Traditional Chinese medicine temporary preparation dosage form recommendation method, system, equipment and medium

Abstract

The invention provides a traditional Chinese medicine temporary prescription preparation dosage form recommendation method, system and device and a medium, and the method comprises the steps: obtaining multi-dimensional prescription information of a target prescription and at least one reference prescription, the multi-dimensional prescription information at least comprises medicine composition information, component index information, medicine property information, function information and major treatment information, the at least one reference prescription comes from a traditional Chinese medicine temporary prescription preparation dosage form recommendation database, and the traditional Chinese medicine temporary prescription preparation dosage form recommendation database further comprises a preparation dosage form corresponding to the at least one reference prescription. The dosage forms of the preparation at least comprise pills, powder, paste and decoction. Calculating the prescription similarity between the target prescription and the at least one reference prescription according to the multi-dimensional prescription information of the target prescription and the multi-dimensional prescription information of the at least one reference prescription; and determining a target reference prescription according to the prescription similarity, and outputting a recommended dosage form according to the dosage form of the target reference prescription. The problem that the dosage form of an existing traditional Chinese medicine temporary preparation is not accurately selected is solved.

Description

Technical Field

[0001] This invention belongs to the field of traditional Chinese medicine technology, specifically, it relates to a method, system, equipment and medium for recommending dosage forms of traditional Chinese medicine prescriptions. Background Technology

[0002] Traditional Chinese medicine (TCM) prescriptions, also known as ad-hoc preparations, refer to the temporary processing of drugs into pills, powders, ointments, or medicated wines at the request of doctors for therapeutic purposes. With the support of national policies and the continuous improvement of people's medical needs and the accelerated pace of life, TCM prescriptions have ushered in a new era of comprehensive development. However, this rapid development has also brought its potential problems to the forefront, posing numerous challenges in clinical application, particularly in dosage form selection. In practice, patients and doctors frequently encounter difficulties in dosage form selection: when to use decoctions, when to use powders, and when to use pills, etc. Different dosage forms not only affect drug absorption and efficacy but may also impact patient compliance and treatment outcomes. Therefore, scientifically addressing the dosage form selection issue for TCM prescriptions has become an important topic in the field of TCM, and is of great significance for promoting the healthy development of TCM prescriptions and improving the quality of medical services.

[0003] In view of this, the present invention is proposed. Summary of the Invention

[0004] The technical problem to be solved by this invention is to overcome the shortcomings of the prior art. The purpose is to provide a method for recommending dosage forms of traditional Chinese medicine prescriptions. By matching reference prescriptions with high similarity in multidimensional prescription information, and recommending dosage forms based on the dosage forms of the reference prescriptions, a dosage form more suitable for the patient's condition can be obtained.

[0005] This invention provides a method for recommending dosage forms of traditional Chinese medicine prescriptions, using the following technical solution: The method involves acquiring multidimensional formula information for a target formula and at least one reference formula. This multidimensional formula information includes at least drug composition information, component index information, medicinal property information, function information, and indication information. The at least one reference formula is sourced from a database of recommended dosage forms for traditional Chinese medicine prescriptions. This database also includes the dosage forms corresponding to the at least one reference formula. The dosage forms include at least pills, powders, ointments, and decoctions. Based on the multidimensional formula information of the target formula and the multidimensional formula information of the at least one reference formula, calculate the formula similarity between the target formula and the at least one reference formula; Based on the similarity of the prescriptions, a target reference prescription is determined, and a recommended dosage form is output based on the dosage form of the target reference prescription.

[0006] This invention also provides a dosage form recommendation system for traditional Chinese medicine prescriptions, which is based on the above-mentioned method for recommending dosage forms for traditional Chinese medicine prescriptions, and adopts the following technical solution: The acquisition module is used to acquire multidimensional prescription information of the target prescription and at least one reference prescription. The multidimensional prescription information includes at least drug composition information, component index information, medicinal property information, function information and indication information. The at least one reference prescription comes from the Chinese medicine prescription preparation dosage form recommendation database. The Chinese medicine prescription preparation dosage form recommendation database also includes the preparation dosage form corresponding to the at least one reference prescription. The preparation dosage form includes at least pills, powders, ointments and decoctions. The calculation module is used to calculate the formula similarity between the target formula and the at least one reference formula based on the multidimensional formula information of the target formula and the multidimensional formula information of the at least one reference formula. The output module is used to determine the target reference prescription based on the prescription similarity and to output a recommended dosage form based on the dosage form of the target reference prescription.

[0007] The present invention also provides an electronic device, which includes a memory and a processor. The memory stores a computer program, and when the computer program is executed by the processor, the processor performs the steps of the method for recommending dosage forms of traditional Chinese medicine prescriptions as described in any one of claims 1-7.

[0008] The present invention also provides a computer-readable storage medium storing a computer program, which, when executed by a processor, causes the processor to perform the steps of the method for recommending dosage forms of traditional Chinese medicine prescriptions as described in any one of claims 1-7.

[0009] In summary, this invention provides a method for recommending dosage forms of traditional Chinese medicine (TCM) prescriptions, a system for recommending dosage forms of TCM prescriptions, an electronic device, and a computer-readable storage medium, which have the following advantages compared with the prior art: This method acquires multidimensional formula information from a database of target formulas and at least one reference formula from a database of recommended dosage forms for traditional Chinese medicine (TCM) prescriptions. Based on this multidimensional formula information, the similarity between the target formula and the at least one reference formula is calculated. A target reference formula is then determined based on this similarity, and a recommended dosage form is output based on the dosage form of the target reference formula. This approach utilizes existing reference formulas as data support while considering the impact of multidimensional information on dosage forms, improving the accuracy and reliability of recommendations and reducing errors caused by subjective judgment. Furthermore, it ensures that the selection of dosage forms for TCM prescriptions no longer relies solely on the individual clinical experience of physicians but rather on a large amount of clinical reference data, thus enhancing the scientific rigor of dosage form selection and accelerating the determination of dosage forms. Attached Figure Description

[0010] The accompanying drawings, as part of this invention, are provided to further illustrate the invention. The illustrative embodiments and descriptions of the invention are used to explain the invention, but do not constitute an undue limitation thereof. Clearly, the drawings described below are merely some embodiments, and those skilled in the art can obtain other drawings based on these drawings without any creative effort.

[0011] In the attached diagram: Figure 1 This is a flowchart of a method for recommending dosage forms of traditional Chinese medicine prescriptions provided by the present invention; Figure 2 This invention provides a structural diagram of a traditional Chinese medicine prescription dosage form recommendation system. Figure 3 This invention provides a thesaurus of thematic terms for classifying the functions of prescriptions. Figure 4 This is a thesaurus of the main indications for traditional Chinese medicine prescriptions provided by the present invention.

[0012] It should be noted that these accompanying drawings and textual descriptions are not intended to limit the scope of the invention in any way, but rather to illustrate the concept of the invention to those skilled in the art by referring to specific embodiments. Detailed Implementation

[0013] To enable those skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present application, and not all embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative effort should fall within the scope of protection of the present application.

[0014] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this application described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0015] The following will describe in detail, with reference to the accompanying drawings, a method and system for recommending dosage forms of traditional Chinese medicine prescriptions according to embodiments of this application. Figure 1 This is a flowchart of a method for recommending dosage forms of traditional Chinese medicine prescriptions provided by the present invention. Figure 1 The method for recommending dosage forms of traditional Chinese medicine prescriptions can be executed by a terminal device or a server, and the method includes: S101: Obtain multidimensional prescription information of the target prescription and at least one reference prescription. The multidimensional prescription information includes at least drug composition information, component index information, medicinal property information, function information, and indication information. The at least one reference prescription comes from the Chinese medicine prescription preparation dosage form recommendation database. The Chinese medicine prescription preparation dosage form recommendation database also includes the preparation dosage form corresponding to the at least one reference prescription. The preparation dosage form includes at least pills, powders, ointments, and decoctions. Specifically, multidimensional prescription information includes prescription information from multiple dimensions, at least including drug composition, component indicators, medicinal properties, functions, and indications. Dosage forms include pills, powders, ointments, and decoctions.

[0016] Specifically, an SQLite database was used to construct and manage a recommended dosage form database for traditional Chinese medicine (TCM) prescriptions. The database includes TCM prescriptions from the 2020 edition of the *Chinese Pharmacopoeia*, the *Dictionary of Clinical Practical Prescriptions*, the *Dictionary of Pills, Powders, Ointments and Pills (Ointment Section)*, and the *Complete Collection of TCM Ointment Prescriptions*. The database also standardizes the drug names and other prescription information, uniformly standardizing the drug names in eligible prescriptions to the pharmacopoeia names.

[0017] As an example, suppose the target formula is Buzhong Yiqi Tang (Supplement the Middle and Benefit Qi Decoction), whose ingredients include Astragalus membranaceus, Glycyrrhiza uralensis, Ginseng, Angelica sinensis, etc.; its component indicators include Astragaloside A, Verbena isoflavone glucoside, Ginsenoside Rg1, Ginsenoside Re, Ginsenoside Rb1, Glycyrrhizin, Glycyrrhizic acid, volatile oil, Ferulic acid, etc.; its medicinal properties include being warm in nature, mainly sweet in taste, primarily entering the spleen and stomach meridians, and being non-toxic; its functions include supplementing the middle and benefiting qi, raising yang and lifting prolapse; and it is mainly used to treat spleen and stomach qi deficiency syndrome, qi deficiency and prolapse syndrome, qi deficiency and fever syndrome, etc.

[0018] S102: Calculate the formula similarity between the target formula and the at least one reference formula based on the multidimensional formula information of the target formula and the multidimensional formula information of the at least one reference formula; Specifically, for each reference prescription, the similarity between the reference prescription and the target prescription is calculated. A high similarity indicates a match across multiple dimensions, with the difference possibly stemming from adjustments made based on the patient's constitution and condition. In this case, the dosage form of the reference prescription can provide doctors with a reference, improving drug absorption and efficacy. This multi-dimensional comprehensive consideration improves the accuracy of similarity calculations. For example, regarding drug composition, if most drugs are identical or belong to the same class, a high degree of similarity can be initially determined. Combining component indicators and medicinal properties, if the main active ingredients or medicinal properties of similar drugs are similar, further corroborates the similarity. Combining functions and indications, a high similarity in the prescription's functions and indications suggests that the patient's constitution and symptoms are similar. This avoids inaccurate conclusions caused by judging from only a single dimension.

[0019] Specifically, formula similarity calculation involves multiple dimensions, so a weighted average method can be used. For example, the similarity between the reference formula and the target formula in five dimensions—drug composition, component indicators, medicinal properties, functions, and indications—can be obtained, and then a weighted average can be applied to obtain the formula similarity. The specific weight values ​​can be set as needed. Other algorithms can also be selected for calculation based on the actual application.

[0020] S103: Based on the similarity of the prescriptions, determine the target reference prescription and output a recommended dosage form based on the dosage form of the target reference prescription.

[0021] Specifically, after obtaining the formula similarity of at least one reference formula, the formula similarities can be ranked, and the reference formula with the highest similarity can be determined as the target reference formula.

[0022] In another embodiment, a similarity threshold can be set to determine reference prescriptions whose similarity to the prescription is greater than the similarity threshold as target reference prescriptions.

[0023] Specifically, as an example, assuming the target reference formula is in pill form, the recommended dosage form output would be pills. The output format can be adaptively adjusted based on the device executing the recommendation method.

[0024] Furthermore, the step of calculating the formula similarity between the target formula and the at least one reference formula based on the multidimensional formula information of the target formula and the multidimensional formula information of at least one reference formula includes: calculating the drug composition similarity, component index similarity, medicinal property similarity, function similarity, and indication similarity between the target formula and the reference formula for each reference formula in the formula database; and calculating the formula similarity between the target formula and the reference formula based on the drug composition similarity, component index similarity, medicinal property similarity, function similarity, and indication similarity.

[0025] Specifically, calculating similarity from multiple dimensions can improve the accuracy of prescription similarity analysis. Through similarity analysis, prescriptions with similar ingredients, efficacy, and indications can be identified as references. For example, among two reference prescriptions for treating spleen deficiency diarrhea, if one has a higher similarity in drug composition and ingredient indicators to the target prescription, the doctor can consider using the dosage form of this similar prescription to better improve efficacy and treatment outcomes.

[0026] Furthermore, the step of calculating the drug composition similarity, component index similarity, medicinal property similarity, function similarity, and indication similarity between the target prescription and the reference prescription includes: calculating the drug composition similarity and component index similarity based on the Jaccard similarity coefficient algorithm; calculating the medicinal property similarity based on the cosine similarity algorithm; and calculating the function similarity and indication similarity based on the LDA topic model.

[0027] Specifically, the Jaccard similarity coefficient measures the similarity of two sets by calculating the ratio of their intersection to their union; that is, it calculates the proportion of common elements in both sets out of their total elements. A higher proportion indicates greater similarity between the two sets. Drug composition and component indicators are precisely presented in the form of sets.

[0028] As an example, suppose the target prescription consists of Astragalus membranaceus, Glycyrrhiza uralensis, and Ginseng; the reference prescription one consists of Astragalus membranaceus, Glycyrrhiza uralensis, Ginseng, Angelica sinensis, and Atractylodes macrocephala; and the reference prescription two consists of Astragalus membranaceus, Angelica sinensis, Poria cocos, Longan, and Polygala tenuifolia. It can be seen that the reference prescription one and the target prescription have 3 common elements, with a similarity of 60%, while the reference prescription two and the target prescription have only 1 common element, with a similarity of 20%. Therefore, it can be determined that the reference prescription one has a higher similarity in drug composition to the target prescription.

[0029] Furthermore, before calculating the similarity of medicinal properties based on the cosine similarity algorithm, the method further includes: the medicinal property information includes at least the information on the four qi, the five flavors, the meridian tropism, and the toxicity; based on a preset assignment method, the vectors of the four qi, the five flavors, the meridian tropism, and the toxicity are determined according to the information on the four qi, the five flavors, the meridian tropism, and the toxicity.

[0030] Specifically, taking the above example of Buzhong Yiqi Decoction, its four properties are warm, its five flavors are mainly sweet, its meridian tropism is spleen and stomach, and its toxicity is non-toxic. This is information in textual form. When calculating similarity, information in numerical form should be used. Therefore, before vectorizing the four properties, five flavors, meridian tropism, and toxicity, it is necessary to assign values ​​to them to quantify the textual information.

[0031] Specifically, according to historical Chinese herbal texts and the 2020 edition of the Chinese Pharmacopoeia, the four natures are divided into nine categories: extremely hot, hot, warm, slightly warm, neutral, cool, slightly cold, cold, and extremely cold. The five flavors are divided into eleven categories: sour, slightly sour, bitter, slightly bitter, sweet, slightly sweet, pungent, slightly pungent, salty, astringent, and bland. The meridians are divided into twelve categories: lung, heart, spleen, stomach, liver, gallbladder, kidney, pericardium, large intestine, small intestine, bladder, and triple burner. The toxicity is divided into four categories: non-toxic, highly toxic, toxic, and slightly toxic. In this embodiment, the total number of the four natures, five flavors, meridians, and toxicity values ​​for each prescription is recorded as 1.

[0032] Specifically, after assigning values, the properties of the entire prescription are vectorized based on the properties of the herbs in each prescription. Taking the four herbs of the Erdong Gao prescription as an example, the prescription consists of Asparagus cochinchinensis and Ophiopogon japonicus, a total of two herbs. The total four-qi value of Erdong Gao is recorded as 1, and the four-qi value of each herb accounts for 1 / 2. Asparagus cochinchinensis is cold in nature, so it is recorded as 1 / 2; Ophiopogon japonicus is slightly cold in nature, so it is recorded as 1 / 2; the remaining four-qi values ​​are 0. The four-qi values ​​of Erdong Gao are converted into a vector as (0, 0, 0, 0, 0, 0, 1 / 2, 1 / 2, 0). It should be noted that if two or more herbs in a prescription have the same four-qi value, then the corresponding four-qi value should be the sum of the four-qi values ​​of multiple herbs. The similarity calculation method for other medicinal property dimensions is the same.

[0033] Furthermore, the calculation of the medicinal property similarity based on the cosine similarity algorithm includes: determining the similarity of the four properties, the similarity of the five flavors, the similarity of the meridian tropism, and the similarity of toxicity based on the cosine similarity algorithm; and determining the medicinal property similarity based on the similarity of the four properties, the similarity of the five flavors, the similarity of the meridian tropism, and the similarity of toxicity.

[0034] Specifically, the cosine similarity algorithm measures the similarity between two vectors by calculating the cosine of the angle between them. In vector space, the smaller the angle between two vectors, the closer their cosine value is to 1, indicating that the two vectors are more similar.

[0035] Specifically, the similarity of drug properties can be calculated using a weighted average method. The weight parameters can be set according to clinical data. If the clinic is more concerned about a certain dimension of drug properties, the weight of that dimension can be increased, and the weights of other dimensions can be adjusted accordingly.

[0036] Furthermore, the calculation of the functional similarity and the therapeutic similarity based on the LDA topic model includes: predefining functional topics and therapeutic topics; generating functional topic probability distribution matrices and therapeutic topic probability distribution matrices respectively through the LDA model; and calculating the functional similarity corresponding to the functional topic probability distribution matrix and the therapeutic similarity corresponding to the therapeutic topic probability distribution matrix based on KL divergence.

[0037] Specifically, the efficacy and indication information in traditional Chinese medicine prescriptions are all textual. The LDA topic model preprocesses the text data, including word segmentation and stop word removal, to obtain a word set. For each topic category, the probability of that topic category is determined based on the correspondence between the topic and different words, as well as the frequency of occurrence of the corresponding words. Based on the probability of each topic category, a topic probability distribution is obtained, and then KL distance is used to calculate the similarity of efficacy and / or indications between prescriptions.

[0038] Specifically, based on the functional classification in *Formulary*, the functional themes are divided into 21 categories, with specific classification keywords as follows: Figure 3 The thesaurus of prescription functions in the text is shown. The main indications are divided into 10 categories based on the eight-principle syndrome categories in "Clinical Terminology of Traditional Chinese Medicine, Part 2: Syndromes," with specific classification thesaurus entries as follows: Figure 4 The thesaurus of prescription indications is shown in the figure.

[0039] Specifically, with Figure 3 Taking Topic 4, "Heat-Clearing Formulas," from the thesaurus of formula function classification, related functional terms include heat-clearing and detoxifying, heat-clearing and fire-purging, and heat-clearing and blood-cooling. Related indication terms include febrile diseases, heat syndromes, and fever. Similar thematic related terms are extracted, and their distribution frequencies are calculated to determine the frequency of Topic 4. This process is repeated for all topics to obtain their frequencies, generating the probability distribution of the formula's functional theme. This allows for the calculation of the functional similarity between the target formula and the reference formula.

[0040] Furthermore, calculating the formula similarity between the target formula and the reference formula based on the drug composition similarity, component index similarity, medicinal property similarity, functional similarity, and indication similarity includes: Based on the drug composition similarity, component index similarity, drug property similarity, function similarity, and indication similarity, determine the dimension weights corresponding to each dimension of similarity; The similarity of the prescriptions is calculated based on the similarity of each dimension and its corresponding dimension weight.

[0041] Specifically, since the similarity scores across different dimensions are not uniform and computation is difficult, it is necessary to calculate the corresponding dimensional weights based on the similarity scores of each dimension. The calculation formula is as follows: b1 = a1 / (a1 + a2 + a3 + a4 + a5) Where a1 represents drug composition similarity, a2 represents component index similarity, a3 represents drug property similarity, a4 represents function similarity, a5 represents indication similarity, and b1 represents the dimension weight corresponding to drug composition similarity.

[0042] b2 = a2 / (a1 + a2 + a3 + a4 + a5) Where b2 is the dimensional weight corresponding to the component index similarity.

[0043] b3 = a3 / (a1 + a2 + a3 + a4 + a5) Where b3 is the dimensional weight corresponding to the drug similarity.

[0044] b4 = a4 / (a1 + a2 + a3 + a4 + a5) Where b4 is the dimensional weight corresponding to the functional similarity.

[0045] b5 = a5 / (a1 + a2 + a3 + a4 + a5) Among them, b5 is the dimension weight corresponding to the similarity of the main treatment.

[0046] Specifically, based on the similarity of each dimension and its corresponding dimension weight, the formula for calculating the similarity of the prescriptions is as follows: M= b1×a1+ b2×a2+ b3×a3+ b4×a4+ b5×a5 Where M represents the similarity between the prescriptions.

[0047] Furthermore, if a particular dimension receives more attention in clinical practice, its weight can be increased, and the weights of other dimensions can be adjusted accordingly.

[0048] Figure 2 This is a structural diagram of a traditional Chinese medicine prescription dosage form recommendation system provided by the present invention. (See diagram below.) Figure 2 As shown, this embodiment of the invention also provides a traditional Chinese medicine (TCM) prescription formulation dosage form recommendation system, which executes the above-described TCM prescription formulation dosage form recommendation method, including: The acquisition module 201 is used to acquire multidimensional prescription information of the target prescription and at least one reference prescription. The multidimensional prescription information includes at least drug composition information, component index information, medicinal property information, function information and indication information. The at least one reference prescription comes from the Chinese medicine prescription preparation dosage form recommendation database. The Chinese medicine prescription preparation dosage form recommendation database also includes the preparation dosage form corresponding to the at least one reference prescription. The preparation dosage form includes at least pills, powders, ointments and decoctions. The calculation module 202 is used to calculate the formula similarity between the target formula and the at least one reference formula based on the multidimensional formula information of the target formula and the multidimensional formula information of the at least one reference formula. The output module 203 is used to determine the target reference prescription based on the prescription similarity and to output a recommended dosage form based on the dosage form of the target reference prescription.

[0049] In summary, this invention provides a method, system, equipment, and medium for recommending dosage forms of traditional Chinese medicine prescriptions, which has the following advantages compared with the prior art: This method acquires multidimensional formula information from a database of target formulas and at least one reference formula from a database of recommended dosage forms for traditional Chinese medicine (TCM) prescriptions. Based on this multidimensional formula information, the similarity between the target formula and the at least one reference formula is calculated. A target reference formula is then determined based on this similarity, and a recommended dosage form is output based on the dosage form of the target reference formula. This approach utilizes existing reference formulas as data support while considering the impact of multidimensional information on dosage forms, improving the accuracy and reliability of recommendations and reducing errors caused by subjective judgment. Furthermore, it ensures that the selection of dosage forms for TCM prescriptions no longer relies solely on the individual clinical experience of physicians but rather on a large amount of clinical reference data, thus enhancing the scientific rigor of dosage form selection and accelerating the determination of dosage forms.

[0050] The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention in any way. Although the present invention has been disclosed above with reference to preferred embodiments, it is not intended to limit the present invention. Any person skilled in the art can make some modifications or alterations to the above-described technical content to create equivalent embodiments without departing from the scope of the present invention. The implementation schemes in the above embodiments can also be further combined or replaced. Any simple modifications, equivalent changes and alterations made to the above embodiments based on the technical essence of the present invention without departing from the scope of the present invention shall still fall within the scope of the present invention.

Claims

1. A method for recommending dosage forms of traditional Chinese medicine prescriptions, characterized in that, include: The method involves acquiring multidimensional formula information for a target formula and at least one reference formula. This multidimensional formula information includes at least drug composition information, component index information, medicinal property information, function information, and indication information. The at least one reference formula is sourced from a database of recommended dosage forms for traditional Chinese medicine prescriptions. This database also includes the dosage forms corresponding to the at least one reference formula. The dosage forms include at least pills, powders, ointments, and decoctions. Based on the multidimensional formula information of the target formula and the multidimensional formula information of the at least one reference formula, calculate the formula similarity between the target formula and the at least one reference formula; Based on the similarity of the prescriptions, a target reference prescription is determined, and a recommended dosage form is output based on the dosage form of the target reference prescription.

2. The method for recommending dosage forms of traditional Chinese medicine prescriptions according to claim 1, characterized in that, The step of calculating the formula similarity between the target formula and the at least one reference formula based on the multidimensional formula information of the target formula and the multidimensional formula information of at least one reference formula includes: For each reference prescription in the prescription database, the similarity of drug composition, component index, medicinal properties, function, and indications between the target prescription and the reference prescription are calculated respectively. The similarity between the target prescription and the reference prescription is calculated based on the similarity of the drug composition, the similarity of the component indicators, the similarity of the medicinal properties, the similarity of the functions, and the similarity of the indications.

3. The method for recommending dosage forms of traditional Chinese medicine prescriptions according to claim 2, characterized in that, The calculation of the drug composition similarity, component index similarity, medicinal property similarity, function similarity, and indication similarity between the target prescription and the reference prescription includes: Based on the Jaccard similarity coefficient algorithm, the similarity of the drug composition and the similarity of the component indicators are calculated respectively. The similarity of the medicinal properties is calculated based on the cosine similarity algorithm; Based on the LDA topic model, the functional similarity and the therapeutic similarity are calculated respectively.

4. The method for recommending dosage forms of traditional Chinese medicine prescriptions according to claim 3, characterized in that, Before calculating the drug similarity based on the cosine similarity algorithm, the method further includes: The medicinal properties information includes at least information on the four natures, five flavors, meridian tropism, and toxicity; Based on the preset assignment method, the four qi vectors, five flavor vectors, meridian tropism vectors, and toxicity vectors are determined according to the four qi information, the five flavor information, the meridian tropism information, and the toxicity information.

5. The method for recommending dosage forms of traditional Chinese medicine prescriptions according to claim 4, characterized in that, The calculation of the drug similarity based on the cosine similarity algorithm includes: Based on the cosine similarity algorithm, the similarity of the four qi, the similarity of the five flavors, the similarity of meridian tropism, and the similarity of toxicity are determined; The similarity of medicinal properties is determined based on the similarity of the four properties, the five flavors, the meridian tropism, and the toxicity.

6. The method for recommending dosage forms of traditional Chinese medicine prescriptions according to claim 3, characterized in that, The calculation of the functional similarity and the therapeutic similarity based on the LDA topic model includes: Predefined functional themes and therapeutic themes; The probability distribution matrix of the application topic and the probability distribution matrix of the treatment topic were generated using the LDA model. The similarity of functions corresponding to the probability distribution matrix of functional topics and the similarity of treatment subjects corresponding to the probability distribution matrix of treatment subjects are calculated based on KL divergence.

7. The method for recommending dosage forms of traditional Chinese medicine prescriptions according to claim 2, characterized in that, The step of calculating the formula similarity between the target formula and the reference formula based on the drug composition similarity, component index similarity, medicinal property similarity, functional similarity, and indication similarity includes: Based on the drug composition similarity, component index similarity, drug property similarity, function similarity, and indication similarity, determine the dimension weights corresponding to each dimension of similarity; The similarity of the prescriptions is calculated based on the similarity of each dimension and its corresponding dimension weight.

8. A dosage form recommendation system for traditional Chinese medicine prescriptions, characterized in that, include: The acquisition module is used to acquire multidimensional prescription information of the target prescription and at least one reference prescription. The multidimensional prescription information includes at least drug composition information, component index information, medicinal property information, function information and indication information. The at least one reference prescription comes from the Chinese medicine prescription preparation dosage form recommendation database. The Chinese medicine prescription preparation dosage form recommendation database also includes the preparation dosage form corresponding to the at least one reference prescription. The preparation dosage form includes at least pills, powders, ointments and decoctions. The calculation module is used to calculate the formula similarity between the target formula and the at least one reference formula based on the multidimensional formula information of the target formula and the multidimensional formula information of the at least one reference formula. The output module is used to determine the target reference prescription based on the prescription similarity and to output a recommended dosage form based on the dosage form of the target reference prescription.

9. An electronic device comprising a memory and a processor, the memory storing a computer program, characterized in that, When the computer program is executed by the processor, it causes the processor to perform the steps of the method for recommending dosage forms of traditional Chinese medicine prescriptions as described in any one of claims 1-7.

10. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by the processor, it causes the processor to perform the steps of the method for recommending dosage forms of traditional Chinese medicine prescriptions as described in any one of claims 1-7.

Images