0010] A second aspect of the present invention provides an apparatus for assessing the type of organic bowel disease based on confidence rule base reasoning, comprising:

[0011] Data collection unit: Used to collect clinical data of patients with organic bowel disease, providing multiple features to assess the type of organic bowel disease;

[0012] Determining Unit: Used to determine the input features and output type of the organic bowel disease assessment model;

[0013] Model building unit: used to obtain sample datasets, divide the sample datasets into training and test sets according to different proportions, and construct a BRB-ER model for assessing organic bowel disease types.

[0014] A third aspect of the present invention provides a storage medium comprising stored instructions, wherein, when the instructions are executed, the device containing the storage medium is controlled to perform the above-described method for assessing the type of organic bowel disease.

[0015] A fourth aspect of the present invention provides an electronic device including a memory and one or more instructions, wherein one or more instructions are stored in the memory and configured to be executed by one or more processors using the above-described method for assessing the type of organic bowel disease.

[0016] The beneficial effects of this invention are as follows:

[0017] This invention establishes a Confidence Rule Base Inference (BRB-ER) model to analyze the nonlinear relationship between input features (past history, symptoms, and auxiliary examinations) and output results (ulcerative diseases, Crohn's disease, intestinal polyps, intestinal adenomas, colorectal cancer, and ischemic bowel diseases). Furthermore, the BRB-ER model is non-encapsulated and can be specifically optimized for different diseases. It not only explains the details but also has higher persuasiveness and inference accuracy, effectively improving the accuracy of organic bowel disease type assessment and demonstrating excellent practical applicability. Attached Figure Description

[0018] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the accompanying drawings used in the embodiments will be briefly described below. Referring to the accompanying drawings will provide a clearer understanding of the features and advantages of the present invention. The drawings are illustrative and should not be construed as limiting the present invention in any way. For those skilled in the art, other drawings can be obtained based on these drawings without any creative effort. Wherein:

[0019] Figure 1 This is a flowchart of the method of the present invention.

[0020] Figure 2 This is a chart showing the results of the assessment of organic bowel disease types.

[0021] Figure 3 This is a diagram illustrating the optimization process for example cases.

[0022] Figure 4 This is a schematic diagram of the structure of an organic intestinal disease type assessment device provided by the present invention.

[0023] Figure 5 This is a schematic diagram of the structure of an electronic device provided by the present invention. Detailed Implementation

[0024] To better understand the above-mentioned objectives, features, and advantages of the present invention, the present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0025] Many specific details are set forth in the following description in order to provide a full understanding of the invention. However, the invention may also be practiced in other ways different from those described herein, and therefore the scope of protection of the invention is not limited to the specific embodiments disclosed below.

[0026] like Figure 1 As shown, the proposed method for assessing organic bowel disease types based on confidence rule base reasoning in this invention determines the input and output features of the BRB-ER model based on clinical data of organic bowel diseases; the collected clinical data of patients with organic bowel diseases are encoded and transformed using expert experience and assigned semantic values ​​to establish an initial rule base, and the rule weights and output confidence parameters of the confidence rule base are set; the input feature vectors are converted into corresponding matching degree calculations; the activation weights corresponding to the activated rules are calculated; the activated rules are reasoned and fused using the ER algorithm to obtain the confidence distribution results of six types of organic bowel diseases; based on the assessment results of organic bowel disease types, the confidence rule bases for colorectal cancer and ischemic bowel disease are analyzed and optimized, and updated to obtain the optimal diagnostic results.

[0027] To facilitate understanding of the above technical solutions of the present invention, the following detailed explanation of the above technical solutions of the present invention will be provided through actual data processing and model building.

[0028] This invention provides a method for assessing the type of organic bowel disease based on confidence rule base reasoning, comprising the following steps:

[0029] S1: Clinical data of patients with organic bowel diseases (OBDs) from a top-tier traditional Chinese medicine hospital between 2019 and 2021 were collected. Based on the doctors' clinical experience, 14 characteristics were provided for diagnosing organic bowel diseases, including history of radiotherapy and chemotherapy (x1), antibiotic use (x2), past medical history (x3), abdominal pain (x4), diarrhea (x5), changes in bowel habits (x6), rectal bleeding (x7), stool shape (x8), perianal abscess (x9), and fecal occult blood (x1). 10 ), colonoscopy (x) 11 ), pathology (x 12 Abdominal CT (x) 13 ) and mesenteric CAT(x 14 ).

[0030] S2: Determine the input features and output types of the organic bowel disease assessment model. Use the 14 features provided by expert experience in step S1 as the input of the diagnostic model. The output of the diagnostic model is six types of organic bowel diseases, such as ulcerative colitis (D1), Crohn's disease (D2), intestinal polyps (D3), intestinal adenoma (D4), colorectal cancer (D5), and ischemic bowel disease (D6).

[0031] S3: After screening, the entire dataset contains 458 samples, arranged chronologically. These include 75 cases each of Crohn's disease, intestinal adenoma, and ischemic bowel disease; 74 cases each of intestinal polyps and colorectal cancer; and 85 cases of ulcerative colitis. 368 samples collected in 2019-2020 were used as training samples for constructing the confidence rule base, and the remaining 90 samples collected in 2021 were used as test samples for model validation. A BRB-ER model for assessing organic bowel disease types was constructed.

[0032] S3-1: The clinical data of patients with organic bowel disease collected are encoded and transformed using expert experience, and semantic values ​​are assigned to establish an initial rule base and set the initial confidence rule base parameters.

[0033] S3-2: Convert the input feature vector into the corresponding matching degree, obtain the overall matching degree of the input samples of organic bowel disease, and calculate the activation weight of the activated rule.

[0034] S3-3: The activated rules are fused using the ER algorithm to calculate the final confidence distribution of the six organic bowel diseases.

[0035] S3-4: Based on the assessment results of the six types of organic bowel diseases obtained in step S3-3, optimize the confidence rule base corresponding to colorectal cancer and ischemic bowel disease. After updating the confidence rule base, repeat steps S3-2 and S3-3 to finally obtain the optimal assessment results of the organic bowel disease types.

[0036] The specific steps of step S3-1 are as follows:

[0037] S3-1-1: Encode and transform the collected clinical data on organic bowel diseases and assign semantic values. For example, input features include past medical history, which has 8 categories (no past medical history, intestinal tumor, inflammatory tumor, infectious tumor, hereditary tumor, functional tumor, and cardiovascular disease history). These 8 categories are assigned semantic values ​​of 0, 1, 2, ..., 7, 8 respectively. The remaining 13 input features are assigned different semantic values ​​according to different categories, just like above.

[0038] S3-1-2: After encoding and transforming clinical data and assigning semantic values, construct a confidence rule base containing L rules to describe the input features x of organic bowel diseases. i With output diagnostic results (D) j Given the complex nonlinear relationship between them, the k-th rule in the established confidence rule base is:

[0039]

[0040] In equation (1), R k (k = 1, ..., L) represents the k-th rule in the confidence rule base, x i (i = 1, ..., T) k ) is the i-th input feature of the k-th rule. Let x represent the i-th input feature in the k-th rule. i The corresponding set of reference values. D j (j=1,...,N) represents the enteropathy output by the confidence rule, β j,k (j=1,...,N,k=1,...,L) represents the confidence level of the j-th output of enteropathy in the k-th rule, L is the number of all rules in the rule base, and T k Let θ be the number of input features used in the k-th rule, N be the total number of possible outcomes in the rule base, and θ be the number of possible outcomes. k Let be the weight of the k-th rule.

[0041] To facilitate understanding, steps S3-1-1 to S3-1-2 are illustrated below:

[0042] The specific features of the input, such as history of radiotherapy and chemotherapy and antibiotic use, are encoded, transformed, and assigned semantic values ​​as follows:

[0043] Table 1. Semantic values ​​and reference values ​​of input features

[0044]

[0045] Construct a confidence rule base containing L rules to describe the input features x of organic bowel diseases. i With output diagnostic results (D) j The complex nonlinear relationship between them is illustrated by the following three rules from the established confidence rule base:

[0046] Table 2 Confidence Rule Base

[0047]

[0048] The specific steps of step S3-2 are as follows:

[0049] S3-2-1: Transform the information of medical features into a confidence distribution form that the BRB-ER model system can recognize, and calculate the sample data relative to the input features x in the confidence rule base. i The degree of matching between different reference values ​​can be represented as a confidence distribution, where the degree of matching between the input features and the reference values ​​is expressed.

[0050] S(x i ,ε i )={(A i,j ,α i,j ),j=1,...,j i}, i=1,...,T k (2)

[0051] In equation (2), S represents the confidence assessment assigned to the input feature, and A i,j This represents the input feature x. i The j-th reference value, α i,j The degree of matching to the reference value, α i,j ≥0 and j i α represents the number of reference values. i,j By inputting sample X = {X i The values ​​are compared with discrete values ​​in the confidence rule base using an XOR operation. XOR is a mathematical operator, and its mathematical symbol is "⊙". The number of "1"s in the XOR result is counted, and then the data is normalized.

[0052] S3-2-2: Based on the confidence distribution form in step S3-2-1, calculate the activation weights of the activated rules in the confidence rule base. The activation weight of the k-th rule is:

[0053]

[0054] In equation (3), This represents a reference value for the i-th input feature in the k-th rule. (It is A) i,j The matching degree of an element is called the individual matching degree, and is calculated according to formula (2) in step (3-2-1). It represents the degree of combination matching, calculated using a multiplicative aggregation function.

[0055] To facilitate understanding, the calculation process of matching degree and activation weight is illustrated with an example, Sample1, as shown below:

[0056] Table 3 Sample Data

[0057]

[0058] The BRB-ER model of this invention has 24 rules (the specific data for three rules are shown in Table 1), all of which are discrete input quantities. The matching degree of the sample input data Sample1 is calculated as follows: When Sample is compared with Rule1, 9 input feature data are equal, denoted as a1 = 9; when Sample is compared with Rule2, 10 input feature data are equal, denoted as a2 = 10; when Sample1 is compared with Rule3, 9 input feature data are equal, denoted as a3 = 8; when Sample is compared with Rule... 24 In comparison, there are 9 identical input feature data points, denoted as a. 24 =3; similarly, the rest a4~a 23 As calculated above.

[0059] Then the data is normalized, transforming the discrete input values ​​of the rules into a confidence distribution of the matching degree: for Rule 1 discrete confidence, transform α... 1,1 = a1 / a1+a2+a3+...+a 24 =9 / 9 + 10 + 8 + ... + 3 = 0.0743; Discrete reliability transformation α 1,2 = a2 / a1+a2+a3+...+a 24 =10 / 9 + 10 + 8 + ... + 3 = 0.0826; Discrete reliability transformation α 1,3 = a3 / a1+a2+a3+...+a 24 =8 / 9 + 10 + 8 + ... + 3 = 0.0661; Discrete reliability transformation α 1,24= a3 / a1+a2+a3+...+a 24 =3 / 9 + 10 + 8 + ... + 3 = 0.0248. Similarly, the rest of α... 1,4 ~α 1,23 As calculated above. Therefore, S(x) i ,ε i )={(A i,j ,α i,j )}={(0, 0.0743), (1, 0.0826), …, (3, 0.0248)}

[0060] The matching degree is known and can be calculated using formula (3). The activation weights for each rule in Sample1 are shown in the table below:

[0061] Table 4 Activation weights for each rule

[0062]

[0063] The specific steps of step S3-3 are as follows:

[0064] S3-3-1: The confidence level of the inference result can be obtained by fusing all activated rules using the ER algorithm, and parsing the confidence level of each output enteropathy:

[0065]

[0066]

[0067] S3-3-2: The confidence distribution of the inference results can be represented as {(D1, β1), (D2, β2), ..., (D...} j ,β j )}, where β j For the j-th output of the k-th rule, the enteropathy D j The final confidence level.

[0068] The bowel disease type with the highest confidence level is selected as the output of the assessment:

[0069] f(x) = argmax(β) j ),j=1,...,N (5)

[0070] To facilitate understanding, steps S3-3-1 to S3-3-2 are illustrated here. According to formula (4), all rules are fused. The result of fusion of all rules in Sample1 is shown below:

[0071] Table 5 Confidence levels of the output results for each rating type

[0072] Output Output disease type Confidence <![CDATA[D1]]> Ulcerative diseases 0.3065 <![CDATA[D2]]> Crohn's disease 0.1387 <![CDATA[D3]]> Intestinal polyps 0.0946 <![CDATA[D4]]> Intestinal adenoma 0.0749 <![CDATA[D5]]> Colorectal cancer 0.1968 <![CDATA[D6]]> Ischemic bowel disease 0.1885

[0073] According to formula (5), the highest confidence level (D1, 0.3065) was selected, and the corresponding intestinal disease type was ulcerative colitis as the type assessment result. As shown in Table 3, the sample data consisted of patients with ulcerative colitis. Therefore, the type assessment result by the BRB-ER model was correct.

[0074] The specific steps of steps S3-4 are as follows:

[0075] After obtaining the assessment results for six types of organic bowel diseases, the accuracy rate for the assessment results of the other four types of organic bowel diseases was 100%, except for colorectal cancer and ischemic bowel. Therefore, the confidence rules for colorectal cancer and ischemic bowel diseases were optimized.

[0076] S3-4-1: In all cases of misdiagnosed colorectal cancer, the cancer was misclassified as a polyp. Calculate the activation weight ω according to step S3-2-2. k At the same time, optimize the rule with the highest activation weight among all rules that are misclassified as polyps, and increase the activation weight of this rule for all cases of misclassified colorectal cancer types.

[0077] S3-4-2: Among all misdiagnosed ischemic bowel disease types, the probability of misdiagnosis as Crohn's disease and intestinal adenoma is relatively high. Calculate the activation weight ω according to step S3-2-2. k At the same time, optimize the two rules with higher values ​​among all misclassified Crohn's disease and intestinal adenoma, and increase the activation weight of these two rules for all cases of ischemic bowel disease types that are misclassified.

[0078] S3-4-3: Obtain the optimized confidence rule base, repeat steps S3-2 and S3-3 to obtain the optimal assessment result for the organic bowel disease type.

[0079] To facilitate understanding, the optimization process of steps S3-4-1 to S3-4-3 is explained in detail below:

[0080] Calculated from steps S3-2 and S3-3 Figure 2 The results of the organic bowel disease classification show that the overall accuracy of organic bowel disease classification on the test set was 88.9%. Except for ischemic bowel disease and colorectal cancer, the accuracy of the other four organic bowel disease classifications was 100%. In all cases of misclassified colorectal cancer, the disease was misclassified as polyps. This was considered when calculating the activation weight ω. k At that time, among all the rules misclassified as polyps, the rule 11 The activation weight is the highest, ω 11 =0.0619. Therefore, in response to this situation, the present invention optimizes rule 11, improving the handling of all misdiagnosed colorectal cancer types in the rule. 11Activation weights, optimized rules 11 Confidence level: {β1=0, β2=0, β3=0.8, β4=0, β5=0.2, β6=0}. Finally, steps S3-2 and S3-4 were repeated, and all three cases of colorectal cancer that were misdiagnosed as polyps were correctly identified.

[0081] Among all cases of misdiagnosed ischemic bowel disease, the probability of misdiagnosis as Crohn's disease and intestinal adenoma was relatively high, when calculating the activation weight ω. k At that time, among all the rules misdiagnosed as Crohn's disease and intestinal adenoma, the rule 15 Rule 18 The activation weight is relatively high, ω 15 =0.0585, ω 18 =0.0525. Therefore, in response to this misclassification, this invention has also optimized rules 15 and 18, improving the accuracy of all misclassified ischemic bowel disease cases in the Rule. 15 and Rule 18 Activation weights, optimized rule 15 Rule 18 The confidence levels were {β1=0, β2=0, β3=0, β4=0.7, β5=0, β6=0.3} and {β1=0, β2=0, β3=0, β4=0, β5=0.9, β6=0.1}, respectively. Steps S3-2 and S3-3 were repeated, and the two misdiagnosed cases of ischemic bowel disease were correctly diagnosed. The final experimental results for the organic bowel disease type assessment are shown in Table 6.

[0082] Table 6. Accuracy of BRB-ER Model Type Assessment Before and After Optimization

[0083]

[0084] This optimization method improved the assessment results of organic bowel disease types in this invention, increasing the test set accuracy to 94.4%. To verify the effectiveness of the method, a case study was used to describe in detail the optimization process for its type assessment, such as... Figure 3 As shown. The three rules after case activation optimization, with rule weights determined by ω. 11 The value increased from 0.063 to 0.071, ω 15 The value decreased from 0.038 to 0.035, ω 18 The confidence level was increased from 0.056 to 0.061, and the final assessment result changed from a misclassification as polyps to a correct assessment of ischemic bowel disease. Therefore, this optimization method provides a more detailed analysis of the nonlinear relationship between the output assessment result and the input features. By adjusting the confidence level of the rule base, it clearly reflects the interpretability of the BRB-ER model and the degree of expert participation.

[0085] and Figure 1 Corresponding to the method for assessing the type of organic bowel disease shown, this embodiment of the invention also provides an apparatus for assessing the type of organic bowel disease, used for... Figure 1 The specific implementation of the method shown is illustrated in the following diagram. Figure 4 As shown, it includes:

[0086] Data collection unit: Used to collect clinical data of patients with organic bowel disease, providing multiple features to assess the type of organic bowel disease;

[0087] Determining Unit: Used to determine the input features and output type of the organic bowel disease assessment model;

[0088] Model building unit: used to obtain sample datasets, divide the sample datasets into training and test sets according to different proportions, and construct a BRB-ER model for assessing organic bowel disease types.

[0089] exist Figure 4 Based on the device shown, the device provided in this embodiment of the invention can be further extended to include multiple units. The functions of each unit can be found in the descriptions of the various embodiments of the organic bowel disease type assessment method provided above, and will not be further illustrated here.

[0090] This invention also provides a storage medium that includes stored instructions, wherein when the instructions are executed, the device containing the storage medium is controlled to perform the organic bowel disease type assessment method as described above.

[0091] This invention also provides an electronic device, the structural schematic of which is shown below. Figure 5 As shown, it specifically includes memory and one or more instructions, wherein one or more instructions are stored in memory and configured to be executed by one or more processors to perform the following operations:

[0092] S1: Collect clinical data of patients with organic bowel disease and provide multiple characteristics to assess the type of organic bowel disease;

[0093] S2: Determine the input characteristics and output type of the organic bowel disease assessment model;

[0094] S3: Obtain the sample dataset, divide the sample dataset into training and test sets according to different proportions, and construct the BRB-ER model for assessing the type of organic bowel disease.

[0095] The various embodiments in this specification are described in a progressive manner. Similar or identical parts between embodiments can be referred to mutually. Each embodiment focuses on describing the differences from other embodiments. In particular, for system or system embodiments, since they are basically similar to method embodiments, the description is relatively simple, and relevant parts can be referred to the descriptions in the method embodiments. The systems and system embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Those skilled in the art can understand and implement this without creative effort.

[0096] Those skilled in the art will further recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of both. To clearly illustrate the interchangeability of hardware and software, the components and steps of the various examples have been generally described in terms of functionality in the foregoing description. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementations should not be considered beyond the scope of this invention.

[0097] The above description of the disclosed embodiments enables those skilled in the art to make or use the invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the invention. Therefore, the invention is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A method for assessing the type of organic bowel disease based on confidence rule base reasoning, characterized in that, include: S1: Collect clinical data of patients with organic bowel disease and provide multiple characteristics to assess the type of organic bowel disease; S2: Determine the input characteristics and output type of the organic bowel disease assessment model; S3: Obtain the sample dataset, divide the sample dataset into training and test sets according to different proportions, and construct the BRB-ER model for assessing the type of organic bowel disease; Step S3 includes: S3-1: Encode and transform the collected clinical data of patients with organic bowel diseases and assign semantic values ​​to establish an initial rule base and set the initial confidence rule base parameters; specifically: After encoding and transforming the collected clinical data on organic bowel diseases and assigning semantic values, a system containing... L A confidence rule base of rules used to describe input features of organic bowel diseases. x i With output diagnostic results D j There exist complex nonlinear relationships between them; S3-2: After converting the input feature vector into the corresponding matching degree and obtaining the overall matching degree of the input samples for organic bowel disease, calculate the activation weights of the activated rules; specifically as follows: S3-2-1: Transform the information of medical features into a confidence distribution form that the BRB-ER model system can recognize, and calculate the sample data relative to the input features in the confidence rule base. x i The matching degree of different reference values ​​is represented by the matching degree of the input features to the reference values ​​as a confidence distribution. S3-2-2: Based on the confidence distribution form in step S3-2-1, calculate the activation weight of the activated rules in the confidence rule base; S3-3: The activated rules are fused using the ER algorithm to calculate the final confidence distribution of organic bowel disease. S3-4: Based on the assessment results of the organic bowel disease type obtained in step S3-3, optimize the confidence rule bases corresponding to colorectal cancer and ischemic bowel disease. After updating the confidence rule bases, repeat steps S3-2 and S3-3 to finally obtain the optimal assessment results; the specific steps are as follows: S3-4-1: In all cases of misdiagnosed colorectal cancer, the cancer was misclassified as a polyp. Calculate the activation weights according to step S3-2-2. ω k At the same time, optimize the rule with the highest activation weight among all rules that are classified as polyps, and increase the activation weight of this rule for all cases that are misclassified as colorectal cancer. S3-4-2: Among all misdiagnosed ischemic bowel disease types, the probability of misdiagnosis as Crohn's disease and intestinal adenoma is relatively high. Calculate the activation weights according to step S3-2-2. ω k At the same time, optimize the two rules with higher scores among all rules for Crohn's disease and intestinal adenoma, and increase the activation weight of these two rules for all cases misclassified as ischemic bowel disease. S3-4-3: Obtain the optimized confidence rule base, repeat steps S3-2 and S3-3 to obtain the optimal assessment result for the organic bowel disease type.

2. The method for assessing the type of organic bowel disease based on confidence rule base reasoning according to claim 1, characterized in that, The specific steps of step S3-3 are as follows: S3-3-1: The confidence of the inference results is obtained by fusing all activated rules using the ER algorithm and parsing the confidence of each output enteropathy. S3-3-2: Select the intestinal disease type with the highest confidence level as the output of the assessment.

3. An organic bowel disease type assessment device based on confidence rule base reasoning, used to implement the method of claim 1, characterized in that, include: Data collection unit: Used to collect clinical data of patients with organic bowel disease, providing multiple features to assess the type of organic bowel disease; Determining Unit: Used to determine the input features and output type of the organic bowel disease assessment model; Model building unit: used to obtain sample datasets, divide the sample datasets into training and test sets according to different proportions, and construct a BRB-ER model for assessing organic bowel disease types.

4. A storage medium, characterized in that, The storage medium includes stored instructions, wherein, when the instructions are executed, the device containing the storage medium is controlled to perform the organic bowel disease type assessment method as described in any one of claims 1 to 2.

5. An electronic device, characterized in that, It includes a memory, and one or more instructions, wherein one or more instructions are stored in the memory and configured to be executed by one or more processors as described in any one of claims 1 to 2 for the assessment of organic bowel disease types.

Images