A method, device, equipment and storage medium for auxiliary determination of efficacy of traditional Chinese medicine prescription

Abstract

The present invention discloses a method, device, equipment and storage medium for auxiliary determination of the efficacy of a Chinese medicine prescription, the method comprising receiving the input of a preset Chinese medicine prescription; generating a drug component vector sequence according to the drug components and dosage in the preset Chinese medicine prescription, and setting the efficacy and toxic and side effect labels of each of the drug components respectively; inputting the drug component vector sequence and the efficacy and toxic and side effect labels of each of the drug components into a Chinese medicine prescription evaluation model and processing them, wherein the Chinese medicine prescription evaluation model comprises a Chinese medicine prescription encoder and a Rank‑SVM multi-label linear classifier; outputting the efficacy and toxic and side effect labels of the preset Chinese medicine prescription as a basis for auxiliary determination of the efficacy of the preset Chinese medicine prescription. The above-mentioned method, device, equipment and storage medium for auxiliary determination of the efficacy of a Chinese medicine prescription can assist Chinese medicine practitioners in ensuring the rationality of Chinese medicine prescriptions, reducing toxic and side effects, improving the efficacy and curative effect of Chinese medicine, and helping patients recover their health as soon as possible.

Description

Technical Field

[0001] The present invention belongs to the field of deep learning technology, and in particular relates to a method, device, equipment and storage medium for auxiliary determination of the efficacy of a traditional Chinese medicine prescription. Background Art

[0002] In recent years, artificial intelligence technology has developed rapidly, which will bring disruptive changes to the entire society and has become an important development strategy for all countries in the future. In particular, the evolution of algorithms centered on deep learning has super evolutionary capabilities. With the support of big data, it can build a large-scale neural network similar to the structure of the human brain through training, which can solve a variety of problems. Among them, the attention mechanism is a data processing method in machine learning, which is widely used in various types of tasks such as natural language processing, image recognition and speech recognition.

[0003] Traditional Chinese medicine is a treasure accumulated by the ancients in long-term medical practice. It is an important part of my country's ancient excellent cultural heritage. Traditional Chinese medicine is mostly plant-based and is the most commonly used. With the development of modern natural science and technology and the national economy, traditional Chinese medicine has also made unprecedented achievements. After long-term medical practice, a variety of drugs are combined and boiled into soups to form a traditional Chinese medicine prescription. Traditional Chinese medicine doctors prescribe medicines according to the actual conditions of patients. To a certain extent, the treatment effect of patients depends on the experience of traditional Chinese medicine doctors, especially some drug combinations will produce toxic side effects. This requires comprehensive consideration of many factors to judge the efficacy of traditional Chinese medicine prescriptions. Often, a prescription is composed of more than a dozen or even dozens of drugs. Therefore, its efficacy is not a simple superposition of medicinal components. In this case, how to more reasonably and accurately judge the efficacy and toxic side effects of traditional Chinese medicine prescriptions has become an urgent problem to be solved. Summary of the invention

[0004] To solve the above problems, the present invention provides a method, device, equipment and storage medium for auxiliary determination of the efficacy of Chinese medicine prescriptions, which can assist Chinese medicine practitioners in ensuring the rationality of Chinese medicine prescriptions, reducing toxic side effects, improving the efficacy and curative effect of Chinese medicine, and helping patients recover their health as soon as possible.

[0005] The present invention provides a method for auxiliary determination of the efficacy of a traditional Chinese medicine prescription, comprising:

[0006] Receiving input of a preset Chinese medicine prescription;

[0007] Generate a drug component vector sequence according to the drug components and dosages in the preset Chinese medicine prescription, and set the efficacy and toxic and side effect labels of each drug component respectively;

[0008] Inputting the drug component vector sequence and the efficacy and toxic side effect labels of each drug component into a traditional Chinese medicine prescription evaluation model and processing them, wherein the traditional Chinese medicine prescription evaluation model includes a traditional Chinese medicine prescription encoder and a Rank-SVM multi-label linear classifier;

[0009] The efficacy and toxic and side effect labels of the preset Chinese medicine prescription are output as a basis for assisting in determining the efficacy of the preset Chinese medicine prescription.

[0010] Preferably, in the above-mentioned Chinese medicine prescription efficacy auxiliary determination method, it also includes training the Chinese medicine prescription evaluation model, including:

[0011] Receive input of traditional Chinese medicine prescription data, perform preprocessing, generate a drug component vector sequence based on drug components and dosages in the traditional Chinese medicine prescription data, and set efficacy and toxic and side effect labels for each drug;

[0012] Based on the multi-head self-attention mechanism, a multi-classification neural network is designed in combination with Rank-SVM, and multi-label supervised learning is performed using the efficacy and side effect labels of each drug. The parameters of the Rank-SVM are adjusted, and the traditional Chinese medicine prescription evaluation model is formed through training.

[0013] Preferably, in the above-mentioned Chinese medicine prescription efficacy auxiliary determination method, it also includes:

[0014] Based on the collected preset Chinese medicine prescriptions and the Chinese medicine prescription evaluation model, secondary training is performed to form a personalized Chinese medicine prescription evaluation model that matches each physician.

[0015] Preferably, in the above-mentioned Chinese medicine prescription efficacy auxiliary determination method, it also includes:

[0016] Based on the collected patient medication feedback information, the traditional Chinese medicine prescription evaluation model is optimized.

[0017] The present invention provides a Chinese medicine prescription efficacy auxiliary determination device, comprising:

[0018] A receiving component, used for receiving an input of a preset Chinese medicine prescription;

[0019] A generating component, used for generating a drug component vector sequence according to the drug components and dosage in the preset traditional Chinese medicine prescription, and setting the efficacy and toxic and side effect labels of each drug component respectively;

[0020] An input component, used for inputting the drug component vector sequence and the efficacy and toxic side effect labels of each drug component into a traditional Chinese medicine prescription evaluation model and processing them, wherein the traditional Chinese medicine prescription evaluation model includes a traditional Chinese medicine prescription encoder and a Rank-SVM multi-label linear classifier;

[0021] The output component is used to output the efficacy and toxic and side effect labels of the preset Chinese medicine prescription as a basis for assisting in determining the efficacy of the preset Chinese medicine prescription.

[0022] Preferably, in the above-mentioned Chinese medicine prescription efficacy auxiliary determination device, it also includes a training component for training the Chinese medicine prescription evaluation model, and the training component includes:

[0023] A preprocessing and generating unit, for receiving input of traditional Chinese medicine prescription data, performing preprocessing, generating a drug component vector sequence based on drug components and dosages in the traditional Chinese medicine prescription data, and setting efficacy and toxic and side effect labels for each drug;

[0024] The adjustment and training unit is used to design a multi-classification neural network based on a multi-head self-attention mechanism in combination with Rank-SVM, use the efficacy and side effect labels of each drug to perform multi-label supervised learning, adjust the parameters of the Rank-SVM, and form the traditional Chinese medicine prescription evaluation model through training.

[0025] Preferably, in the above-mentioned Chinese medicine prescription efficacy auxiliary determination device, it also includes:

[0026] The secondary training component is used to perform secondary training based on the collected preset Chinese medicine prescriptions and the Chinese medicine prescription evaluation model to form a personalized Chinese medicine prescription evaluation model that matches each physician.

[0027] Preferably, in the above-mentioned Chinese medicine prescription efficacy auxiliary determination device, it also includes:

[0028] The optimization component is used to optimize the traditional Chinese medicine prescription evaluation model based on the collected patient medication feedback information.

[0029] The present invention provides a Chinese medicine prescription efficacy auxiliary determination device, comprising:

[0030] Memory for storing computer programs;

[0031] A processor is used to implement the steps of the method for auxiliary determination of the efficacy of a traditional Chinese medicine prescription as described in any one of the above items when executing the computer program.

[0032] The present invention provides a storage medium having a computer program stored thereon, and when the computer program is executed by a processor, the steps of the method for auxiliary determination of the efficacy of a traditional Chinese medicine prescription as described in any one of the above items are implemented.

[0033] It can be known from the above description that the above-mentioned Chinese medicine prescription efficacy auxiliary determination method provided by the present invention includes first receiving the input of a preset Chinese medicine prescription, then generating a drug component vector sequence according to the drug components and dosage in the preset Chinese medicine prescription, respectively setting the efficacy and toxic and side effect labels of each of the drug components, and then inputting the drug component vector sequence and the efficacy and toxic and side effect labels of each of the drug components into the Chinese medicine prescription evaluation model and processing them, wherein the Chinese medicine prescription evaluation model includes a Chinese medicine prescription encoder and a Rank-SVM multi-label linear classifier, and finally outputting the efficacy and toxic and side effect labels of the preset Chinese medicine prescription as a basis for auxiliary determination of the efficacy of the preset Chinese medicine prescription, so as to assist Chinese medicine practitioners in ensuring the rationality of the Chinese medicine prescription, reducing toxic and side effects, improving the efficacy and curative effect of Chinese medicine, and helping patients recover their health as soon as possible. The above-mentioned device, equipment and storage medium provided by the present invention have the same advantages as the above-mentioned method. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on the provided drawings without paying creative work.

[0035] Figure 1 A schematic diagram of an embodiment of a method for auxiliary determination of efficacy of a traditional Chinese medicine prescription provided by the present invention;

[0036] Figure 2 A schematic diagram of an embodiment of a device for assisting in determining the efficacy of a traditional Chinese medicine prescription provided by the present invention;

[0037] Figure 3 A schematic diagram of an embodiment of a Chinese medicine prescription efficacy auxiliary determination device provided by the present invention;

[0038] Figure 4 This is a schematic diagram of the Chinese medicine prescription encoder used in this application. DETAILED DESCRIPTION

[0039] The core of the present invention is to provide a method, device, equipment and storage medium for auxiliary determination of the efficacy of Chinese medicine prescriptions, which can assist Chinese medicine practitioners in ensuring the rationality of Chinese medicine prescriptions, reducing toxic and side effects, improving the efficacy and curative effect of Chinese medicine, and helping patients recover their health as soon as possible.

[0040] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0041] The implementation example of the auxiliary determination method of the efficacy of a traditional Chinese medicine prescription provided by the present invention is as follows: Figure 1 As shown, Figure 1 This is a schematic diagram of an embodiment of a method for auxiliary determination of efficacy of a traditional Chinese medicine prescription provided by the present invention, the method comprising the following steps:

[0042] S1: receiving input of a preset Chinese medicine prescription;

[0043] It should be noted that Chinese medicine doctors prescribe Chinese medicine prescriptions according to the condition of the disease and structure them, that is, extract text and then form one-hot vectors to form a sequence of drug vectors, which are input into the Chinese medicine prescription evaluation model. The input here can be a large number of Chinese medicine prescriptions. Since the amount of data is relatively large, and the data unit format and other contents entered by different doctors are not necessarily consistent, and the prescription data may be a text description, it is necessary to extract the content, which can be converted into a one-hot code according to the code table. This is the so-called preprocessing process. Using the preprocessing process, the input preset Chinese medicine prescription is converted into a unified format of encoding.

[0044] S2: Generate a drug component vector sequence based on the drug components and dosage in the preset traditional Chinese medicine prescription, and set the efficacy and toxicity labels of each drug component respectively;

[0045] It should be noted that each medicinal material in the drug composition of the traditional Chinese medicine prescription can adopt a one-hot vector, and its dosage can be used as a dimension to generate a drug component vector. The drug component vectors of each medicinal material in the traditional Chinese medicine prescription are combined together to form a drug component vector sequence, which can represent a traditional Chinese medicine prescription.

[0046] S3: Input the drug component vector sequence and the efficacy and toxic side effect labels of each drug component into the traditional Chinese medicine prescription evaluation model and process them, wherein the traditional Chinese medicine prescription evaluation model includes a traditional Chinese medicine prescription encoder and a Rank-SVM multi-label linear classifier;

[0047] It should be noted that the efficacy of the Chinese medicine prescription can be reflected by multiple labels to reflect the actual efficacy of the prescription, and the toxic and side effects of the Chinese medicine prescription can also be represented by multiple labels. Figure 4 As shown, Figure 4This is a schematic diagram of the Chinese medicine prescription encoder used in this application. The Chinese medicine prescription encoder is composed of multiple multi-head self-attention networks and fully connected network layers. The encoding vector of the Chinese medicine prescription is generated according to the input vector. The Rank-SVM multi-label linear classifier uses the idea of ​​maximum margin to process multi-label data. It is responsible for marking the encoding vector generated by the Chinese medicine prescription encoder with multiple labels to determine its efficacy and toxic side effects.

[0048] S4: Output the efficacy and toxic side effect labels of the preset Chinese medicine prescription as a basis for assisting in determining the efficacy of the preset Chinese medicine prescription.

[0049] Specifically, in this step, the Chinese medicine prescription evaluation model outputs the efficacy and toxicity labels according to the input vector sequence, and assists the Chinese medicine physician in judging whether the prescription is appropriate. It can be seen that this judgment method based on deep learning can better discover the connection between the various medicinal materials in the prescription, and more accurately judge the efficacy and toxic side effects of the prescription prescribed by the Chinese medicine physician, and assist the Chinese medicine physician to improve the rationality of the Chinese medicine prescription, reduce toxic side effects, and improve the efficacy and efficacy of Chinese medicine.

[0050] From the above description, it can be seen that in the embodiment of the above-mentioned Chinese medicine prescription efficacy auxiliary determination method provided by the present invention, since it includes first receiving the input of a preset Chinese medicine prescription, and then generating a drug component vector sequence according to the drug components and dosage in the preset Chinese medicine prescription, and setting the efficacy and toxic and side effect labels of each drug component respectively, and then inputting the drug component vector sequence and the efficacy and toxic and side effect labels of each drug component into the Chinese medicine prescription evaluation model and processing, wherein the Chinese medicine prescription evaluation model includes a Chinese medicine prescription encoder and a Rank-SVM multi-label linear classifier, and finally outputting the efficacy and toxic and side effect labels of the preset Chinese medicine prescription as a basis for auxiliary determination of the efficacy of the preset Chinese medicine prescription, thereby assisting Chinese medicine practitioners to ensure the rationality of the Chinese medicine prescription, reduce toxic and side effects, improve the efficacy and therapeutic effect of Chinese medicine, and help patients recover their health as soon as possible.

[0051] In a specific embodiment of the above-mentioned Chinese medicine prescription efficacy auxiliary determination method, the method further includes training a Chinese medicine prescription evaluation model, including:

[0052] Receive the input of traditional Chinese medicine prescription data, perform preprocessing, generate a drug component vector sequence based on the drug components and dosage in the traditional Chinese medicine prescription data, and set efficacy and toxic side effect labels for each drug;

[0053] Based on the multi-head self-attention mechanism, a multi-classification neural network was designed in combination with Rank-SVM. The efficacy and side effect labels of each drug were used for multi-label supervised learning. The parameters of Rank-SVM were adjusted to form a traditional Chinese medicine prescription evaluation model through training.

[0054] Specifically, during the training process, the efficacy and toxic side effect labels of the traditional Chinese medicine prescriptions are used for multi-label supervised learning, mainly adjusting the parameters of the Rank-SVM multi-label linear classifier, and finally forming the above-mentioned traditional Chinese medicine prescription evaluation model through training based on massive data.

[0055] In another specific embodiment of the above-mentioned method for auxiliary determination of efficacy of traditional Chinese medicine prescription, it also includes:

[0056] Based on the collected preset TCM prescriptions and TCM prescription evaluation models, secondary training is carried out to form a personalized TCM prescription evaluation model that matches each physician.

[0057] In this way, secondary training can be carried out according to the actual medication use of TCM practitioners to form a targeted and personalized evaluation model, further improving the accuracy of the model and the rationality of the prescriptions prescribed by TCM practitioners.

[0058] In another specific embodiment of the above-mentioned Chinese medicine prescription efficacy auxiliary determination method, it also includes:

[0059] Based on the collected patient medication feedback information, the traditional Chinese medicine prescription evaluation model is optimized.

[0060] In other words, timely feedback can be provided based on the patient's actual medication use, and the Chinese medicine prescription evaluation model can be continuously optimized.

[0061] An embodiment of a Chinese medicine prescription efficacy auxiliary determination device provided by the present invention is as follows: Figure 2 As shown, Figure 2 This is a schematic diagram of an embodiment of a Chinese medicine prescription efficacy auxiliary determination device provided by the present invention, the device comprising:

[0062] The receiving component 201 is used to receive the input of a preset Chinese medicine prescription. It should be noted that a Chinese medicine physician prescribes a Chinese medicine prescription according to the condition, structures it, that is, extracts text and then forms a one-hot vector to form a drug vector sequence, which is input into the Chinese medicine prescription evaluation model. The input here can be a massive amount of Chinese medicine prescriptions. Since the amount of data is relatively large, and the data unit format and other contents input by different physicians are not necessarily consistent, and the prescription data may be a text description, it is necessary to extract the content, which can be converted into a one-hot code according to the code table. This is the so-called preprocessing process. The preprocessing process is used to convert the input preset Chinese medicine prescription into a code in a unified format;

[0063] The generating component 202 is used to generate a drug component vector sequence according to the drug components and dosages in the preset Chinese medicine prescription, and respectively set the efficacy and toxicity labels of each drug component. It should be noted that each medicinal material in the drug composition of the Chinese medicine prescription can adopt a one-hot vector, and at the same time, its dosage is used as a dimension to generate a drug component vector. The drug component vectors of each medicinal material in the Chinese medicine prescription are combined together to form a drug component vector sequence, so as to represent a Chinese medicine prescription;

[0064] The input component 203 is used to input the drug component vector sequence and the efficacy and toxic and side effect labels of each drug component into the traditional Chinese medicine prescription evaluation model and process them, wherein the traditional Chinese medicine prescription evaluation model includes a traditional Chinese medicine prescription encoder and a Rank-SVM multi-label linear classifier. It should be noted that the efficacy of the traditional Chinese medicine prescription can be reflected by multi-labeling the actual efficacy of the prescription, and the toxic and side effects of the traditional Chinese medicine prescription can also be represented by multi-labeling. Moreover, the traditional Chinese medicine prescription encoder is composed of a plurality of multi-head self-attention networks and fully connected networks layered superposition, and generates the encoding vector of the traditional Chinese medicine prescription according to the input vector. The Rank-SVM multi-label linear classifier uses the maximum interval idea to process multi-label data, and is responsible for marking the encoding vector generated by the traditional Chinese medicine prescription encoder with multiple labels to determine its efficacy and toxic and side effects;

[0065] The output component 204 is used to output the efficacy and toxic side effect labels of the preset Chinese medicine prescription as a basis for assisting in determining the efficacy of the preset Chinese medicine prescription. Specifically, the Chinese medicine prescription evaluation model outputs its efficacy and toxic effect labels based on the input vector sequence to assist the Chinese medicine practitioner in determining whether the prescription is appropriate. It can be seen that this deep learning-based determination method can better discover the connection between the various medicinal materials in the prescription, more accurately determine the efficacy and toxic side effects of the prescription prescribed by the Chinese medicine practitioner, and assist the Chinese medicine practitioner in improving the rationality of the Chinese medicine prescription, reducing toxic side effects, and improving the efficacy and efficacy of Chinese medicine.

[0066] From the above description, it can be seen that the embodiment of the above-mentioned Chinese medicine prescription efficacy auxiliary determination device provided by the present invention can assist Chinese medicine practitioners to ensure the rationality of Chinese medicine prescriptions, reduce toxic side effects, improve the efficacy and curative effect of Chinese medicine, and help patients recover their health as soon as possible.

[0067] In a specific embodiment of the above-mentioned Chinese medicine prescription efficacy auxiliary determination device, it also includes a training component for training the Chinese medicine prescription evaluation model, and the training component includes:

[0068] A preprocessing and generating unit, for receiving the input of the traditional Chinese medicine prescription data, performing preprocessing, generating a drug component vector sequence based on the drug components and dosage in the traditional Chinese medicine prescription data, and setting efficacy and toxic side effect labels for each drug;

[0069] The adjustment and training unit is used to design a multi-classification neural network based on the multi-head self-attention mechanism combined with Rank-SVM, use the efficacy and side effect labels of each drug for multi-label supervised learning, adjust the parameters of Rank-SVM, and form a traditional Chinese medicine prescription evaluation model through training.

[0070] Specifically, during the training process, the efficacy and toxic side effect labels of the traditional Chinese medicine prescriptions are used for multi-label supervised learning, mainly adjusting the parameters of the Rank-SVM multi-label linear classifier, and finally forming the above-mentioned traditional Chinese medicine prescription evaluation model through training based on massive data.

[0071] In another specific embodiment of the above-mentioned Chinese medicine prescription efficacy auxiliary determination device, it also includes:

[0072] The secondary training component is used to perform secondary training based on the collected preset Chinese medicine prescriptions and Chinese medicine prescription evaluation models to form a personalized Chinese medicine prescription evaluation model that matches each physician.

[0073] In this way, secondary training can be carried out according to the actual medication use of TCM practitioners to form a targeted and personalized evaluation model, further improving the accuracy of the model and the rationality of the prescriptions prescribed by TCM practitioners.

[0074] In another specific embodiment of the above-mentioned Chinese medicine prescription efficacy auxiliary determination device, it also includes:

[0075] The optimization component is used to optimize the traditional Chinese medicine prescription evaluation model based on the collected patient medication feedback information.

[0076] In other words, timely feedback can be provided based on the patient's actual medication use, and the Chinese medicine prescription evaluation model can be continuously optimized.

[0077] An embodiment of a Chinese medicine prescription efficacy auxiliary determination device provided by the present invention is as follows: Figure 3 As shown, Figure 3 A schematic diagram of an embodiment of a Chinese medicine prescription efficacy auxiliary determination device provided by the present invention, the device comprising:

[0078] Memory 301, used for storing computer programs;

[0079] The processor 302 is used to implement the steps of any of the above methods for auxiliary determination of efficacy of traditional Chinese medicine prescriptions when executing a computer program.

[0080] In an embodiment of a storage medium provided by the present invention, a computer program is stored on the storage medium, and when the computer program is executed by a processor, the steps of any of the above methods for auxiliary determination of the efficacy of a traditional Chinese medicine prescription are implemented.

[0081] The above-mentioned Chinese medicine prescription efficacy auxiliary determination device and storage medium provided by the present invention have the same advantages as the above-mentioned method and device, which will not be repeated here.

[0082] The above embodiment is described in detail below with a specific example:

[0083] Take Da Chaihu Tang as an example, its ingredients are Bupleurum (c1), Scutellaria (c2), Rhubarb (c3), Citrus aurantium (c4), Pinellia (c5), White Peony (c6), Jujube (c7), and Ginger (c8). The dosage of one dose is Bupleurum 12g, Scutellaria 9g, Rhubarb 6g, Citrus aurantium 9g, Pinellia 9g, Peony 9g, 4 pieces of Jujube (15g), and Ginger 15g, forming a vector sequence s: (c1,12), v2 (c2,9), v3 (c3,6), v4 (c4,9), v5 (c5,9), v6 (c6,9), v7 (c7,15), and v8 (c8,15); common efficacy and side effect labels include analgesia, distension, laxative, anti-reflux, heat-clearing, dampness-removing, turbidity-removing, antipyretic, and anti-inflammatory. The following is a detailed description of the above method with this example.

[0084] 1. Determining the efficacy and toxic side effects of a Chinese medicine prescription includes the following steps:

[0085] Step 101: Collect massive amounts of traditional Chinese medicine prescription data, perform preprocessing, find out all the medicinal ingredients for one-hot encoding, and set labels for efficacy and toxic side effects;

[0086] Step 102: According to the composition and dosage of the Chinese medicine prescription, a drug component vector sequence is formed, and multiple labels of efficacy and toxic side effects are added;

[0087] Step 103: training the Chinese medicine prescription encoder, using unsupervised learning to generate a Chinese medicine prescription encoding vector based on a sequence of constituent medicinal material vectors of the Chinese medicine prescription;

[0088] Step 104: jointly training the Chinese medicine prescription encoder and the Rank-SVM multi-label linear classifier;

[0089] Step 105: During the training process, the efficacy and toxicity labels of the Chinese medicine prescription are used to perform multi-label supervised learning, mainly adjusting the parameters of the Rank-SVM multi-label linear classifier;

[0090] Step 106: Based on the massive data, through training, the traditional Chinese medicine prescription evaluation model is finally formed;

[0091] Step 107: The Chinese medicine doctor prescribes a Chinese medicine prescription according to the condition, structures it into a drug vector sequence (for example, the Chinese medicine doctor prescribes a paper prescription, and can use image recognition to extract the composition and dosage of the medicinal materials), and inputs it into the Chinese medicine prescription evaluation model;

[0092] Step 108: The Chinese medicine prescription evaluation model outputs the efficacy and toxicity labels according to the input vector sequence to assist the Chinese medicine physician in determining whether the prescription is appropriate;

[0093] Step 109: The TCM physician may collect the data of the TCM prescriptions he / she prescribes, and perform secondary training based on the TCM prescription evaluation model to form a targeted personalized evaluation model;

[0094] Step 110: Provide feedback on the results based on the patient's actual medication use and continuously optimize the TCM prescription evaluation model.

[0095] The above description of the disclosed embodiments enables one skilled in the art to implement or use the present invention. Various modifications to these embodiments will be apparent to one skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but rather to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A method for auxiliary determination of the efficacy of a traditional Chinese medicine prescription, characterized in that: include: Receiving input of a preset Chinese medicine prescription; Generate a drug component vector sequence according to the drug components and dosage in the preset traditional Chinese medicine prescription, and set the efficacy and toxic and side effect labels of each drug component respectively; Inputting the drug component vector sequence and the efficacy and toxic side effect labels of each drug component into a traditional Chinese medicine prescription evaluation model and processing them, wherein the traditional Chinese medicine prescription evaluation model includes a traditional Chinese medicine prescription encoder and a Rank-SVM multi-label linear classifier; Outputting the efficacy and toxic and side effect labels of the preset Chinese medicine prescription as a basis for assisting in determining the efficacy of the preset Chinese medicine prescription; The method also includes training the traditional Chinese medicine prescription evaluation model, including: Receive input of traditional Chinese medicine prescription data, perform preprocessing, generate a drug component vector sequence based on drug components and dosages in the traditional Chinese medicine prescription data, and set efficacy and toxic and side effect labels for each drug; Based on the multi-head self-attention mechanism, a multi-classification neural network is designed in combination with Rank-SVM, and multi-label supervised learning is performed using the efficacy and side effect labels of each drug. The parameters of the Rank-SVM are adjusted, and the traditional Chinese medicine prescription evaluation model is formed through training.

2. The method for auxiliary determination of efficacy of a traditional Chinese medicine prescription according to claim 1, characterized in that: Also includes: Based on the collected preset Chinese medicine prescriptions and the Chinese medicine prescription evaluation model, secondary training is performed to form a personalized Chinese medicine prescription evaluation model that matches each physician.

3. The method for auxiliary determination of efficacy of a traditional Chinese medicine prescription according to claim 1, characterized in that: Also includes: Based on the collected patient medication feedback information, the traditional Chinese medicine prescription evaluation model is optimized.

4. A Chinese medicine prescription efficacy auxiliary determination device, characterized in that: include: A receiving component, used for receiving an input of a preset Chinese medicine prescription; A generating component, used for generating a drug component vector sequence according to the drug components and dosage in the preset traditional Chinese medicine prescription, and setting the efficacy and toxic and side effect labels of each drug component respectively; An input component, used for inputting the drug component vector sequence and the efficacy and toxic side effect labels of each drug component into a traditional Chinese medicine prescription evaluation model and processing them, wherein the traditional Chinese medicine prescription evaluation model includes a traditional Chinese medicine prescription encoder and a Rank-SVM multi-label linear classifier; An output component, used to output the efficacy and toxic and side effect labels of the preset Chinese medicine prescription as a basis for assisting in determining the efficacy of the preset Chinese medicine prescription; It also includes a training component for training the traditional Chinese medicine prescription evaluation model, and the training component includes: A preprocessing and generating unit, used for receiving the input of traditional Chinese medicine prescription data, performing preprocessing, generating a drug component vector sequence based on the drug components and dosage in the traditional Chinese medicine prescription data, and setting efficacy and toxic and side effect labels for each of the drugs; The adjustment and training unit is used to design a multi-classification neural network based on a multi-head self-attention mechanism in combination with Rank-SVM, perform multi-label supervised learning using the efficacy and toxic side effect labels of each of the drugs, adjust the parameters of the Rank-SVM, and form the traditional Chinese medicine prescription evaluation model through training.

5. The Chinese medicine prescription efficacy auxiliary determination device according to claim 4, characterized in that: Also includes: The secondary training component is used to perform secondary training based on the collected preset Chinese medicine prescriptions and the Chinese medicine prescription evaluation model to form a personalized Chinese medicine prescription evaluation model that matches each physician.

6. The Chinese medicine prescription efficacy auxiliary determination device according to claim 4, characterized in that: Also includes: The optimization component is used to optimize the traditional Chinese medicine prescription evaluation model based on the collected patient medication feedback information.

7. A Chinese medicine prescription efficacy auxiliary determination device, characterized in that: include: Memory for storing computer programs; A processor is used to implement the steps of the method for auxiliary determination of the efficacy of a traditional Chinese medicine prescription as claimed in any one of claims 1 to 3 when executing the computer program.

8. A storage medium, characterized in that: The storage medium stores a computer program, which, when executed by a processor, implements the steps of the method for auxiliary determination of the efficacy of a traditional Chinese medicine prescription as claimed in any one of claims 1 to 3.

Images