CN114372112A

CN114372112A - Empirical prescription data processing method, system, terminal and storage medium based on traditional Chinese medicine names

Info

Publication number: CN114372112A
Application number: CN202111626541.1A
Authority: CN
Inventors: 王元福; 贾声声; 赵旭
Original assignee: Shanghai National Group Health Technology Co ltd
Current assignee: Shanghai National Group Health Technology Co ltd
Priority date: 2021-12-28
Filing date: 2021-12-28
Publication date: 2022-04-19

Abstract

The application provides an empirical prescription data processing method, system, terminal and storage medium based on traditional Chinese medicine names, which comprises the following steps: converting unstructured data into structured data based on traditional Chinese medicine literature; the structured data comprises syndrome data, symptom data and prescription data; clustering different prescription data corresponding to the same syndrome data to form a syndrome-based cluster which takes the syndrome data as a core and is aggregated with a plurality of prescription data; clustering different prescription data corresponding to the same symptom data to form a symptom-based cluster which takes the symptom data as a core and is aggregated with a plurality of prescription data; and performing fusion calculation on the syndrome-based clustering cluster and the symptom-based clustering cluster to generate a prescription data set. The invention selects high-frequency prescription data as a final prescription data set, thereby accelerating the standardization speed in the field of traditional Chinese medicine and assisting the traditional Chinese medicine to save time and energy.

Description

Empirical prescription data processing method, system, terminal and storage medium based on traditional Chinese medicine names

Technical Field

The application relates to the technical field of intelligent medical treatment and machine learning models, in particular to an empirical prescription data processing method, an empirical prescription data processing system, an empirical prescription data processing terminal and a storage medium based on traditional Chinese medicine names.

Background

With the improvement of the living standard of all people, more and more people pay attention to the health status of the people, so that medical treatment is greatly developed. The western medicine has a corresponding medicine for each disease condition, can accurately treat corresponding diseases, and can purchase corresponding medicines for treatment even at home, thereby greatly reducing the medical cost of patients and saving the time and energy of doctors.

However, in the traditional Chinese medicine field, at present, various traditional Chinese medicine data are scattered and cannot form effective association, for example, different prescription data can be correspondingly associated according to the same disease condition, and the use standard of each prescription data is different from person to person, so that the time and energy of the traditional Chinese medicine are greatly consumed, and the payment cost of a patient is increased.

Therefore, how to effectively process prescription data and how to establish an effective data association relationship so as to assist the traditional Chinese medicine to save time and energy of the traditional Chinese medicine becomes a technical problem to be solved urgently in the field.

Disclosure of Invention

In view of the above-mentioned shortcomings of the prior art, the present application aims to provide a method, a system, a terminal and a storage medium for processing empirical prescription data based on traditional Chinese medicine names, so as to solve the problems in the prior art.

To achieve the above and other related objects, a first aspect of the present application provides a method for processing empirical prescription data based on traditional Chinese medicine names, comprising: converting unstructured data into structured data based on traditional Chinese medicine literature; the structured data comprises syndrome data, symptom data and prescription data; clustering different prescription data corresponding to the same syndrome data to form a syndrome-based cluster which takes the syndrome data as a core and is aggregated with a plurality of prescription data; clustering different prescription data corresponding to the same symptom data to form a symptom-based cluster which takes the symptom data as a core and is aggregated with a plurality of prescription data; and performing fusion calculation on the syndrome-based clustering cluster and the symptom-based clustering cluster to generate a prescription data set.

In some embodiments of the first aspect of the present application, after primary clustering is performed on different prescription data corresponding to the same syndrome data, secondary clustering is performed on different syndrome data corresponding to the same prescription data, and the secondary clustering result is used to match the primary clustering result, so as to obtain the complemented syndrome-based clustering cluster.

In some embodiments of the first aspect of the present application, after performing primary clustering on different prescription data corresponding to the same symptom data, performing secondary clustering on different symptom data corresponding to the same prescription data, and matching the primary clustering result with the secondary clustering result to obtain the complemented symptom-based clustering cluster.

In some embodiments of the first aspect of the present application, performing a fusion calculation on the syndrome-based cluster and the symptom-based cluster generates a corresponding prescription data set, which includes: searching an association relation between symptoms and syndrome types based on a preset knowledge graph; acquiring a corresponding clustering cluster based on the syndrome, and acquiring a corresponding clustering cluster based on the symptom; and counting the medicinal materials and/or added and subtracted medicines which appear at high frequency according to the syndrome-based clustering cluster and the symptom-based clustering cluster to form the prescription data set.

In some embodiments of the first aspect of the present application, a generation manner of each of the cluster clusters includes: firstly, inputting the structured data into a coding end of a self-coder for coding; and inputting the coded data into a clustering module for clustering to obtain a cluster.

In some embodiments of the first aspect of the present application, the clustering module comprises a coarse clustering module and a fine clustering module; the rough clustering module is used for determining the number of clustering centers and clustering center points; and the fine clustering module is used for performing fine clustering on the basis of the determined number of the clustering centers and the clustering center points to obtain clustering clusters.

In some embodiments of the first aspect of the present application, the process of determining the number of cluster centers and the cluster center point by the rough clustering module includes: firstly, determining the number of clustering centers, and then initializing the position of the central point of each clustering center; the initialization mode of the central point position of each clustering center comprises the following steps: randomly selecting a data point from the data set input into the rough clustering module as a temporary clustering center; calculating a distance value between each data point in the data set and the temporary clustering center; selecting a data point corresponding to the maximum distance value as a new clustering center; and repeating the steps until all the clustering centers with the same number as the clustering centers are selected, and taking the clustering centers as the initialized center point positions.

To achieve the above and other related objects, a second aspect of the present application provides an empirical prescription data processing system based on traditional Chinese medicine names, comprising: a structuring module for converting unstructured data into structured data based on traditional Chinese medicine literature; the structured data comprises syndrome data, symptom data and prescription data; the syndrome type clustering module is used for clustering different prescription data corresponding to the same syndrome type data to form a syndrome type-based clustering cluster which takes the syndrome type data as a core and is aggregated with a plurality of prescription data; the symptom clustering module is used for clustering different prescription data corresponding to the same symptom data to form a symptom-based clustering cluster which takes the symptom data as a core and is aggregated with a plurality of prescription data; and the fusion module is used for performing fusion calculation on the syndrome-based clustering cluster and the symptom-based clustering cluster to generate a prescription data set.

To achieve the above and other related objects, a third aspect of the present application provides a computer-readable storage medium having stored thereon a computer program which, when executed by a processor, implements the method for processing empirical prescription data based on traditional Chinese medicine names.

To achieve the above and other related objects, a fourth aspect of the present application provides an electronic terminal comprising: a processor and a memory; the memory is used for storing computer programs, and the processor is used for executing the computer programs stored by the memory so as to enable the terminal to execute the empirical prescription data processing method based on traditional Chinese medicine names.

As described above, the empirical prescription data processing method, system, terminal and storage medium based on traditional Chinese medicine names of the present application have the following beneficial effects: the invention adopts a deep clustering method, and establishes corresponding syndrome cluster and symptom cluster based on the incidence relation between syndrome data and prescription data and between symptom data and prescription data; and performing fusion calculation on the two clustering clusters, and selecting high-frequency prescription data as a final prescription data set, so that the standardization speed in the field of traditional Chinese medicine is increased, and the traditional Chinese medicine is assisted to save time and energy.

Drawings

FIG. 1 is a flow chart illustrating a method for processing empirical prescription data based on the names of traditional Chinese medicines according to an embodiment of the present application.

Fig. 2 is a schematic structural diagram of an auto-encoder according to an embodiment of the present application.

FIG. 3 is a block diagram of an exemplary empirical prescription data processing system based on traditional Chinese medicine names.

Fig. 4 is a schematic structural diagram of an electronic terminal according to an embodiment of the present application.

Detailed Description

The following description of the embodiments of the present application is provided by way of specific examples, and other advantages and effects of the present application will be readily apparent to those skilled in the art from the disclosure herein. The present application is capable of other and different embodiments and its several details are capable of modifications and/or changes in various respects, all without departing from the spirit of the present application. It is to be noted that the features in the following embodiments and examples may be combined with each other without conflict.

It should be noted that in the following description, as used herein, the singular forms "a," "an," and "the" are intended to include the plural forms as well, unless the context indicates otherwise. It will be further understood that the terms "comprises," "comprising," and/or "comprising," when used in this specification, specify the presence of stated features, operations, elements, components, items, species, and/or groups, but do not preclude the presence, or addition of one or more other features, operations, elements, components, items, species, and/or groups thereof. The terms "or" and/or "as used herein are to be construed as inclusive or meaning any one or any combination. Thus, "A, B or C" or "A, B and/or C" means "any of the following: a; b; c; a and B; a and C; b and C; A. b and C ". An exception to this definition will occur only when a combination of elements, functions or operations are inherently mutually exclusive in some way.

In order to solve the problems in the prior art, the invention aims to excavate the increase and decrease change rules of different prescriptions for the same disease condition based on a depth clustering algorithm in the field of artificial intelligence, and form a universal prescription for different people, thereby not only helping doctors save more time and energy, but also saving treatment cost for patients.

In order to make the objects, technical solutions and advantages of the present invention more apparent, the technical solutions in the embodiments of the present invention are further described in detail by the following embodiments in conjunction with the accompanying drawings. It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.

It should be explained that, for the professional vocabularies in the technical fields of traditional Chinese medicine and artificial intelligence, etc., the following explanations are made:

syndrome type: it refers to the result of qualitative classification of the relatively stable syndrome expressed at a certain stage in the disease process under the guidance of the syndrome differentiation theory of traditional Chinese medicine. For example, the eight principles can be classified into yin syndrome, yang syndrome, exterior syndrome, interior syndrome, cold syndrome, heat syndrome, deficiency syndrome and excess syndrome; for example, the syndrome of qi-blood can be classified into qi-deficiency syndrome, qi-stagnation syndrome, qi-adverse syndrome, qi-sinking syndrome, blood-deficiency syndrome, blood-heat syndrome, and blood stasis syndrome; the classification can also be made according to the differentiation of syndromes of six meridians, the differentiation of syndromes of wei-defensive qi, ying-nutrient blood, the differentiation of syndromes of zang-fu organs, etc., which are not listed one by one.

Symptoms are: refers to the subjective abnormal feeling or some objective pathological changes of the patient caused by a series of abnormal changes of function, metabolism and morphological structure in the body in the disease process; it mainly refers to the symptoms of TCM, such as: aversion to cold, fever, night sweat, fatigue, sallow complexion, bitter taste in the mouth, dry tongue, etc., which are not listed one by one.

An auto-encoder: the artificial neural network is used in semi-supervised learning and unsupervised learning, and has the function of performing characterization learning on input information by taking the input information as a learning target. Generally, a self-encoder includes two parts, an encoder (encoder) and a decoder (decoder); the auto-encoder can learn the implicit features of the input data, which is called encoding, while reconstructing the original input data with the learned new features, which is called decoding.

Example one

Fig. 1 is a schematic flow chart showing a method for processing empirical prescription data based on the names of traditional Chinese medicines according to an embodiment of the present invention.

It should be noted that the empirical prescription data processing method based on the traditional Chinese medicine name in the embodiment can be applied to various hardware devices; for example, the present invention is applied to an arm (advanced RISC machines) controller, an fpga (field Programmable Gate array) controller, an soc (system on chip) controller, a dsp (digital Signal processing) controller, or an mcu (micro controller unit) controller; the method is also applied to personal computers such as desktop computers, notebook computers, tablet computers, smart phones, smart bracelets, smart watches, smart helmets, smart televisions and the like; the present invention is also applicable to, for example, a server, where the server may be arranged on one or more entity servers according to various factors such as functions and loads, or may be formed by a distributed or centralized server cluster, and this embodiment is not limited.

It should be noted that the empirical prescription data processing method based on the traditional Chinese medicine name provided by the invention is essentially a data processing method rather than a disease diagnosis method, and the execution subject of the empirical prescription data processing method is computer equipment rather than a disease diagnosis method suitable for individuals such as doctors.

In this embodiment, the empirical prescription data processing method based on traditional Chinese medicine names mainly comprises steps S11 to S14, and the following steps will be described in detail with reference to specific embodiments and drawings of the specification.

Step S11: converting unstructured data into structured data based on traditional Chinese medicine literature; the structured data includes syndrome data, prescription data, and symptom data.

The data of the present invention are derived from traditional Chinese medicine documents such as ancient medical books and classical books of modern traditional Chinese medicine, for example, Shen nong Ben Cao Jing, Huang Di Nei Jing Su, Shang Han Bing Lun, Ben Cao gang mu, etc., and the present embodiment is not limited thereto.

In some examples, the converting unstructured data into structured data based on the literature of traditional Chinese medicine comprises: based on a semantic analysis algorithm (such as an NLP (non-line segment) semantic analysis algorithm and the like), extracting various traditional Chinese medicine elements from unstructured data in traditional Chinese medicine documents, and preprocessing the traditional Chinese medicine elements to form structured data; the Chinese medicinal elements comprise symptoms, syndrome types and prescriptions, wherein the prescriptions at least comprise prescription components, prescription dosages and the like. Accordingly, the structured data includes at least syndrome data, prescription data, symptom data, and the like, accordingly.

Further, the above-mentioned preprocessing of the data of the elements of traditional Chinese medicine mainly includes preprocessing modes such as data cleaning, data integration, data change, data reduction, etc. Data cleaning mainly refers to cleaning data by filling in missing values, smoothing noisy data, identifying or deleting outliers and solving inconsistency, and is used for achieving the purposes of format standardization, abnormal data cleaning, error correction, repeated data cleaning and the like. Data integration mainly refers to combining and uniformly storing data in a plurality of data sources. Data transformation mainly refers to converting data into a form suitable for data mining by means of smooth aggregation, data generalization, data normalization and the like. Data reduction mainly means that the data volume is very large when data is mined, long time is needed for mining analysis on a small amount of data, and data reduction can be used for obtaining reduction representation of a data set so as to shorten analysis time on the premise of keeping data integrity. It should be understood that the above data pre-processing examples are provided for illustrative purposes, but should not be construed as limiting.

Step S12: clustering different prescription data corresponding to the same syndrome data to form a syndrome-based cluster which takes the syndrome data as a core and is aggregated with a plurality of prescription data.

In some examples, the manner of clustering different prescription data corresponding to the same syndrome data includes, but is not limited to, a traditional clustering method and a deep learning clustering method; the traditional clustering methods include, for example, a K-means clustering method, a graph-based clustering method, a density-based clustering method, a network-based clustering method, a hierarchical clustering method, or a fuzzy-based clustering method; the Deep learning Clustering method includes a Deep learning Clustering algorithm based on an auto-encoder, and specific examples include a dcn (Deep Clustering network) Deep learning Clustering algorithm, a den (Deep Embedding network) Deep learning Clustering algorithm, a dsc (Deep sub Clustering network) Deep learning Clustering algorithm, a DMC (Deep Multi-modified Clustering) Deep learning Clustering algorithm, and the like. Taking the DCN deep learning algorithm as an example, the clustering mode is to combine an automatic encoder and a k-means algorithm, train an automatic encoder in advance and then optimize the reconstruction loss and the k-means loss.

Exemplarily, the self-encoder in this embodiment may be divided into an encoding end, a hidden layer, and a decoding end, the structured data is used as input data, the high-dimensional data is encoded into low-dimensional data through the self-encoder after passing through the neural network layer, and then decoded and restored into original data, and the hidden layer structure is that the number of hidden nodes of each hidden layer is 10, 5, 2, 5, and 10; clustering is performed by using data at layer 3 in the hidden layer, and AE in AE clustering can be regarded as a regular rule for restricting features to reduce degradation risk.

In some examples, after primary clustering is performed on different prescription data corresponding to the same syndrome data, secondary clustering is performed on different syndrome data corresponding to the same prescription data, and the secondary clustering result is used for matching the primary clustering result to obtain a completed syndrome-based clustering cluster. In the secondary clustering, the prescription data is used as a clustering center, the syndrome data is used as clustering data, the clustering data is input into a self-encoder to be encoded and decoded, the fourth layer data in the hiding is input into a clustering model to be clustered, and a data set taking the prescription as a core is obtained.

Specifically, because the syndrome is used as the data of the central cluster, a prescription corresponding to one syndrome may be omitted; therefore, in this example, the syndrome data is used as the center to perform primary clustering on the different prescription data surrounding the syndrome data, and then the prescription data is used as the center to perform secondary clustering on the different syndrome data surrounding the prescription data; matching the secondary clustering result with the primary clustering result, and if the secondary clustering result is completely matched with the primary clustering result, indicating that the primary clustering result is complete; if the two clustering results are not completely matched, the primary clustering result is not complete, the missing items are added into the primary clustering result through the secondary clustering result, and finally, complete whole clustering data taking the syndrome as the center are obtained.

In some examples, the generation manner of the syndrome-based cluster includes: firstly, inputting the structured data into a coding end of a self-coder for coding; and inputting the coded data into a clustering module for clustering to obtain a syndrome-based clustering cluster.

Further, the clustering module comprises a coarse clustering module and a fine clustering module; the rough clustering module is used for determining the number of clustering centers and clustering center points; and the fine clustering module is used for performing fine clustering on the basis of the determined number of the clustering centers and the clustering center points to obtain clustering clusters.

The process of determining the number of the clustering centers and the clustering center points by the rough clustering module comprises the following steps: firstly, determining the number of clustering centers, and then initializing the position of the central point of each clustering center; the initialization mode of the central point position of each clustering center comprises the following steps: randomly selecting a data point from the data set input into the rough clustering module as a temporary clustering center; calculating a distance value between each data point in the data set and the temporary clustering center; selecting a data point corresponding to the maximum distance value as a new clustering center; and repeating the steps until all the clustering centers with the same number as the clustering centers are selected, and taking the clustering centers as the initialized center point positions.

For example, the coarse clustering module may use the Canopy clustering module and the fine clustering module may use the K-means clustering module. Therefore, the Canopy algorithm can be used to select the K value in the K-means clustering module, which determines the initial number of clustering centers and the clustering center point for the K-means algorithm by rough clustering in advance. It is worth noting that, different from the traditional clustering algorithm, the Canopy algorithm has the greatest characteristic that the k value (namely the number of clusters) does not need to be specified in advance, so that the Canopy algorithm has great practical application value; however, the conventional clustering algorithm needs to determine the K value, which affects the processing speed, so that the Canopy algorithm is superior in processing speed, and therefore, in the embodiment, the Canopy algorithm is used for performing coarse clustering on data to obtain the K value and approximate K central points, and then the K-means algorithm is used for further performing fine clustering. Therefore, the clustering algorithm in the form of Canopy + K-means adopted by the embodiment has obvious advantages in clustering speed and precision.

Step S13: clustering different prescription data corresponding to the same symptom data to form a symptom-based cluster which takes the symptom data as a core and gathers a plurality of prescription data.

In some examples, after the first clustering of the different prescription data corresponding to the same symptom data, the second clustering of the different symptom data corresponding to the same prescription data is performed, and the symptom-based clustering cluster is completed by matching the second clustering result with the first clustering result. In the secondary clustering, the prescription data is used as a clustering center, the symptom data is used as clustering data, the clustering data is input into a self-encoder to be encoded and decoded, the fourth layer data in the hiding is input into a clustering model to be clustered, and a data set taking the prescription as a core is obtained.

Specifically, since the data of clustering is used as a center of symptoms, a prescription corresponding to one symptom may be omitted; therefore, in this example, the symptom data is first clustered around the different prescription data, and then the prescription data is used as the center to perform secondary clustering around the different symptom data; matching the secondary clustering result with the primary clustering result, and if the secondary clustering result is completely matched with the primary clustering result, indicating that the primary clustering result is complete; if the two clustering results are not completely matched, the primary clustering result is not complete, missing items are added into the primary clustering result through the secondary clustering result, and finally, complete all clustering data taking symptoms as centers are obtained.

In some examples, the manner in which the symptom-based cluster is generated includes: firstly, inputting the structured data into a coding end of a self-coder for coding; and inputting the coded data into a clustering module for clustering to obtain a symptom-based clustering cluster.

Step S14: and performing fusion calculation on the syndrome-based clustering cluster and the symptom-based clustering cluster to generate a corresponding prescription data set.

In some examples, the process of performing a fusion calculation on the syndrome-based cluster and the symptom-based cluster includes:

firstly, the incidence relation between symptoms and syndrome types is searched based on a preset knowledge graph.

For example, based on the created knowledge graph, the corresponding relationship between symptoms and syndromes is obtained by querying the knowledge graph, for example, syndromes are kidney-yang deficiency, decline of fire from the gate of life, etc., the corresponding symptoms are lassitude, etc., and the corresponding medicine prescriptions are six-ingredient rehmannia pills, zuogui pills, dugui pills, etc. by querying.

Secondly, acquiring a corresponding syndrome-based cluster based on syndrome, and acquiring a corresponding symptom-based cluster based on symptom.

And finally, counting the medicinal materials and/or added and subtracted medicines which appear at high frequency according to the syndrome-based clustering cluster and the symptom-based clustering cluster to form the prescription data set. It should be understood that the term "high frequency" as used herein refers to a condition where the frequency of occurrence exceeds a certain threshold, the threshold is also set by a user, a default value may also be set, and the present embodiment is not limited thereto.

Example two

The present embodiment will further explain the processing procedure of empirical prescription data based on traditional Chinese medicine names, which is as follows:

extracting initial data from the traditional Chinese medicine literature, performing preprocessing such as data cleaning on the initial data, mining to obtain structured text data, inputting the structured text data into an encoding end of a self-encoder to perform encoding operation, and using a hidden layer (for example, the layer 3 can be used in the embodiment) as data input by clustering. One line of data is one sample (sample), 300 samples are in one batch (batch), and 300 rounds of training (epoch) are performed.

The self-encoder in this embodiment is an unsupervised learning technique, and performs characterization learning by using a neural network. That is, it is equivalent to setting a barrier in a network, forcing the original input to compress the neural network architecture represented by the knowledge, but if there is correlation between the input features, it is possible to learn the structure.

The structure of the self-encoder is shown in fig. 2: input multiple dimensionsAccording to [ X1, X2, X3 … X10]The dimension reduction treatment is changed into [ h1, h2 … h5 ]]Inputting the reduced data into a clustering layer for clustering to form a clustering result a1 and a clustering result a2, wherein the clustering result is low-dimensional data

Outputting, and then performing dimension-raising processing on the low-dimensional data to obtain

This is roughly the case for the process of the self-encoder. The unmarked data set and the frame can be taken as a task supervision learning problem to be responsible for outputting new data

I.e. the reconstruction of the original input X. This network can be implemented by minimizing reconstruction errors

Training, by minimizing reconstruction errors, is meant a measure of the difference between the original input and the reconstruction.

In some examples, an auto-encoder is used to obtain an efficient data representation, so after training is completed, the decoder is preferably removed, i.e. only the encoder is kept, and the output of the encoder can be directly used as the input of the subsequent machine learning model, so that the third layer of the hidden layer is used as the input of the cluster.

In some examples, the syndrome-based clustering cluster is formed by taking the syndrome as a clustering key point and taking prescription details as clustering features, and finding similar symptom sets to form prescription sets of similar symptom sets, wherein the prescription sets can be connected by using a K-means clustering module, and the third layer of the self-encoder is used as a follow-up of the self-encoder.

Further, the value of K in the K-means clustering module can be selected by using a Canopy algorithm, and the initial clustering center number and the clustering center point are determined for the K-means algorithm in a rough clustering mode in advance. It is worth noting that, different from the traditional clustering algorithm, the Canopy algorithm has the greatest characteristic that the k value (namely the number of clusters) does not need to be specified in advance, so that the Canopy algorithm has great practical application value; however, the conventional clustering algorithm needs to determine the K value, which affects the processing speed, so that the Canopy algorithm is superior in processing speed, and therefore, in the embodiment, the Canopy algorithm is used for performing coarse clustering on data to obtain the K value and approximate K central points, and then the K-means algorithm is used for further performing fine clustering. Therefore, the clustering algorithm in the form of Canopy + K-means adopted by the embodiment has obvious advantages in clustering speed and precision.

The principle and process of the clustering algorithm in the form of Canopy + K-means employed in this embodiment are described as follows:

step 1: the initial data set L is sorted according to a certain rule (the rule can be arbitrary, for example, the formula medicinal material corresponding to the syndrome type and symptoms appears according to the frequency as a main sorting rule), and once the rule is determined, the rule is not changed. Meanwhile, setting initial distance thresholds T1 and T2, and enabling T1 to be larger than T2.

Step 2: randomly selecting a sample data vector A from the initial data set L, and calculating the distance d between the sample data vector A and other sample data vectors in the initial data set L by using a distance calculation method.

And step 3: according to the distance d, sample data vectors with the distance d smaller than the distance threshold T1 are divided into a Canopy list, and the sample data vectors with the distance d smaller than the distance threshold T2 are removed from the initial data set L.

And 4, step 4: and (5) repeating the steps 2 and 3 until the initial data set L is empty, and ending the algorithm.

Further, considering that the position of the initialized centroid has a great influence on the clustering result and the clustering time, after the K value is determined, the initialized centroid is correspondingly optimized as follows:

step 1: in the input initial data set L, a point is randomly selected as a cluster center a 1.

Step 2: based on each point X in the initial dataset L, the distance between it and the cluster center a1 is calculated, which can be expressed as:

A(X)＝argmin‖Xi-μr‖22，r＝1,2,…k selected.

and step 3: the point with the largest A (X) is taken as the new cluster center.

And 4, step 4: and (3) taking the currently selected new clustering center as a centroid, removing the new clustering center, repeating the steps 2 and 3 on the initial data set until K centroids are selected, and finally taking the K centroids as initialized centroids.

Further, after completing the centroid initialization, the following steps are also executed: calculating the distance between each datum and the K clustering centers by using the Euclidean distance; dividing the data with the closest distance into a class, and determining a clustering cluster; recalculating the intra-cluster distance and updating the cluster center; and repeating the steps until the clustering center is not changed, and finally obtaining all clustering clusters. The method is used for respectively obtaining the clustering cluster based on the syndrome type and the clustering cluster based on the symptom.

And finally, performing fusion calculation on the basis of the syndrome-based clustering cluster and the symptom-based clustering cluster, and counting high-frequency co-occurring medicinal materials and high-frequency plus-minus prescriptions.

EXAMPLE III

Fig. 3 is a schematic structural diagram of an empirical prescription data processing system based on traditional Chinese medicine names according to an embodiment of the present invention. The empirical prescription data processing system 300 in this embodiment includes: a structuring module 301, a syndrome clustering module 302, a symptom clustering module 303, and a fusion module 304.

The structuring module 301 is configured to convert unstructured data into structured data based on the literature of traditional Chinese medicine; the structured data includes syndrome data, symptom data, and prescription data. The syndrome clustering module 302 is configured to cluster different prescription data corresponding to the same syndrome data to form a syndrome-based cluster that takes the syndrome data as a core and aggregates a plurality of prescription data. The symptom clustering module 303 is configured to cluster different prescription data corresponding to the same symptom data to form a symptom-based cluster that takes the symptom data as a core and gathers a plurality of prescription data. The fusion module 304 is configured to perform fusion calculation on the syndrome-based cluster and the symptom-based cluster to generate a prescription data set.

It should be noted that the above-mentioned empirical prescription data processing system based on traditional Chinese medicine names is similar to the empirical prescription data processing system based on traditional Chinese medicine names, and therefore, the detailed description thereof is omitted.

It should be understood that the division of the modules of the above system is only a logical division, and the actual implementation may be wholly or partially integrated into one physical entity or may be physically separated. And these modules can be realized in the form of software called by processing element; or may be implemented entirely in hardware; and part of the modules can be realized in the form of calling software by the processing element, and part of the modules can be realized in the form of hardware. For example, the fusion module may be a processing element that is set up separately, or may be implemented by being integrated in a chip of the system, or may be stored in a memory of the system in the form of program code, and the function of the fusion module may be called and executed by a processing element of the system. Other modules are implemented similarly. In addition, all or part of the modules can be integrated together or can be independently realized. The processing element described herein may be an integrated circuit having signal processing capabilities. In implementation, each step of the above method or each module above may be implemented by an integrated logic circuit of hardware in a processor element or an instruction in the form of software.

For example, the above modules may be one or more integrated circuits configured to implement the above methods, such as: one or more Application Specific Integrated Circuits (ASICs), or one or more microprocessors (DSPs), or one or more Field Programmable Gate Arrays (FPGAs), among others. For another example, when one of the above modules is implemented in the form of a Processing element scheduler code, the Processing element may be a general-purpose processor, such as a Central Processing Unit (CPU) or other processor capable of calling program code. For another example, these modules may be integrated together and implemented in the form of a system-on-a-chip (SOC).

Fig. 4 is a schematic structural diagram of an electronic terminal according to an embodiment of the present invention. This example provides an electronic terminal, includes: a processor 41, a memory 42, a communicator 43; the memory 42 is connected with the processor 41 and the communicator 43 through a system bus and is used for completing mutual communication, the memory 42 is used for storing computer programs, the communicator 43 is used for communicating with other equipment, and the processor 41 is used for running the computer programs so as to enable the electronic terminal to execute the steps of the empirical prescription data processing method based on traditional Chinese medicine names.

The above-mentioned system bus may be a Peripheral Component Interconnect (PCI) bus, an Extended Industry Standard Architecture (EISA) bus, or the like. The system bus may be divided into an address bus, a data bus, a control bus, and the like. For ease of illustration, only one thick line is shown, but this does not mean that there is only one bus or one type of bus. The communication interface is used for realizing communication between the database access device and other equipment (such as a client, a read-write library and a read-only library). The Memory may include a Random Access Memory (RAM), and may further include a non-volatile Memory (non-volatile Memory), such as at least one disk Memory.

The Processor may be a general-purpose Processor, and includes a Central Processing Unit (CPU), a Network Processor (NP), and the like; the Integrated Circuit may also be a Digital Signal Processor (DSP), an Application Specific Integrated Circuit (ASIC), a Field Programmable Gate Array (FPGA) or other Programmable logic device, a discrete Gate or transistor logic device, or a discrete hardware component.

The invention also provides a computer readable storage medium, on which a computer program is stored, which, when executed by a processor, implements the method for processing empirical prescription data based on traditional Chinese medicine names.

Those of ordinary skill in the art will understand that: all or part of the steps for implementing the above method embodiments may be performed by hardware associated with a computer program. The aforementioned computer program may be stored in a computer readable storage medium. When executed, the program performs steps comprising the method embodiments described above; and the aforementioned storage medium includes: various media that can store program codes, such as ROM, RAM, magnetic or optical disks.

In the embodiments provided herein, the computer-readable and writable storage medium may include read-only memory, random-access memory, EEPROM, CD-ROM or other optical disk storage, magnetic disk storage or other magnetic storage devices, flash memory, a USB flash drive, a removable hard disk, or any other medium that can be used to store desired program code in the form of instructions or data structures and that can be accessed by a computer. Also, any connection is properly termed a computer-readable medium. For example, if the instructions are transmitted from a website, server, or other remote source using a coaxial cable, fiber optic cable, twisted pair, Digital Subscriber Line (DSL), or wireless technologies such as infrared, radio, and microwave, then the coaxial cable, fiber optic cable, twisted pair, DSL, or wireless technologies such as infrared, radio, and microwave are included in the definition of medium. It should be understood, however, that computer-readable-writable storage media and data storage media do not include connections, carrier waves, signals, or other transitory media, but are intended to be non-transitory, tangible storage media. Disk and disc, as used in this application, includes Compact Disc (CD), laser disc, optical disc, Digital Versatile Disc (DVD), floppy disk and blu-ray disc where disks usually reproduce data magnetically, while discs reproduce data optically with lasers.

In summary, the application provides an empirical prescription data processing method, system, terminal and storage medium based on traditional Chinese medicine names, a deep clustering method is adopted, and corresponding syndrome cluster and symptom cluster are established based on the incidence relation between syndrome data and prescription data and between symptom data and prescription data; and performing fusion calculation on the two clustering clusters, and selecting high-frequency prescription data as a final prescription data set, so that the standardization speed in the field of traditional Chinese medicine is increased, and the traditional Chinese medicine is assisted to save time and energy. Therefore, the application effectively overcomes various defects in the prior art and has high industrial utilization value.

The above embodiments are merely illustrative of the principles and utilities of the present application and are not intended to limit the application. Any person skilled in the art can modify or change the above-described embodiments without departing from the spirit and scope of the present application. Accordingly, it is intended that all equivalent modifications or changes which can be made by those skilled in the art without departing from the spirit and technical concepts disclosed in the present application shall be covered by the claims of the present application.

Claims

1. An empirical prescription data processing method based on traditional Chinese medicine names is characterized by comprising the following steps:

converting unstructured data into structured data based on traditional Chinese medicine literature; the structured data comprises syndrome data, symptom data and prescription data;

clustering different prescription data corresponding to the same syndrome data to form a syndrome-based cluster which takes the syndrome data as a core and is aggregated with a plurality of prescription data;

clustering different prescription data corresponding to the same symptom data to form a symptom-based cluster which takes the symptom data as a core and is aggregated with a plurality of prescription data;

and performing fusion calculation on the syndrome-based clustering cluster and the symptom-based clustering cluster to generate a prescription data set.

2. The method for processing empirical prescription data based on traditional Chinese medicine names according to claim 1, comprising:

and after primary clustering is carried out on different prescription data corresponding to the same prescription data, secondary clustering is carried out on different prescription data corresponding to the same prescription data, and the secondary clustering result is matched with the primary clustering result to obtain the complemented syndrome-based clustering cluster.

3. The method for processing empirical prescription data based on traditional Chinese medicine names according to claim 1, comprising:

and after the primary clustering is carried out on the different prescription data corresponding to the same symptom data, the secondary clustering is carried out on the different symptom data corresponding to the same prescription data, and the secondary clustering result is matched with the primary clustering result to obtain the complemented symptom-based clustering cluster.

4. The method for processing empirical prescription data based on traditional Chinese medicine names according to claim 1, wherein fusion calculation is performed on the syndrome-based cluster and the symptom-based cluster to generate a corresponding prescription data set, and the method comprises the following steps:

searching an association relation between symptoms and syndrome types based on a preset knowledge graph;

acquiring a corresponding clustering cluster based on the syndrome, and acquiring a corresponding clustering cluster based on the symptom;

and counting the medicinal materials and/or added and subtracted medicines which appear at high frequency according to the syndrome-based clustering cluster and the symptom-based clustering cluster to form the prescription data set.

5. The empirical prescription data processing method based on traditional Chinese medicine names of claim 1, wherein the generation mode of each cluster comprises: firstly, inputting the structured data into a coding end of a self-coder for coding; and inputting the coded data into a clustering module for clustering to obtain a cluster.

6. The empirical prescription data processing method based on traditional Chinese medicine names of claim 5, wherein the clustering module comprises a coarse clustering module and a fine clustering module; the rough clustering module is used for determining the number of clustering centers and clustering center points; and the fine clustering module is used for performing fine clustering on the basis of the determined number of the clustering centers and the clustering center points to obtain clustering clusters.

7. The method for processing empirical prescription data based on traditional Chinese medicine names of claim 6, wherein the process of determining the number of clustering centers and the clustering center point by the rough clustering module comprises: firstly, determining the number of clustering centers, and then initializing the position of the central point of each clustering center; the initialization mode of the central point position of each clustering center comprises the following steps:

randomly selecting a data point from the data set input into the rough clustering module as a temporary clustering center;

calculating a distance value between each data point in the data set and the temporary clustering center;

selecting a data point corresponding to the maximum distance value as a new clustering center;

and repeating the steps until all the clustering centers with the same number as the clustering centers are selected, and taking the clustering centers as the initialized center point positions.

8. An empirical prescription data processing system based on traditional Chinese medicine names, comprising:

a structuring module for converting unstructured data into structured data based on traditional Chinese medicine literature; the structured data comprises syndrome data, symptom data and prescription data;

the syndrome type clustering module is used for clustering different prescription data corresponding to the same syndrome type data to form a syndrome type-based clustering cluster which takes the syndrome type data as a core and is aggregated with a plurality of prescription data;

the symptom clustering module is used for clustering different prescription data corresponding to the same symptom data to form a symptom-based clustering cluster which takes the symptom data as a core and is aggregated with a plurality of prescription data;

and the fusion module is used for performing fusion calculation on the syndrome-based clustering cluster and the symptom-based clustering cluster to generate a prescription data set.

9. A computer-readable storage medium, on which a computer program is stored, wherein the computer program, when executed by a processor, implements the method for processing empirical prescription data based on traditional Chinese medicine names according to any one of claims 1 to 7.

10. An electronic terminal, comprising: a processor and a memory;

the memory is used for storing a computer program;

the processor is used for executing the computer program stored in the memory so as to enable the terminal to execute the empirical prescription data processing method based on traditional Chinese medicine names according to any one of claims 1 to 7.