CN112102952A - Method for identifying pathological category based on distance calculation method and related device - Google Patents

Abstract

The embodiment of the present invention discloses a method for identifying a pathological category based on a distance calculation method. By acquiring characteristic parameters to be diagnosed, the characteristic parameters to be diagnosed include N characteristic parameters corresponding to N types of characteristics; The standard parameters include M×N standard parameters of N kinds of features corresponding to M kinds of pathological categories respectively; using a preset distance calculation method to calculate the predicted distance between the feature parameters to be diagnosed and the standard parameters corresponding to each kind of pathological category, to obtain M prediction distances; determine the pathological category corresponding to the characteristic parameters to be diagnosed according to the size of the M predicted distances, and realize automatic pathological diagnosis by calculating and comparing the characteristic parameters to be diagnosed and the standard parameters corresponding to each pathological category one by one. And improve the objectivity and accuracy of pathological diagnosis. In addition, a system, computer equipment and storage medium for identifying pathological categories based on the distance calculation method are also proposed.

Description

Method and related equipment for identifying pathological categories based on distance calculation method

technical field

The invention relates to the field of computer technology, and in particular, to a method and related equipment for identifying a pathological category based on a distance calculation method.

Background technique

Pathological diagnosis is to study the cause of disease, pathogenesis, as well as the morphological structure, functional and metabolic changes of the diseased body during the disease process, and the outcome of the disease, so as to provide the necessary theoretical basis and practical basis for the diagnosis, treatment and prevention of the disease. Pathological diagnosis is the most reliable among various tumor examination methods. Pathological diagnosis is regarded as the "gold standard" and the final diagnosis of the disease.

From clinical symptoms to traditional HE pathology, there are many similar lesions, which are easily confused with a variety of diseases, including benign and malignant diseases. Once misdiagnosed, it will cause serious medical malpractice. In the process of disease diagnosis, differential diagnosis with many similar lesions is required. For example, in tumor diseases, there will be different levels of expression of various proteins and mutations of many different genes. The occurrence of these proteins and genes is called characteristic parameters. On the other hand, neoplastic diseases are called consequential phenomena. The most intuitive and detectable results of doctors are characteristic parameters. It is necessary to comprehensively analyze the expression of characteristic parameters, and integrate their own accumulated experience and knowledge in books and literature. Finally, predict the resulting phenomenon. The doctor sees the report of the test results and the imaging report, and then combines the medical history and clinical data to comprehensively select N more likely lesions, and then through comprehensive analysis and consultation and discussion, the most inclined diagnosis results are obtained. Finally, predict the resulting phenomenon. However, the current phenomenon of predicting results based on the experience of different levels of individuals has a serious impact of personal subjectivity. In addition, among different medical institutions and different doctors, the interpretation of consequential phenomena is quite different, which reduces the objectivity of pathological category identification, thereby affecting the quality of diagnosis and treatment to a certain extent, making the accuracy of pathological category identification not high. The missed rate of pathological diagnosis is high.

SUMMARY OF THE INVENTION

Based on this, it is necessary to propose a method, system, computer equipment and storage medium for identifying pathological categories based on the distance calculation method to improve the objectivity and accuracy of pathological diagnosis.

A method for identifying a pathological category based on a distance calculation method, the method comprising:

Obtaining feature parameters to be diagnosed, the feature parameters to be diagnosed include N feature parameters corresponding to N kinds of features, and N is a natural number;

Acquiring preset standard parameters, where the preset standard parameters include M×N standard parameters of N features corresponding to M types of pathological categories, where M is a natural number;

Calculate the predicted distances between the characteristic parameters to be diagnosed and the standard parameters corresponding to each pathological category by using a preset distance calculation method to obtain M predicted distances;

The pathological category corresponding to the to-be-diagnosed feature parameter is determined according to the size of the M predicted distances.

A system for identifying pathological categories based on a distance calculation method, the system comprising:

a first parameter obtaining module, configured to obtain characteristic parameters to be diagnosed, wherein the characteristic parameters to be diagnosed include N characteristic parameters corresponding to N kinds of features, and N is a natural number;

The second parameter acquisition module is configured to acquire preset standard parameters, where the preset standard parameters include M×N standard parameters of N features corresponding to M types of pathological categories, where M is a natural number;

a calculation module, configured to calculate the predicted distances between the characteristic parameters to be diagnosed and the standard parameters corresponding to each pathological category by using a preset distance calculation method, to obtain M predicted distances;

A diagnosis module, configured to determine the pathological category corresponding to the characteristic parameter to be diagnosed according to the size of the M predicted distances.

A computer device includes a memory and a processor, the memory stores a computer program, and when the computer program is executed by the processor, the processor performs the following steps:

Obtaining feature parameters to be diagnosed, the feature parameters to be diagnosed include N feature parameters corresponding to N kinds of features, and N is a natural number;

Acquiring preset standard parameters, where the preset standard parameters include M×N standard parameters of N features corresponding to M types of pathological categories, where M is a natural number;

Calculate the predicted distances between the characteristic parameters to be diagnosed and the standard parameters corresponding to each pathological category by using a preset distance calculation method to obtain M predicted distances;

The pathological category corresponding to the to-be-diagnosed feature parameter is determined according to the size of the M predicted distances.

A computer-readable storage medium storing a computer program, when executed by a processor, the computer program causes the processor to perform the following steps:

Obtaining feature parameters to be diagnosed, the feature parameters to be diagnosed include N feature parameters corresponding to N kinds of features, and N is a natural number;

Acquiring preset standard parameters, where the preset standard parameters include M×N standard parameters of N features corresponding to M types of pathological categories, where M is a natural number;

Calculate the predicted distances between the characteristic parameters to be diagnosed and the standard parameters corresponding to each pathological category by using a preset distance calculation method to obtain M predicted distances;

The pathological category corresponding to the to-be-diagnosed feature parameter is determined according to the size of the M predicted distances.

The above-mentioned method, system, computer equipment and storage medium for identifying pathological categories based on the distance calculation method, by obtaining the characteristic parameters to be diagnosed; obtaining preset standard parameters; using the preset distance calculation method to respectively calculate the characteristic parameters to be diagnosed and each The predicted distances between the standard parameters corresponding to the various pathological categories are obtained, and M predicted distances are obtained; the pathological categories corresponding to the characteristic parameters to be diagnosed are determined according to the size of the M predicted distances, and the pathological categories corresponding to the characteristic parameters to be diagnosed are determined by adopting a variety of methods based on distance calculation. Pathological diagnosis improves the accuracy and objectivity of pathological diagnosis.

Description of drawings

In order to explain the embodiments of the present invention or the technical solutions in the prior art more clearly, the following briefly introduces the accompanying drawings that need to be used in the description of the embodiments or the prior art. Obviously, the accompanying drawings in the following description are only These are some embodiments of the present invention. For those of ordinary skill in the art, other drawings can also be obtained according to these drawings without creative efforts.

in:

1 is a flowchart of a method for identifying a pathological category based on a distance calculation method in one embodiment;

2 is a flowchart of a method for calculating a predicted distance in one embodiment;

3 is a flowchart of a method for calculating a predicted distance in another embodiment;

4 is a flowchart of a method for determining a pathological category in one embodiment;

5 is a structural block diagram of a system for identifying a pathological category based on a distance calculation method in one embodiment;

FIG. 6 is a structural block diagram of a computer device in one embodiment.

Detailed ways

The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, but not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

As shown in FIG. 1 , in one embodiment, a method for identifying a pathological category based on a distance calculation method is provided. The method for identifying a pathological category based on a distance calculation method can be applied to both a terminal and a server. This implementation The example is applied to the server as an example. The method for identifying pathological categories based on the distance calculation method specifically includes the following steps:

Step 102: Obtain characteristic parameters to be diagnosed, where the characteristic parameters to be diagnosed include N characteristic parameters corresponding to N types of characteristics, where N is a natural number.

The to-be-diagnosed feature parameter refers to a parameter used to reflect the pathological feature of the to-be-diagnosed pathological slice, and the to-be-diagnosed feature parameter includes multiple feature parameters corresponding to multiple features. In a specific embodiment, the pathological section to be diagnosed is an epithelial ovarian malignant tumor, and the corresponding seven characteristic parameters may be the values corresponding to Pax-8, WT-1, CA125, P53, CEA, ER and PVHL. Specifically, the to-be-diagnosed characteristic parameter can be obtained after analyzing the pathological section by a pathological analysis instrument.

Step 104: Acquire preset standard parameters, where the preset standard parameters include M×N standard parameters of N features corresponding to M types of pathological categories, where M is a natural number.

Among them, the preset standard parameters are parameters set according to the size or range of N characteristic parameters under each pathological category, and the standard parameters correspond one-to-one with the characteristic parameters to be diagnosed, that is, each pathological category contains N standard parameters, so , M pathological categories contain M×N standard parameters. Continue to take the ovarian epithelial malignant tumor in step S102 as an example, which corresponds to 7 characteristic parameters, and its pathological categories include: serous adenocarcinoma, mucinous adenocarcinoma, endometrioid adenocarcinoma, clear cell adenocarcinoma and metastasis Gonadal carcinoma. There are N standard parameters corresponding to each pathological category. For example, for the 7 standard parameters whose pathological category is serous adenocarcinoma, the corresponding values of Pax-8, WT-1, CA125, P53, CEA, ER and PVHL are 95 respectively. %, 95%, 95%, 95%, 95%, 75% and 5%.

Step 106 , using a preset distance calculation method to calculate the predicted distances between the characteristic parameters to be diagnosed and the standard parameters corresponding to each pathological category, respectively, to obtain M predicted distances.

Wherein, the preset distance calculation method refers to a preset quantification method for comparing the similarity degree between the characteristic parameter to be diagnosed and the standard parameter. The distance calculation method may be one or more of the distance measures of Euclidean distance, Minkowski distance, Manhattan distance, Chebyshev distance, cosine similarity and/or Pearson correlation coefficient. The selection can be made according to the characteristics of each distance itself and the characteristics of the standard parameters and/or the characteristic parameters to be diagnosed. The prediction distance refers to the quantified value of the similarity degree between the characteristic parameter to be diagnosed and the standard parameter corresponding to each pathological category. Specifically, the N characteristic parameters in the characteristic parameters to be diagnosed and the N standard parameters corresponding to the M pathological categories are respectively calculated according to a preset distance calculation method, thereby obtaining M predicted distances. Understandably, the predicted distance between the characteristic parameter to be diagnosed and the standard parameter corresponding to each pathological category is calculated by the preset distance calculation method, which realizes the specific quantification of the similarity between the characteristic parameter to be diagnosed and the standard parameter, and improves the predicted distance. Calculated objectivity. And because the preset distance calculation method includes a variety of distance metrics, the accuracy of the predicted distance calculation is improved.

Step 108: Determine the pathological category corresponding to the feature parameter to be diagnosed according to the size of the M predicted distances.

Specifically, the pathological category corresponding to the characteristic parameter to be diagnosed is determined according to the specific numerical values corresponding to the M predicted distances and the positive correlation or inverse correlation between the magnitude of the predicted distance and the degree of similarity. For example, 5 pathological categories: serous adenocarcinoma, mucinous adenocarcinoma, endometrioid adenocarcinoma, clear cell adenocarcinoma, and metastatic adenocarcinoma correspond to the predicted distances: 0.7, 0.5, 0.4, 0.4, 0.5, and the predicted distances are the same as The degree of similarity is positively correlated, that is, the larger the predicted distance, the higher the degree of similarity, and the pathological category corresponding to the feature parameter to be diagnosed is the pathological category corresponding to the predicted distance of 0.7, namely serous adenocarcinoma. Understandably, by calculating and comparing the characteristic parameters to be diagnosed with the standard parameters corresponding to each pathological category one by one, the pathological diagnosis is automated, and the objectivity and accuracy of the pathological diagnosis are ensured.

In the above method for identifying pathological categories based on the distance calculation method, by acquiring characteristic parameters to be diagnosed, the characteristic parameters to be diagnosed include N characteristic parameters corresponding to N kinds of features; and preset standard parameters are obtained, and the preset standard parameters include M kinds of pathological categories M×N standard parameters corresponding to N kinds of features respectively; use a preset distance calculation method to calculate the predicted distance between the feature parameter to be diagnosed and the standard parameter corresponding to each pathological category, and obtain M predicted distances; according to M The size of the predicted distance determines the pathological category corresponding to the characteristic parameter to be diagnosed. By using a variety of distance calculation methods to calculate and compare the characteristic parameter to be diagnosed with the standard parameters corresponding to each pathological category one by one, the pathological diagnosis can be automated. , and improve the objectivity and accuracy of pathological diagnosis.

As shown in FIG. 2, in one embodiment, a preset distance calculation method is used to calculate the predicted distances between the feature parameters to be diagnosed and the standard parameters corresponding to each pathological category, respectively, to obtain M predicted distances, including:

Step 106A, respectively calculating the first distance and/or the second distance between each characteristic parameter and the standard parameters corresponding to M types of pathological categories, to obtain M×N characteristic distances;

In step 106B, the N feature distances corresponding to each pathological category are respectively fused and calculated to obtain M predicted distances.

Among them, the first distance and the second distance are two types of distances, which are classified according to the correlation with the degree of similarity. For example, the first distance is positively correlated with the similarity degree, and the second distance is inversely correlated with the similarity degree. Feature distance refers to the distance between a single feature parameter and the corresponding standard parameter. The feature distance can also be Euclidean distance, Minkowski distance, Manhattan distance, Chebyshev distance, cosine similarity and/or One or more of the distance measures for the Pearson correlation coefficient. Specifically, the distance between the N characteristic parameters and the standard parameters corresponding to M types of pathological categories is calculated respectively, and the distance is the first distance and/or the second distance, so as to obtain M×N characteristic distances; then each pathological category corresponds to The N feature distances are fused and calculated to obtain M predicted distances. The fusion calculation refers to a processing method that integrates multiple indicators for calculation. For example, the weight of each indicator can be set according to the importance of each indicator, and then the weighted sum calculation can be performed, or the automatic calculation can be performed according to preset rules. Adaptive fusion approach. It can be understood that, in this embodiment, by performing fusion calculation on each feature distance, the influence of each feature distance on the predicted distance is considered in fusion, which further ensures the accuracy of determining the predicted distance.

As shown in Figure 3, in one embodiment, the N feature distances corresponding to each pathological category are respectively fused and calculated to obtain M predicted distances, including:

Step 106B1, obtain the preset weight corresponding to each feature distance;

Step 106B2: Perform weighted calculation according to the N characteristic distances corresponding to each pathological category and the corresponding preset weights to obtain M predicted distances.

Specifically, a preset weight of each feature distance is first determined, and the preset weight can be set according to the influence of each feature distance on the correctness of the pathological diagnosis. Then, weighted calculation is performed according to the N characteristic distances corresponding to each pathological category and the corresponding preset weights to obtain M predicted distances. It can be understood that, in this embodiment, the predicted distance is obtained through weighted calculation, and the method of fusion calculation is simple and fast, thereby improving the speed of calculation of the predicted distance.

In one embodiment, the first distance is at least one of Euclidean distance, Minkowski distance, Manhattan distance and Chebyshev distance, and the second distance is cosine similarity and/or Pearson correlation coefficient.

其中，dist(X₁,Y₁)表示为欧几里得距离，x_i表示为第i个特征参数，y_i表示为与第i个特征参数对应的第i个标准参数。明可夫斯基距离(Minkowski Distance)是欧几里得距离的推广，是对多个距离度量公式的概括性的表述，其公式为：

其中，dist(X₂,Y₂)表示为明可夫斯基距离，x_i表示为第i个特征参数，y_i表示为与第i个特征参数对应的第i个标准参数，p为常数。曼哈顿距离(Manhattan Distance)来源于城市区块距离，是将多个维度上的距离进行求和后的结果，其公式为：

其中，dist(X₃,Y₃)表示为曼哈顿距离，x_i表示为第i个特征参数，y_i表示为与第i个特征参数对应的第i个标准参数。切比雪夫距离(ChebyshevDistance)是向量空间中的一种度量，二个点之间的距离定义是其各坐标数值差绝对值的最大值，其公式为：

其中，dist(X₄,Y₄)表示为切比雪夫距离，x_i表示为第i个特征参数，y_i表示为与第i个特征参数对应的第i个标准参数。且欧几里得距离、明可夫斯基距离、曼哈顿距离及切比雪夫距离与相似程度呈负相关关系，即第一距离越大，则相似度越低。第二距离为余弦相似度和/或皮尔森相关系数，余弦相似度(CosineSimilarity)是用向量空间中两个向量夹角的余弦值作为衡量两个个体间差异的大小，其公式为：

其中，sim(X,Y)表示为余弦相似度，x表示为特征参数，y表示为与特征参数x对应的标准参数。皮尔森相关系数(Pearson Correlation Coefficient)是用来衡量两个数据集合是否在一条线上面，它用来衡量定距变量间的线性关系，其公式为：

其中，r(X,Y)表示为皮尔森相关系数，x表示为特征参数，y表示为与特征参数x对应的标准参数。余弦相似度和皮尔森相关系数与相似程度呈负相关关系，即第二距离越大，则相似程度越高。可以理解地，第一距离及第二距离有各自的距离度量场景，因此，根据待诊断特征参数的应用场景选取合适的第一距离或者第二距离，以进一步提高病理诊断得准确性。Among them, the first distance is at least one of Euclidean distance, Minkowski distance, Manhattan distance and Chebyshev distance. The absolute distance between points, whose formula is:

Among them, dist(X ₁ , Y ₁ ) represents the Euclidean distance, _xi represents the ith characteristic parameter, and _yi represents the ith standard parameter corresponding to the ith characteristic parameter. Minkowski Distance is a generalization of Euclidean distance and a generalized expression of multiple distance measurement formulas. The formula is:

Among them, dist(X ₂ , Y ₂ ) is the Minkowski distance, x _i is the i-th characteristic parameter, y _i is the i-th standard parameter corresponding to the i-th characteristic parameter, and p is a constant . Manhattan Distance is derived from the city block distance, which is the result of summing the distances in multiple dimensions. The formula is:

Among them, dist(X ₃ , Y ₃ ) represents the Manhattan distance, _xi represents the ith feature parameter, and y _i represents the ith standard parameter corresponding to the ith feature parameter. Chebyshev distance (ChebyshevDistance) is a measure in the vector space, the distance between two points is defined as the maximum value of the absolute value of the difference between their coordinates, and its formula is:

Among them, dist(X ₄ , Y ₄ ) represents the Chebyshev distance, _xi represents the ith feature parameter, and y _i represents the ith standard parameter corresponding to the ith feature parameter. And Euclidean distance, Minkowski distance, Manhattan distance and Chebyshev distance are negatively correlated with similarity, that is, the larger the first distance, the lower the similarity. The second distance is the cosine similarity and/or the Pearson correlation coefficient. The cosine similarity (CosineSimilarity) uses the cosine value of the angle between the two vectors in the vector space as a measure of the difference between two individuals. The formula is:

Among them, sim(X, Y) represents the cosine similarity, x represents the feature parameter, and y represents the standard parameter corresponding to the feature parameter x. The Pearson Correlation Coefficient is used to measure whether two data sets are on a line. It is used to measure the linear relationship between spaced variables. The formula is:

Among them, r(X, Y) is the Pearson correlation coefficient, x is the characteristic parameter, and y is the standard parameter corresponding to the characteristic parameter x. The cosine similarity and the Pearson correlation coefficient are negatively correlated with the similarity, that is, the larger the second distance, the higher the similarity. Understandably, the first distance and the second distance have their own distance measurement scenarios. Therefore, an appropriate first distance or second distance is selected according to the application scenario of the feature parameter to be diagnosed to further improve the accuracy of pathological diagnosis.

In one embodiment, acquiring the preset weight corresponding to each feature distance includes: when the feature distance is the first distance, the preset weight is a negative number; when the feature distance is the second distance, the preset weight is a positive number.

Specifically, when the feature distance is the first distance, the preset weight is a negative number, and when the feature distance is the second distance, the preset weight is a positive number. Since the first distance is negatively correlated with the similarity, the corresponding preset weight is The weight is determined as a negative number, and the second distance is positively correlated with the degree of similarity, and the corresponding preset weight is determined as a positive number, so that the predicted distance calculated based on the preset weight is positively correlated with the degree of similarity, which can facilitate Pathological diagnosis must be carried out according to the predicted distance to further improve the efficiency of pathological diagnosis.

In one embodiment, the N feature distances corresponding to each pathological category are respectively fused and calculated to obtain M predicted distances, including: when the feature distance is the first distance, the predicted distance and the feature distance are negatively correlated; when When the feature distance is the second distance, the predicted distance is positively correlated with the feature distance.

As shown in Figure 4, in one embodiment, the pathological category corresponding to the feature parameter to be diagnosed is determined according to the size of the M predicted distances, including:

Step 108A, determining the proportion of each predicted distance in the M predicted distances as the probability of the corresponding pathological category;

Step 108B: Determine the pathological category corresponding to the characteristic parameter to be diagnosed according to the probability of each pathological category.

In this embodiment, the ratio of each predicted distance to the sum of M predicted distances is calculated, the ratio is determined as the probability of the pathological category corresponding to the feature parameter to be diagnosed, and the feature parameter to be diagnosed is determined according to the probability of each pathological category the corresponding pathological category. For example, 5 pathological categories: serous adenocarcinoma, mucinous adenocarcinoma, endometrioid adenocarcinoma, clear cell adenocarcinoma and metastatic adenocarcinoma correspond to the predicted distances: 0.7, 0.5, 0.4, 0.4, 0.5, corresponding to each pathology The probabilities of the categories are 28%, 20%, 16%, 16%, and 20% in sequence. Therefore, the serous adenocarcinoma with a probability of the pathological category of 28% is the pathological category corresponding to the characteristic parameters to be diagnosed. Understandably, by determining the proportion of each predicted distance in the M predicted distances as the probability of the corresponding pathological category as the basis for pathological diagnosis, not only the calculation is simple and fast, but also the accuracy of pathological diagnosis and the efficiency of pathological diagnosis are improved.

As shown in Figure 5, in one embodiment, a system for identifying pathological categories based on a distance calculation method is proposed, and the system includes:

The first parameter obtaining module 502 is configured to obtain characteristic parameters to be diagnosed, wherein the characteristic parameters to be diagnosed include N characteristic parameters corresponding to N kinds of features, and N is a natural number;

The second parameter obtaining module 504 is configured to obtain preset standard parameters, where the preset standard parameters include M×N standard parameters of N features corresponding to M types of pathological categories, where M is a natural number;

The calculation module 506 is used to calculate the predicted distance between the characteristic parameter to be diagnosed and the standard parameter corresponding to each pathological category by using a preset distance calculation method, and obtain M predicted distances;

The diagnosis module 508 is configured to determine the pathological category corresponding to the characteristic parameter to be diagnosed according to the size of the M predicted distances.

In one embodiment, the computing module includes:

A distance calculation unit, configured to respectively calculate the first distance and/or the second distance between each of the characteristic parameters and the standard parameters corresponding to the M types of pathological categories, to obtain M×N characteristic distances;

The distance fusion unit is used to fuse and calculate the N characteristic distances corresponding to each pathological category to obtain the M predicted distances.

In one embodiment, the distance fusion unit includes:

The weight acquisition subunit is used to acquire the preset weight corresponding to each feature distance;

The fusion calculation subunit is configured to perform weighted calculation according to the N characteristic distances corresponding to each pathological category and the corresponding preset weights, to obtain the M predicted distances.

In one embodiment, the diagnostic module includes:

a probability calculation unit, configured to determine the proportion of each of the predicted distances in the M predicted distances as the probability of the corresponding pathological category;

A pathological diagnosis unit, configured to determine the pathological category corresponding to the characteristic parameter to be diagnosed according to the probability of each pathological category.

Figure 6 shows an internal structure diagram of a computer device in one embodiment. The computer device may specifically be a server, and the server includes but is not limited to high-performance computers and high-performance computer clusters. As shown in Figure 6, the computer device includes a processor, memory and a network interface connected by a system bus. Wherein, the memory includes a non-volatile storage medium and an internal memory. The non-volatile storage medium of the computer device stores an operating system and also stores a computer program, which, when executed by the processor, enables the processor to implement a method for identifying a pathological category based on a distance calculation method. A computer program may also be stored in the internal memory, and when executed by the processor, the computer program may cause the processor to execute a method for identifying a pathological category based on a distance calculation method. Those skilled in the art can understand that the structure shown in FIG. 6 is only a block diagram of a partial structure related to the solution of the present application, and does not constitute a limitation on the computer equipment to which the solution of the present application is applied. Include more or fewer components than shown in the figures, or combine certain components, or have a different arrangement of components.

In one embodiment, the method for identifying a pathological category based on a distance calculation method provided by the present application can be implemented in the form of a computer program, and the computer program can be executed on a computer device as shown in FIG. 6 . The memory of the computer device can store various program templates that make up the system for identifying pathological categories based on the distance calculation method. For example, the first parameter acquisition module 502 , the second parameter acquisition module 504 , the calculation module 506 , and the diagnosis module 508 .

A computer device, comprising a memory, a processor and a computer program stored in the memory and running on the processor, the processor implements the following steps when executing the computer program: acquiring characteristic parameters to be diagnosed, The to-be-diagnosed feature parameters include N feature parameters corresponding to N kinds of features, where N is a natural number; obtain preset standard parameters, where the preset standard parameters include M×N of N kinds of features corresponding to M kinds of pathological categories respectively standard parameters, M is a natural number; the predicted distance between the characteristic parameter to be diagnosed and the standard parameter corresponding to each pathological category is calculated by using a preset distance calculation method, and M predicted distances are obtained; according to the M predicted distances The size of the distance determines the pathological category corresponding to the feature parameter to be diagnosed.

In one embodiment, calculating the predicted distance between the characteristic parameter to be diagnosed and the standard parameter corresponding to each pathological category by using a preset distance calculation method, to obtain M predicted distances, includes: calculating each The first distance and/or the second distance between the feature parameter and the standard parameters corresponding to the M pathological categories, to obtain M×N feature distances; respectively, perform fusion calculation on the N feature distances corresponding to each pathological category , to obtain M the predicted distances.

In one embodiment, performing fusion calculation on N feature distances corresponding to each pathological category to obtain M predicted distances, including: obtaining a preset weight corresponding to each feature distance; The N characteristic distances corresponding to the pathological categories and the corresponding preset weights are weighted and calculated to obtain the M predicted distances.

In one embodiment, the first distance is at least one of Euclidean distance, Minkowski distance, Manhattan distance and Chebyshev distance, and the second distance is cosine similarity and/or or the Pearson correlation coefficient.

In one embodiment, the acquiring a preset weight corresponding to each feature distance includes: when the feature distance is a first distance, the preset weight is a negative number; when the feature distance is a second distance , the preset weight is a positive number.

In one embodiment, performing fusion calculation on N feature distances corresponding to each pathological category to obtain M predicted distances, including: when the feature distance is a first distance, the predicted distance is negatively correlated with the feature distance; when the feature distance is the second distance, the predicted distance is positively correlated with the feature distance.

In one embodiment, the determining the pathological category corresponding to the feature parameter to be diagnosed according to the size of the M predicted distances includes: determining the proportion of each predicted distance in the M predicted distances as the corresponding pathology The probability of the category; the pathological category corresponding to the characteristic parameter to be diagnosed is determined according to the probability of each pathological category.

A computer-readable storage medium storing a computer program, characterized in that, when the computer program is executed by a processor, the following steps are implemented: acquiring characteristic parameters to be diagnosed, and the characteristic parameters to be diagnosed include: N feature parameters corresponding to N kinds of features, where N is a natural number; obtain preset standard parameters, where the preset standard parameters include M×N standard parameters of N features corresponding to M kinds of pathological categories, where M is a natural number ; Calculate the predicted distances between the characteristic parameters to be diagnosed and the standard parameters corresponding to each pathological category by using a preset distance calculation method, and obtain M predicted distances; Determine the predicted distances according to the size of the M predicted distances The pathological category corresponding to the diagnostic feature parameter.

In one embodiment, calculating the predicted distance between the characteristic parameter to be diagnosed and the standard parameter corresponding to each pathological category by using a preset distance calculation method, to obtain M predicted distances, includes: calculating each The first distance and/or the second distance between the feature parameter and the standard parameters corresponding to the M pathological categories, to obtain M×N feature distances; respectively, perform fusion calculation on the N feature distances corresponding to each pathological category , to obtain M the predicted distances.

In one embodiment, performing fusion calculation on N feature distances corresponding to each pathological category to obtain M predicted distances, including: obtaining a preset weight corresponding to each feature distance; The N characteristic distances corresponding to the pathological categories and the corresponding preset weights are weighted and calculated to obtain the M predicted distances.

In one embodiment, the first distance is at least one of Euclidean distance, Minkowski distance, Manhattan distance and Chebyshev distance, and the second distance is cosine similarity and/or or the Pearson correlation coefficient.

In one embodiment, the acquiring a preset weight corresponding to each feature distance includes: when the feature distance is a first distance, the preset weight is a negative number; when the feature distance is a second distance , the preset weight is a positive number.

In one embodiment, performing fusion calculation on N feature distances corresponding to each pathological category to obtain M predicted distances, including: when the feature distance is a first distance, the predicted distance is negatively correlated with the feature distance; when the feature distance is the second distance, the predicted distance is positively correlated with the feature distance.

In one embodiment, the determining the pathological category corresponding to the feature parameter to be diagnosed according to the size of the M predicted distances includes: determining the proportion of each predicted distance in the M predicted distances as the corresponding pathology The probability of the category; the pathological category corresponding to the characteristic parameter to be diagnosed is determined according to the probability of each pathological category.

Those of ordinary skill in the art can understand that all or part of the processes in the methods of the above embodiments can be implemented by instructing relevant hardware through a computer program, and the program can be stored in a non-volatile computer-readable storage medium , when the program is executed, it may include the flow of the above-mentioned method embodiments. Wherein, any reference to memory, storage, database or other medium used in the various embodiments provided in this application may include non-volatile and/or volatile memory. Nonvolatile memory may include read only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in various forms such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous chain Road (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), etc.

The technical features of the above embodiments can be combined arbitrarily. In order to make the description simple, all possible combinations of the technical features in the above embodiments are not described. However, as long as there is no contradiction in the combination of these technical features It is considered to be the range described in this specification.

The above-mentioned embodiments only represent several embodiments of the present application, and the descriptions thereof are relatively specific and detailed, but should not be construed as a limitation on the scope of the patent of the present application. It should be pointed out that for those skilled in the art, without departing from the concept of the present application, several modifications and improvements can be made, which all belong to the protection scope of the present application. Therefore, the scope of protection of the patent of the present application shall be subject to the appended claims.

Claims (11)

1. A method for identifying a category of pathology based on a distance calculation method, comprising:

acquiring characteristic parameters to be diagnosed, wherein the characteristic parameters to be diagnosed comprise N characteristic parameters corresponding to N characteristics, and N is a natural number;

acquiring preset standard parameters, wherein the preset standard parameters comprise M multiplied by N standard parameters of N characteristics respectively corresponding to M pathological categories, and M is a natural number;

respectively calculating the prediction distances between the characteristic parameters to be diagnosed and the standard parameters corresponding to each pathology category by adopting a preset distance calculation method to obtain M prediction distances;

and determining the pathological type corresponding to the characteristic parameter to be diagnosed according to the size of the M predicted distances.

2. The method for identifying pathological types according to claim 1, wherein the step of calculating the predicted distance between the characteristic parameter to be diagnosed and the standard parameter corresponding to each pathological type by using a preset distance calculation method to obtain M predicted distances comprises:

respectively calculating a first distance and/or a second distance between each characteristic parameter and the standard parameters corresponding to the M pathological categories to obtain M multiplied by N characteristic distances;

and respectively carrying out fusion calculation on the N characteristic distances corresponding to each pathology category to obtain M predicted distances.

3. The method according to claim 2, wherein the fusion calculation of the N characteristic distances corresponding to each pathology category to obtain M predicted distances comprises:

acquiring a preset weight corresponding to each characteristic distance;

and performing weighted calculation according to the N characteristic distances corresponding to each pathology category and the corresponding preset weight to obtain M predicted distances.

4. The method according to claim 2, wherein the first distance is at least one of a euclidean distance, a minkowski distance, a manhattan distance, and a chebyshev distance, and the second distance is a cosine similarity and/or a pearson correlation coefficient.

5. The method for identifying pathological types according to claim 4, wherein the obtaining of the preset weight corresponding to each characteristic distance comprises:

when the characteristic distance is a first distance, the preset weight is a negative number;

and when the characteristic distance is a second distance, the preset weight is a positive number.

6. The method according to claim 4, wherein the fusion calculation of the N characteristic distances corresponding to each pathology category to obtain M predicted distances comprises:

when the feature distance is a first distance, the predicted distance is inversely related to the feature distance;

when the characteristic distance is a second distance, the predicted distance is positively correlated with the characteristic distance.

7. The method for identifying pathological categories according to claim 1, wherein the determining pathological categories corresponding to the feature parameters to be diagnosed according to the magnitude of the M predicted distances includes:

determining the proportion of each predicted distance in the M predicted distances as the probability of the corresponding pathological category;

and determining the pathological type corresponding to the characteristic parameter to be diagnosed according to the probability of each pathological type.

8. A system for identifying a category of pathology based on a distance calculation method, the system comprising:

the system comprises a first parameter acquisition module, a second parameter acquisition module and a parameter analysis module, wherein the first parameter acquisition module is used for acquiring characteristic parameters to be diagnosed, the characteristic parameters to be diagnosed comprise N characteristic parameters corresponding to N characteristics, and N is a natural number;

the second parameter acquisition module is used for acquiring preset standard parameters, wherein the preset standard parameters comprise M multiplied by N standard parameters of N characteristics respectively corresponding to M pathological categories, and M is a natural number;

the calculation module is used for respectively calculating the prediction distances between the characteristic parameters to be diagnosed and the standard parameters corresponding to each pathology category by adopting a preset distance calculation method to obtain M prediction distances;

and the diagnosis module is used for determining the pathological type corresponding to the characteristic parameter to be diagnosed according to the size of the M prediction distances.

9. The system for identifying a pathology category according to claim 8, wherein said calculation module comprises:

the distance calculation unit is used for calculating a first distance and/or a second distance between each characteristic parameter and the standard parameters corresponding to the M pathological categories respectively to obtain M multiplied by N characteristic distances;

and the distance fusion unit is used for respectively carrying out fusion calculation on the N characteristic distances corresponding to each pathology category to obtain M predicted distances.

10. A computer device comprising a memory, a processor and a computer program stored in the memory and executable on the processor, wherein the processor when executing the computer program performs the steps of the method for identifying a category of pathology based on a distance calculation method according to any one of claims 1 to 7.

11. A computer-readable storage medium, in which a computer program is stored which, when being executed by a processor, carries out the steps of the method for identifying a category of pathology based on a distance calculation method according to any one of claims 1 to 7.

Images