CN111883263A - Method, device, equipment and storage medium for assisting in judging drug effect of traditional Chinese medicine prescription - Google Patents
Method, device, equipment and storage medium for assisting in judging drug effect of traditional Chinese medicine prescription Download PDFInfo
- Publication number
- CN111883263A CN111883263A CN202010746846.5A CN202010746846A CN111883263A CN 111883263 A CN111883263 A CN 111883263A CN 202010746846 A CN202010746846 A CN 202010746846A CN 111883263 A CN111883263 A CN 111883263A
- Authority
- CN
- China
- Prior art keywords
- traditional chinese
- chinese medicine
- efficacy
- prescription
- medicine prescription
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
- G16H70/00—ICT specially adapted for the handling or processing of medical references
- G16H70/40—ICT specially adapted for the handling or processing of medical references relating to drugs, e.g. their side effects or intended usage
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
- G06F18/2411—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on the proximity to a decision surface, e.g. support vector machines
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Physics & Mathematics (AREA)
- Artificial Intelligence (AREA)
- Evolutionary Computation (AREA)
- Bioinformatics & Cheminformatics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Health & Medical Sciences (AREA)
- Bioinformatics & Computational Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Biology (AREA)
- General Health & Medical Sciences (AREA)
- Computational Linguistics (AREA)
- Pharmacology & Pharmacy (AREA)
- Biomedical Technology (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Chemical & Material Sciences (AREA)
- Medicinal Chemistry (AREA)
- Biophysics (AREA)
- Toxicology (AREA)
- Epidemiology (AREA)
- Medical Informatics (AREA)
- Primary Health Care (AREA)
- Public Health (AREA)
- Medical Treatment And Welfare Office Work (AREA)
- Medical Preparation Storing Or Oral Administration Devices (AREA)
Abstract
The invention discloses a method, a device, equipment and a storage medium for assisting in judging the drug effect of a traditional Chinese medicine prescription, wherein the method comprises the steps of receiving the input of a preset traditional Chinese medicine prescription; generating a medicine component vector sequence according to the medicine components and the dosage in the preset traditional Chinese medicine prescription, and respectively setting the efficacy and toxic and side effect labels of each medicine component; inputting the drug component vector sequences and the efficacy and toxic and side effect labels of each drug component into a traditional Chinese medicine prescription evaluation model and processing the traditional Chinese medicine prescription evaluation model, wherein the traditional Chinese medicine prescription evaluation model comprises a traditional Chinese medicine prescription encoder and a Rank-SVM multi-label linear classifier; and outputting the efficacy and toxic and side effect labels of the preset traditional Chinese medicine prescription as a basis for auxiliary judgment of the efficacy of the preset traditional Chinese medicine prescription. The method, the device, the equipment and the storage medium for assisting in judging the drug effect of the traditional Chinese medicine prescription can assist a traditional Chinese medicine teacher in ensuring the rationality of the traditional Chinese medicine prescription, reduce toxic and side effects, improve the efficacy and the curative effect of the traditional Chinese medicine and help a patient to recover the health as soon as possible.
Description
Technical Field
The invention belongs to the technical field of deep learning, and particularly relates to a method, a device, equipment and a storage medium for assisting in judging the drug effect of a traditional Chinese medicine prescription.
Background
In recent years, the development of artificial intelligence technology is rapid, the whole society can be changed subversively, the artificial intelligence technology becomes an important development strategy of future countries, particularly, the algorithm evolution taking deep learning as a core has super-strong evolutionary capability, and a large-scale neural network similar to a human brain structure is constructed through training under the support of big data, so that various problems can be solved. Among them, Attention Mechanism (Attention Mechanism) is a data processing method in machine learning, and is widely applied to various tasks of different types, such as natural language processing, image recognition, speech recognition, and the like.
The traditional Chinese medicine is a treasure accumulated by the ancient people in long-term medical practice, is an important component of ancient excellent cultural heritage in China, and is most commonly used due to the fact that the traditional Chinese medicine is made of plant medicinal materials. With the development of modern natural science and technology and national economy, the Chinese pharmacology has achieved unprecedented achievements. Through long-term medical practice, a plurality of medicines are combined and decocted to prepare a decoction, so that a traditional Chinese medicine formula can be formed. The traditional Chinese medicine prescriptions are prepared by traditional Chinese medicine doctors according to the actual conditions of patients, the treatment effect of the patients depends on the experience of the traditional Chinese medicine doctors to a certain extent, particularly, certain medicine combinations can generate toxic and side effects, so that various factors need to be integrated to judge the efficacy of the traditional Chinese medicine prescriptions, and one prescription is usually composed of more than ten medicines or even dozens of medicines, so the efficacy is not the superposition of simple medicinal material components. Under the circumstances, how to judge the efficacy and toxic and side effects of the traditional Chinese medicine more reasonably and accurately becomes a problem to be solved urgently.
Disclosure of Invention
In order to solve the problems, the invention provides a method, a device, equipment and a storage medium for assisting in judging the drug effect of a traditional Chinese medicine prescription, which can assist a traditional Chinese medicine doctor in ensuring the rationality of the traditional Chinese medicine prescription, reduce toxic and side effects, improve the efficacy and curative effect of the traditional Chinese medicine and help a patient to recover the health as soon as possible.
The invention provides a method for assisting in judging the efficacy of a traditional Chinese medicine prescription, which comprises the following steps:
receiving input of a preset traditional Chinese medicine formula;
generating a medicine component vector sequence according to the medicine components and the dosage in the preset traditional Chinese medicine prescription, and respectively setting the efficacy and toxic and side effect labels of each medicine component;
inputting the drug component vector sequences and the efficacy and toxic and side effect labels of each drug component into a traditional Chinese medicine prescription evaluation model and processing the traditional Chinese medicine prescription evaluation model, wherein the traditional Chinese medicine prescription evaluation model comprises a traditional Chinese medicine prescription encoder and a Rank-SVM multi-label linear classifier;
and outputting the efficacy and toxic and side effect labels of the preset traditional Chinese medicine prescription as a basis for auxiliary judgment of the efficacy of the preset traditional Chinese medicine prescription.
Preferably, in the method for assisting in determining the efficacy of a Chinese medicinal formulation, training the Chinese medicinal formulation evaluation model further comprises:
receiving input of traditional Chinese medicine prescription data, preprocessing, generating a medicine component vector sequence based on medicine components and dosage in the traditional Chinese medicine prescription data, and setting an efficacy and toxic and side effect label for each medicine;
based on a multi-head self-attention mechanism, a multi-classification neural network is designed by combining with a Rank-SVM, multi-label supervised learning is carried out by using the efficacy and toxic and side effect labels of each medicament, the parameters of the Rank-SVM are adjusted, and the Chinese medicinal prescription evaluation model is formed through training.
Preferably, the method for auxiliary determination of the efficacy of a Chinese medicinal composition further comprises:
and performing secondary training based on the collected preset traditional Chinese medicine formulas and the traditional Chinese medicine formula evaluation model to form a personalized traditional Chinese medicine formula evaluation model matched with each doctor.
Preferably, the method for auxiliary determination of the efficacy of a Chinese medicinal composition further comprises:
and optimizing the traditional Chinese medicine prescription evaluation model based on the acquired patient medication feedback information.
The invention provides a device for assisting in judging the drug effect of a traditional Chinese medicine prescription, which comprises:
the receiving component is used for receiving the input of a preset traditional Chinese medicine formula;
the generating component is used for generating a medicine component vector sequence according to the medicine components and the dosage in the preset traditional Chinese medicine prescription and respectively setting the efficacy and toxic and side effect labels of each medicine component;
the input component is used for inputting the medicine component vector sequences and the efficacy and toxic and side effect labels of each medicine component into a traditional Chinese medicine prescription evaluation model and processing the traditional Chinese medicine prescription evaluation model, wherein the traditional Chinese medicine prescription evaluation model comprises a traditional Chinese medicine prescription encoder and a Rank-SVM multi-label linear classifier;
and the output component is used for outputting the efficacy and toxic and side effect labels of the preset traditional Chinese medicine prescription as a basis for assisting in judging the efficacy of the preset traditional Chinese medicine prescription.
Preferably, the device for assisting in determining the efficacy of a chinese medicinal prescription further includes a training unit for training the model for evaluating a chinese medicinal prescription, and the training unit includes:
the preprocessing and generating unit is used for receiving the input of the traditional Chinese medicine prescription data, preprocessing the traditional Chinese medicine prescription data, generating a medicine component vector sequence based on the medicine components and the dosage in the traditional Chinese medicine prescription data, and setting an efficacy and toxic and side effect label for each medicine;
and the adjusting and training unit is used for designing a multi-classification neural network based on a multi-head self-attention mechanism and combined with a Rank-SVM, performing multi-label supervised learning by using the efficacy and toxic and side effect labels of each drug, adjusting the parameters of the Rank-SVM, and forming the traditional Chinese medicine prescription evaluation model through training.
Preferably, the device for assisting in determining the drug effect of a Chinese medicinal composition further comprises:
and the secondary training part is used for carrying out secondary training on the basis of the collected preset traditional Chinese medicine formulas and the traditional Chinese medicine formula evaluation model to form a personalized traditional Chinese medicine formula evaluation model matched with each doctor.
Preferably, the device for assisting in determining the drug effect of a Chinese medicinal composition further comprises:
and the optimizing component is used for optimizing the traditional Chinese medicine prescription evaluation model based on the acquired patient medication feedback information.
The invention provides a device for assisting in judging the efficacy of a traditional Chinese medicine prescription, which comprises:
a memory for storing a computer program;
a processor for implementing the steps of the method for assisting in determining the efficacy of a Chinese medicinal formulation when executing the computer program.
The invention provides a storage medium, wherein a computer program is stored on the storage medium, and when being executed by a processor, the computer program realizes the steps of the method for assisting in judging the drug effect of the traditional Chinese medicine prescription.
As can be seen from the above description, the method for assisting in determining the drug effect of a traditional Chinese medicine prescription provided by the present invention includes receiving an input of a preset traditional Chinese medicine prescription, generating a vector sequence of the drug components according to the drug components and the amounts of the drug components in the preset traditional Chinese medicine prescription, setting the efficacy and toxicity/side effects of each of the drug components, inputting the vector sequence of the drug components and the efficacy and toxicity/side effects of each of the drug components into a traditional Chinese medicine prescription evaluation model, and processing the same, wherein the traditional Chinese medicine prescription evaluation model includes a traditional Chinese medicine prescription encoder and a Rank-SVM multi-label linear classifier, and outputting the efficacy and toxicity/side effects of the preset traditional Chinese medicine prescription as a basis for assisting in determining the drug effect of the preset traditional Chinese medicine prescription, thereby assisting a traditional Chinese medical practitioner in ensuring the rationality of the traditional Chinese medicine prescription and, improving the efficacy and curative effect of the traditional Chinese medicine and helping the patient to recover the health as soon as possible. The present invention provides the above-described apparatus, device and storage medium with the same advantages as the above-described method.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, it is obvious that the drawings in the following description are only embodiments of the present invention, and for those skilled in the art, other drawings can be obtained according to the provided drawings without creative efforts.
FIG. 1 is a schematic diagram of an embodiment of a method for assisting in determining the efficacy of a Chinese medicinal composition according to the present invention;
FIG. 2 is a schematic view of an embodiment of an auxiliary determination device for determining the efficacy of a Chinese medicinal composition according to the present invention;
FIG. 3 is a schematic diagram of an embodiment of an auxiliary determination device for determining the drug effect of a Chinese medicinal composition according to the present invention;
FIG. 4 is a schematic diagram of a Chinese medicine prescription encoder as used herein.
Detailed Description
The core of the invention is to provide a method, a device, equipment and a storage medium for assisting in judging the efficacy of a traditional Chinese medicine prescription, which can assist a traditional Chinese medicine doctor in ensuring the rationality of the traditional Chinese medicine prescription, reduce toxic and side effects, improve the efficacy and curative effect of the traditional Chinese medicine and help the patient to recover the health as soon as possible.
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
Fig. 1 shows an embodiment of a method for assisting in determining the efficacy of a Chinese medicinal formulation, and fig. 1 is a schematic diagram of an embodiment of a method for assisting in determining the efficacy of a Chinese medicinal formulation, the method comprising the following steps:
s1: receiving input of a preset traditional Chinese medicine formula;
it should be noted that, a traditional Chinese medicine doctor develops a traditional Chinese medicine prescription according to the state of an illness, structurizes the traditional Chinese medicine prescription, namely extracts characters and then forms onehot vectors to form a medicine vector sequence, inputs the medicine vector sequence into a traditional Chinese medicine prescription evaluation model, the input traditional Chinese medicine prescription can be massive, because the data volume is large, the contents of data unit formats and the like input by different doctors are not consistent, and the prescription data can be character description, the contents need to be extracted and can be converted into one-hot codes according to a code table, which is a so-called preprocessing process, and the input preset traditional Chinese medicine prescription is converted into codes with a unified format by utilizing the preprocessing process.
S2: generating a medicine component vector sequence according to the medicine components and the dosage in a preset traditional Chinese medicine prescription, and respectively setting the efficacy and toxic and side effect labels of each medicine component;
it should be noted that each medicinal material in the medicinal composition of the Chinese medicinal prescription can adopt one-hot vector, the dose is taken as one dimension to generate medicinal component vector, and the medicinal component vectors of each medicinal material in the Chinese medicinal prescription are combined together to form a medicinal component vector sequence, so that one Chinese medicinal prescription can be represented.
S3: inputting the vector sequence of the medicinal components and the efficacy and toxic and side effect labels of each medicinal component into a Chinese medicinal prescription evaluation model and processing, wherein the Chinese medicinal prescription evaluation model comprises a Chinese medicinal prescription encoder and a Rank-SVM multi-label linear classifier;
it should be noted that the efficacy of a Chinese medicinal formula can reflect the actual efficacy of the formula by labeling, and the toxic and side effects of the Chinese medicinal formula can also be expressed by labeling. Moreover, the chinese medicine encoder is shown in fig. 4, fig. 4 is a schematic diagram of the chinese medicine encoder employed in the present application, the chinese medicine encoder is formed by layering and superimposing a plurality of multi-headed self-attention networks and full-connected networks, the encoding vector of the chinese medicine is generated according to the inputted vector, the Rank-SVM multi-tag linear classifier processes multi-tag data using the idea of maximum interval, and is responsible for labeling the encoding vector generated by the chinese medicine encoder with multiple tags to determine the efficacy and toxicity thereof.
S4: and outputting the efficacy and toxic and side effect labels of the preset traditional Chinese medicine prescription as a basis for auxiliary judgment of the efficacy of the preset traditional Chinese medicine prescription.
Specifically, in the step, the Chinese medicinal prescription evaluation model outputs the efficacy and toxicity effect labels according to the input vector sequence, and assists the Chinese medic in judging whether the prescription is appropriate, so that the judgment method based on deep learning can better find the relation among the medicinal materials in the prescription, more accurately judge the efficacy and toxic and side effects of the prescription prescribed by the Chinese medic, assist the Chinese medic in improving the rationality of the Chinese medicinal prescription, reduce toxic and side effects and improve the efficacy and curative effect of the Chinese medicament.
As can be seen from the above description, in the embodiment of the method for assisting in determining the drug effect of a traditional Chinese medicine prescription, the method includes receiving an input of a preset traditional Chinese medicine prescription, generating a vector sequence of the drug components according to the drug components and the dosage in the preset traditional Chinese medicine prescription, setting the efficacy and toxicity/side effects labels of each drug component, inputting the vector sequence of the drug components and the efficacy and toxicity/side effects labels of each drug component into a traditional Chinese medicine prescription evaluation model, and processing the same, wherein the traditional Chinese medicine prescription evaluation model includes a traditional Chinese medicine prescription encoder and a Rank-SVM multi-label linear classifier, and outputting the efficacy and toxicity/or side effects labels of the preset traditional Chinese medicine prescription as a basis for assisting in determining the drug effect of the preset traditional Chinese medicine prescription, so that a physician can be assisted in ensuring the rationality of the traditional Chinese medicine prescription, reducing toxicity/side, help the patient to recover the health as soon as possible.
In a specific embodiment of the method for assisting in determining the drug effect of a Chinese medicinal formulation, the method further comprises training a Chinese medicinal formulation evaluation model, comprising:
receiving the input of the Chinese medicine prescription data, preprocessing, generating a medicine component vector sequence based on the medicine components and the dosage in the Chinese medicine prescription data, and setting an efficacy and toxic and side effect label for each medicine;
based on a multi-head self-attention mechanism, a multi-classification neural network is designed by combining with a Rank-SVM, multi-label supervised learning is carried out by using the efficacy and toxic and side effect labels of each medicament, the parameters of the Rank-SVM are adjusted, and a Chinese medicinal prescription evaluation model is formed through training.
Specifically, in the training process, the efficacy and toxic and side effect labels of the traditional Chinese medicine formula are used for multi-label supervised learning, the parameters of the Rank-SVM multi-label linear classifier are mainly adjusted, and the traditional Chinese medicine formula evaluation model is finally formed through training based on mass data.
In another embodiment of the method for assisting in determining the efficacy of a Chinese medicinal composition, the method further comprises:
and performing secondary training based on the collected preset traditional Chinese medicine formulas and the traditional Chinese medicine formula evaluation model to form a personalized traditional Chinese medicine formula evaluation model matched with each doctor.
Therefore, secondary training can be performed according to the actual medication situation of the traditional Chinese medical practitioner, a targeted personalized evaluation model is formed, the accuracy of the model is further improved, and the rationality of the prescription prescribed by the traditional Chinese medical practitioner is improved.
In another embodiment of the method for assisting in determining the efficacy of a Chinese medicinal composition, the method further comprises:
and optimizing the traditional Chinese medicine prescription evaluation model based on the acquired patient medication feedback information.
That is to say, the evaluation model of the traditional Chinese medicine prescription can be continuously optimized by timely feeding back according to the actual medication condition of the patient.
Fig. 2 shows an embodiment of a device for assisting in determining the efficacy of a Chinese medicinal formulation, and fig. 2 is a schematic diagram of an embodiment of a device for assisting in determining the efficacy of a Chinese medicinal formulation according to the present invention, the device comprising:
the receiving component 201 is used for receiving input of preset traditional Chinese medicine formulas, and it needs to be explained that a traditional Chinese medicine doctor develops a traditional Chinese medicine formula according to an illness state, structurizes the traditional Chinese medicine formula, namely extracts characters and then forms onehot vectors to form a medicine vector sequence, and inputs the medicine vector sequence into a traditional Chinese medicine formula evaluation model, wherein the input traditional Chinese medicine formula can be massive, because the data volume is large, the contents of data unit formats and the like input by different doctors are not necessarily consistent, and the formula data can be described in characters, the contents need to be extracted, and can be converted into one-hot codes according to a code table, which is a so-called preprocessing process, and the input preset traditional Chinese medicine formula is converted into codes with a unified format by utilizing the preprocessing process;
the generating component 202 is configured to generate a drug component vector sequence according to drug components and dosages in a preset traditional Chinese medicine prescription, and set efficacy and toxic and side effect labels of each drug component respectively, it should be noted that each drug component in the drug composition of the traditional Chinese medicine prescription can adopt one-hot vectors, and the dosage of each drug component is taken as a dimension to generate drug component vectors, and the drug component vectors of each drug component in the traditional Chinese medicine prescription are combined together to form a drug component vector sequence, so that a traditional Chinese medicine prescription can be represented;
the input component 203 is used for inputting the vector sequences of the medicinal components and the efficacy and toxic and side effect labels of each medicinal component into a Chinese medicinal prescription evaluation model and processing the Chinese medicinal prescription evaluation model, wherein the Chinese medicinal prescription evaluation model comprises a Chinese medicinal prescription encoder and a Rank-SVM multi-label linear classifier, and it needs to be noted that the efficacy of the Chinese medicinal prescription can reflect the actual medicinal effect of the prescription by labeling multiple labels, and the toxic and side effect of the Chinese medicinal prescription can also be expressed in a multi-label labeling manner. The Chinese medicine encoder is formed by overlapping a plurality of multi-head self-attention networks and full-connection networks in a layered mode, the encoding vector of the Chinese medicine is generated according to the input vector, the Rank-SVM multi-label linear classifier processes multi-label data by using the idea of maximum interval, and the Rank-SVM multi-label linear classifier is responsible for printing the encoding vector generated by the Chinese medicine encoder with multiple labels to determine the medicine effect and the toxic and side effects of the Chinese medicine encoder;
the output component 204 is used for outputting the efficacy and toxic and side effect labels of the preset traditional Chinese medicine formulas as a basis for assisting in judging the efficacy of the preset traditional Chinese medicine formulas, specifically, the traditional Chinese medicine formula evaluation model outputs the efficacy and toxic and side effect labels according to the input vector sequence, and assists the traditional Chinese medicine practitioner in judging whether the formula is appropriate, so that the judgment method based on deep learning can better find the relation among the medicinal materials in the formula, more accurately judge the efficacy and toxic and side effects of the prescription prescribed by the traditional Chinese practitioner, assist the traditional Chinese practitioner in improving the rationality of the traditional Chinese medicine formulas, reduce toxic and side effects, and improve the efficacy and curative effect of the traditional Chinese medicine.
The embodiment of the device for assisting in judging the drug effect of the traditional Chinese medicine prescription provided by the invention can assist a traditional Chinese medicine doctor to ensure the rationality of the traditional Chinese medicine prescription, reduce toxic and side effects, improve the efficacy and curative effect of the traditional Chinese medicine and help a patient to recover the health as soon as possible.
In an embodiment of the device for assisting in determining the efficacy of a chinese medicinal formulation, the device further includes a training component for training an evaluation model of the chinese medicinal formulation, and the training component includes:
the preprocessing and generating unit is used for receiving the input of the Chinese medicine prescription data, preprocessing the Chinese medicine prescription data, generating a medicine component vector sequence based on the medicine components and the dosage in the Chinese medicine prescription data, and setting an efficacy and toxic side effect label for each medicine;
and the adjusting and training unit is used for designing a multi-classification neural network based on a multi-head self-attention mechanism and combined with a Rank-SVM, performing multi-label supervised learning by using the efficacy and toxic and side effect labels of each medicament, adjusting the parameters of the Rank-SVM, and forming a Chinese medicinal prescription evaluation model through training.
Specifically, in the training process, the efficacy and toxic and side effect labels of the traditional Chinese medicine formula are used for multi-label supervised learning, the parameters of the Rank-SVM multi-label linear classifier are mainly adjusted, and the traditional Chinese medicine formula evaluation model is finally formed through training based on mass data.
In another embodiment of the device for assisting in determining the efficacy of a Chinese medicinal composition, the device further comprises:
and the secondary training part is used for carrying out secondary training based on the collected preset traditional Chinese medicine formulas and the traditional Chinese medicine formula evaluation model to form a personalized traditional Chinese medicine formula evaluation model matched with each doctor.
Therefore, secondary training can be performed according to the actual medication situation of the traditional Chinese medical practitioner, a targeted personalized evaluation model is formed, the accuracy of the model is further improved, and the rationality of the prescription prescribed by the traditional Chinese medical practitioner is improved.
In another embodiment of the device for assisting in determining the efficacy of a Chinese medicinal composition, the device further comprises:
and the optimizing component is used for optimizing the traditional Chinese medicine prescription evaluation model based on the acquired patient medication feedback information.
That is to say, the evaluation model of the traditional Chinese medicine prescription can be continuously optimized by timely feeding back according to the actual medication condition of the patient.
Fig. 3 shows an embodiment of a device for assisting in determining the efficacy of a Chinese medicinal formulation according to the present invention, and fig. 3 is a schematic view of an embodiment of a device for assisting in determining the efficacy of a Chinese medicinal formulation according to the present invention, the device including:
a memory 301 for storing a computer program;
the processor 302 is used for implementing the steps of the method for assisting in determining the drug effect of a Chinese medicinal preparation when executing a computer program.
In an embodiment of the storage medium provided by the present invention, the storage medium stores a computer program, and the computer program, when executed by the processor, implements the steps of the method for assisting in determining the drug effect of a Chinese medicinal formulation.
The device for assisting in judging the drug effect of the traditional Chinese medicine and the storage medium have the same advantages as the method and the device, and are not repeated herein.
The above embodiment is described in detail below with a specific example:
taking DACHAIHU decoction as an example, the medicinal materials comprise radix bupleuri (c1), radix Scutellariae (c2), radix et rhizoma Rhei (c3), fructus Aurantii Immaturus (c4), rhizoma Pinelliae (c5), radix Paeoniae alba (c6), fructus Jujubae (c7), and rhizoma Zingiberis recens (c8), and one dosage is radix bupleuri 12g, radix Scutellariae 9g, radix et rhizoma Rhei 6g, fructus Aurantii Immaturus 9g, rhizoma Pinelliae 9g, radix Paeoniae 9g, 4 fructus Jujubae (15g), and rhizoma Zingiberis recens 15g, and form vector sequence s (c1,12), v2(c2,9), v3(c3,6), v4(c4,9), v5(c5,9), v6(c6,9), v7(c7,15), v8(c8, 15); the common labels of efficacy and toxic and side effects are pain relief, flatulence removal, constipation relief, adverse qi lowering, heat clearing, dampness removing, turbid pathogen descending, exterior syndrome relieving, inflammation resistance and the like. The above method will be described in detail with reference to this example.
Firstly, judging the drug effect and toxic and side effect of the traditional Chinese medicine formula, comprising the following steps:
step 101: collecting mass data of the traditional Chinese medicine prescription, preprocessing, finding out all medicinal material compositions for one-hot coding, and setting labels for efficacy and toxic and side effects;
step 102: forming a vector sequence of medicinal components according to the composition and the dosage of the Chinese medicinal materials, and simultaneously marking multiple labels of efficacy and toxic and side effects;
step 103: training the Chinese medicine prescription encoder, and generating Chinese medicine prescription encoding vectors based on the constituent medicinal material vector sequences of the Chinese medicine prescriptions by adopting unsupervised learning;
step 104: training the Chinese medicine encoder and a Rank-SVM multi-label linear classifier in a combined manner;
step 105: in the training process, the efficacy and toxic and side effect labels of the traditional Chinese medicine are used for multi-label supervised learning, and the parameters of the Rank-SVM multi-label linear classifier are mainly adjusted;
step 106: based on mass data, the Chinese medicine prescription evaluation model is finally formed through training;
step 107: the traditional Chinese medicine doctor prescribes a traditional Chinese medicine prescription according to the state of illness, the traditional Chinese medicine prescription is structured to form a medicine vector sequence (for example, the traditional Chinese medicine doctor prescribes a paper prescription, and can extract the composition and the dosage of medicinal materials by image recognition), and the medicine vector sequence is input into the traditional Chinese medicine prescription evaluation model;
step 108: the Chinese medicine prescription evaluation model outputs the medicine effect and toxicity effect labels according to the input vector sequence, and assists a Chinese doctor to judge whether the prescription is appropriate;
step 109: traditional Chinese medicine prescriptions prescribed by a traditional Chinese medicine doctor can be collected by the traditional Chinese medicine doctor, and secondary training is carried out based on the traditional Chinese medicine prescription evaluation model to form a targeted personalized evaluation model;
step 110: and feeding back results according to the actual medication condition of the patient, and continuously optimizing the traditional Chinese medicine prescription evaluation model.
The previous description of the disclosed embodiments is provided to enable any person skilled in the art to make or use the present invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the generic principles defined herein may be applied to other embodiments without departing from the spirit or scope of the invention. Thus, the present invention is not intended to be limited to the embodiments shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.
Claims (10)
1. An auxiliary determination method for the drug effect of a traditional Chinese medicine prescription is characterized by comprising the following steps:
receiving input of a preset traditional Chinese medicine formula;
generating a medicine component vector sequence according to the medicine components and the dosage in the preset traditional Chinese medicine prescription, and respectively setting the efficacy and toxic and side effect labels of each medicine component;
inputting the drug component vector sequences and the efficacy and toxic and side effect labels of each drug component into a traditional Chinese medicine prescription evaluation model and processing the traditional Chinese medicine prescription evaluation model, wherein the traditional Chinese medicine prescription evaluation model comprises a traditional Chinese medicine prescription encoder and a Rank-SVM multi-label linear classifier;
and outputting the efficacy and toxic and side effect labels of the preset traditional Chinese medicine prescription as a basis for auxiliary judgment of the efficacy of the preset traditional Chinese medicine prescription.
2. The method for assisting in determining the efficacy of a Chinese medicinal formulation according to claim 1, further comprising training the Chinese medicinal formulation evaluation model, comprising:
receiving input of traditional Chinese medicine prescription data, preprocessing, generating a medicine component vector sequence based on medicine components and dosage in the traditional Chinese medicine prescription data, and setting an efficacy and toxic and side effect label for each medicine;
based on a multi-head self-attention mechanism, a multi-classification neural network is designed by combining with a Rank-SVM, multi-label supervised learning is carried out by using the efficacy and toxic and side effect labels of each medicament, the parameters of the Rank-SVM are adjusted, and the Chinese medicinal prescription evaluation model is formed through training.
3. The method for assisting in determining the efficacy of a Chinese medicinal formulation according to claim 1, further comprising:
and performing secondary training based on the collected preset traditional Chinese medicine formulas and the traditional Chinese medicine formula evaluation model to form a personalized traditional Chinese medicine formula evaluation model matched with each doctor.
4. The method for assisting in determining the efficacy of a Chinese medicinal formulation according to claim 1, further comprising:
and optimizing the traditional Chinese medicine prescription evaluation model based on the acquired patient medication feedback information.
5. An auxiliary judgment device for the drug effect of a traditional Chinese medicine prescription is characterized by comprising:
the receiving component is used for receiving the input of a preset traditional Chinese medicine formula;
the generating component is used for generating a medicine component vector sequence according to the medicine components and the dosage in the preset traditional Chinese medicine prescription and respectively setting the efficacy and toxic and side effect labels of each medicine component;
the input component is used for inputting the medicine component vector sequences and the efficacy and toxic and side effect labels of each medicine component into a traditional Chinese medicine prescription evaluation model and processing the traditional Chinese medicine prescription evaluation model, wherein the traditional Chinese medicine prescription evaluation model comprises a traditional Chinese medicine prescription encoder and a Rank-SVM multi-label linear classifier;
and the output component is used for outputting the efficacy and toxic and side effect labels of the preset traditional Chinese medicine prescription as a basis for assisting in judging the efficacy of the preset traditional Chinese medicine prescription.
6. The apparatus according to claim 5, further comprising a training unit for training the evaluation model of the prescription, wherein the training unit comprises:
the preprocessing and generating unit is used for receiving the input of the traditional Chinese medicine prescription data, preprocessing the traditional Chinese medicine prescription data, generating a medicine component vector sequence based on the medicine components and the dosage in the traditional Chinese medicine prescription data, and setting an efficacy and toxic and side effect label for each medicine;
and the adjusting and training unit is used for designing a multi-classification neural network based on a multi-head self-attention mechanism and combined with a Rank-SVM, performing multi-label supervised learning by using the efficacy and toxic and side effect labels of each drug, adjusting the parameters of the Rank-SVM, and forming the traditional Chinese medicine prescription evaluation model through training.
7. The device for assisting in determining the efficacy of a Chinese medicinal formulation according to claim 5, further comprising:
and the secondary training part is used for carrying out secondary training on the basis of the collected preset traditional Chinese medicine formulas and the traditional Chinese medicine formula evaluation model to form a personalized traditional Chinese medicine formula evaluation model matched with each doctor.
8. The device for assisting in determining the efficacy of a Chinese medicinal formulation according to claim 5, further comprising:
and the optimizing component is used for optimizing the traditional Chinese medicine prescription evaluation model based on the acquired patient medication feedback information.
9. An auxiliary judgment device for the drug effect of a traditional Chinese medicine prescription, which is characterized by comprising:
a memory for storing a computer program;
a processor for implementing the steps of the method for assisting in determining the efficacy of a Chinese medicinal formulation according to any one of claims 1 to 4 when executing the computer program.
10. A storage medium, wherein a computer program is stored on the storage medium, and when the computer program is executed by a processor, the method for assisting in determining the efficacy of a Chinese medicinal formulation according to any one of claims 1 to 4 is implemented.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010746846.5A CN111883263A (en) | 2020-07-29 | 2020-07-29 | Method, device, equipment and storage medium for assisting in judging drug effect of traditional Chinese medicine prescription |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010746846.5A CN111883263A (en) | 2020-07-29 | 2020-07-29 | Method, device, equipment and storage medium for assisting in judging drug effect of traditional Chinese medicine prescription |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN111883263A true CN111883263A (en) | 2020-11-03 |
Family
ID=73201188
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010746846.5A Pending CN111883263A (en) | 2020-07-29 | 2020-07-29 | Method, device, equipment and storage medium for assisting in judging drug effect of traditional Chinese medicine prescription |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111883263A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112420153A (en) * | 2020-11-26 | 2021-02-26 | 济南浪潮高新科技投资发展有限公司 | Method for improving traditional Chinese medicine prescription based on GAN |
| CN112820359A (en) * | 2021-02-24 | 2021-05-18 | 北京中医药大学东直门医院 | Liver injury prediction method, apparatus, device, medium, and program product |
| CN116072305A (en) * | 2023-02-17 | 2023-05-05 | 北京中兴正远科技有限公司 | Clinical trial data acquisition system |
-
2020
- 2020-07-29 CN CN202010746846.5A patent/CN111883263A/en active Pending
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112420153A (en) * | 2020-11-26 | 2021-02-26 | 济南浪潮高新科技投资发展有限公司 | Method for improving traditional Chinese medicine prescription based on GAN |
| CN112420153B (en) * | 2020-11-26 | 2022-11-15 | 山东浪潮科学研究院有限公司 | Method for improving traditional Chinese medicine prescription based on GAN |
| CN112820359A (en) * | 2021-02-24 | 2021-05-18 | 北京中医药大学东直门医院 | Liver injury prediction method, apparatus, device, medium, and program product |
| CN116072305A (en) * | 2023-02-17 | 2023-05-05 | 北京中兴正远科技有限公司 | Clinical trial data acquisition system |
| CN116072305B (en) * | 2023-02-17 | 2023-09-29 | 北京中兴正远科技有限公司 | Clinical trial data acquisition system |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111883263A (en) | Method, device, equipment and storage medium for assisting in judging drug effect of traditional Chinese medicine prescription | |
| CN108447536A (en) | Medication information processing method based on genius morbi and disease degree information | |
| CN108597587A (en) | A kind of veteran TCM doctor's experience intelligence succession and clinical aid decision-making system and method | |
| CN102841937A (en) | Health intelligent inquiring system | |
| CN105550526B (en) | A kind of tcm prescription effect quantitative approach and system based on Chinese medicine attribute | |
| CN106709520A (en) | Topic model based medical record classification method | |
| CN107095828A (en) | A kind of daily chemical product composition with skin barrier repair function and its production and use | |
| CN106692881A (en) | Traditional Chinese medicine composition for treating uterus myoma and preparation method and application of traditional Chinese medicine composition | |
| CN108648791A (en) | Prescriptions of traditional Chinese medicine method for pushing and system based on ICD codes | |
| CN109659005A (en) | Information-pushing method and system | |
| CN114417014A (en) | Basic knowledge system construction method taking human as center | |
| CN112420153B (en) | Method for improving traditional Chinese medicine prescription based on GAN | |
| CN114093455A (en) | Internet hospital rational medication engine construction method based on artificial intelligence | |
| Vaishnavi et al. | Comparative Study of Different ML And DL Algorithms For Indian Medicinal Leaf Detection | |
| CN102357191B (en) | Pharmaceutical composition for treating hashimoto disease | |
| CN112185503A (en) | Intelligent auxiliary diagnosis system and method for traditional Chinese medicine | |
| Lee et al. | Utilization Trends and Concentration Ratio of Korean Medicine: Based on the National Health Insurance Data | |
| CN101085062A (en) | Medicine for curing parkinson's disease and its preparing method | |
| Zhang et al. | Construction and Application of Prescription Knowledge Graph Based on Deep Learning | |
| CN116955497A (en) | Classification method for Chinese patent medicine data | |
| Mahalanabish | Deep Learning based Medicinal Plants Leaf Recognition | |
| Chen et al. | Parametric design of orthotics for abnormal arches in school-age children | |
| Liu et al. | Based on data mining to explore the medication rule of Professor Shao Zheng-bin in the treatment of coronary heart disease combined with type2 diabetes mellitus after PCI. | |
| CN107998205A (en) | A kind of Chinese herbs paste for treating rheumatism and preparation method thereof | |
| CN118280514A (en) | Method for determining combination of core medicine pair, related device and computer storage medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |