Drug monitoring and management system based on clinical pharmacy

Abstract

The invention relates to the technical field of medicine monitoring and management, aims to improve the accuracy and rationality of prescription auditing and realize intelligent monitoring and management of medicine matching, dosage evaluation and interaction analysis, and particularly discloses a medicine monitoring and management system based on clinical pharmacy. Comprising a prescription identification module, a medicine combination analysis module, a medicine dosage analysis module, a medicine combination analysis module, a comprehensive analysis module, a management and control processing module, a display terminal and a cloud database. According to the method, standardized prescription data information is obtained by identifying, analyzing and processing key fields of a prescription, then based on a set algorithm model, the matching degree, dosage rationality and interaction risk of drug matching are evaluated respectively, corresponding evaluation values are calculated, the evaluation values are comprehensively analyzed to obtain a drug comprehensive evaluation value, and the drug comprehensive evaluation value is obtained. Whether the prescription is abnormal or not is judged, if the prescription is abnormal, the prescription is managed, controlled and analyzed, comprehensive monitoring and management of the prescription medicine are achieved, and medication safety of a patient is guaranteed.

Description

Technical Field

[0001] The present invention relates to the technical field of drug monitoring and management, in particular to a drug monitoring and management system based on clinical pharmacy. Background Art

[0002] The continuous development of clinical pharmacy has made the safety and rationality of drug therapy an important focus of medical management. Rational use of drugs can not only improve the treatment effect, but also effectively reduce the risk of drug use and reduce unnecessary medical costs. However, in actual clinical applications, due to the wide variety of drugs and complex prescription structures, medical institutions still face many challenges in drug monitoring and management, such as unreasonable drug matching, lack of standardization of drug dosage, and high risk of drug interactions. At present, most medical institutions still rely on manual review of prescriptions to evaluate drug rationality, but this method has limitations such as low review efficiency, reliance on the experience of reviewers, and difficulty in fully covering potential drug risks. In addition, although some intelligent auxiliary drug monitoring systems have the function of early warning of adverse drug reactions, they often lack comprehensive analysis of drug matching, rationality of dosage and interaction in prescriptions, resulting in insufficient accuracy and reliability of the system in actual application, making it difficult to meet the needs of clinical rational drug use; In order to solve the above defects, a technical solution is now provided. Summary of the invention

[0003] The purpose of the present invention is to solve the problems that the existing drug monitoring and management process relies on manual experience to fully cover potential drug risks, and lacks comprehensive analysis of prescription drug matching, dosage rationality and interaction, resulting in insufficient accuracy and reliability, and proposes a drug monitoring and management system based on clinical pharmacy.

[0004] The purpose of the present invention can be achieved through the following technical solutions: Drug monitoring and management system based on clinical pharmacy, including: The drug combination analysis module is used to evaluate the matching degree of the drug combinations in the prescription and calculate the matching evaluation value of the drug combinations, where: The matching assessment value of drug pairing is determined by the indication matching value, target overlap value, and drug duplication value; The drug dosage analysis module is used to evaluate the rationality of the drug dosage in the prescription and calculate the reasonable evaluation value of the drug dosage, where: The reasonable evaluation value of drug dosage is determined by the dosage, frequency, course of treatment and patient characteristics of each drug; The drug combination analysis module is used to evaluate the risk of drug interactions in prescriptions and calculate the risk assessment value of drug interactions, where: The risk assessment value of drug interactions is jointly determined by the number of matching groups of substance ingredient types at each risk level; The comprehensive analysis module is used to extract the matching evaluation value of drug combination, the reasonable evaluation value of drug dosage and the risk evaluation value of drug interaction, and perform normalization processing to obtain the comprehensive evaluation value of the drug in the prescription, thereby determining whether to generate a prescription abnormality signal; The control and processing module is used to receive prescription abnormality signals, thereby performing control and analysis on the prescription and obtaining a recommended solution for prescription drug adjustment.

[0005] Furthermore, it also includes a prescription recognition module for identifying and analyzing key fields of the prescription, obtaining standardized prescription data information, and converting it into a structured data format and transmitting it to a cloud database for storage; The specific process of identifying and analyzing the key fields of the prescription is as follows: By detecting whether there is an unfilled blank field in the prescription form, if it is detected that there is no unfilled blank field in the prescription form, a prescription form passing signal is generated; Based on the generated prescription signal, the layout structure of the prescription is parsed, the position and content of key fields are identified, the prescription data information is obtained, and the prescription data information is subjected to text standardization and drug entity matching to obtain standardized prescription data information.

[0006] Furthermore, the process of solving the indication matching value is as follows: By extracting the names of the drugs in the prescription and the characteristics of the patient's disease, the names of the drugs in the prescription and the characteristics of the patient's disease are obtained; According to the standardized drug indication database, the indications of each drug in the prescription are matched to obtain the corresponding indication list of each drug; By comparing and analyzing the indications corresponding to each drug with the patient's disease characteristics, according to the formula: , get the indication matching value S Indication , where D represents the patient's disease characteristics, I i represents the indication corresponding to the i-th drug, Sim(D, I i ) represents the similarity value between indications and diseases, and N represents the total number of drugs; The process of solving the similarity value is as follows: The description language of the patient's disease characteristics is processed with standard medical terminology to obtain a description text of the disease; Extract key medical terms from the list of indications corresponding to the drug, remove stop words and modifiers that are not keywords, and obtain the description text of the drug indications; The description texts of diseases and drug indications are assigned overlapping keywords to obtain similar values.

[0007] Furthermore, the process of solving the target overlap value is as follows: By matching the names of each drug with the drug target type table, the target type of each drug is obtained, and the target overlap value S is calculated. Target , the specific calculation formula is: , where T i and T i* represents the target set of the i-th and i*-th drugs, and i≠i*, 1≤i<i*≤N, |T i ∩T i* | represents the number of common targets of two drugs, |T i ∪T i* | represents the total number of targets of the two drugs.

[0008] Furthermore, the process of solving the drug duplication value is as follows: The drug duplication value S is calculated by matching the drug name with the WHO ATC code table to obtain the hierarchy of each drug. Duplication , the specific calculation formula is: , where C i and C i* represents the hierarchy of the i-th and i*-th drugs, δ(C i , C i* ) represents the hierarchical similarity value. If C i =C i* (indicated that the levels are exactly the same), then the level similarity value is 1. If C i =1 / 3C i* (indicated by the same level 1 / 3), the level similarity value is 0.75. If C i =1 / 2C i* (indicated as level 1 / 2 are the same), the level similarity value is 0.5, otherwise, the level similarity value is 0.

[0009] Furthermore, the process of solving the matching evaluation value of drug combination is as follows: Extract indication matching value S Indication , target overlap value S Target and drug repeat value S Duplication The value of is normalized according to the formula: , the matching evaluation value SP of the drug combination is obtained, where e represents the set natural constant, μ1, μ2 and μ3 represent the set weight coefficients respectively, and μ1>μ2>μ3.

[0010] Furthermore, the process of solving the reasonable evaluation value of drug dosage is as follows: By extracting the dosage Y of each drug in the prescription Dose i, frequency Y Fre i 、Course Y Treat i And the patient characteristics influence coefficient λ for comprehensive analysis, according to the formula: , and obtain the reasonable evaluation value GP of drug dosage, where Q Dose i , Q Fre i and Q Treat i Represent the standard dose, standard frequency and standard course of treatment of the i-th drug, ΔY Dose i , ΔY Fre i and ΔY Treat i They represent the set allowable dose error value, allowable frequency error value and allowable treatment course error value of the i-th drug, μ4, μ5 and μ6 represent the set weight coefficients, and μ4>μ6>μ5; Among them, the patient characteristics influence coefficient is determined by the patient's height, weight, age, serum creatinine and treatment frequency in the prescription.

[0011] Furthermore, the process of solving the number of material component type matching groups for each risk level is as follows: By extracting the types of material components of each drug in the prescription and constructing a set, each drug in the prescription is combined in pairs to form a drug combination set; In each drug combination, the types of material components contained therein are then combined in pairs to form a set of material component type matching groups; Compare the set of material component type matching groups with the material component type risk table to determine the risk level of each material component type matching group. According to the risk level of each material component type matching group, each material component type matching group falls into the corresponding risk level. Count the number of material component type matching groups at each risk level to obtain the number of material component type matching groups at each risk level and mark it as W. zhing x , where x represents the risk level, x=1, 2, 3. When x=1, it indicates high-level risk, when x=2, it indicates medium-level risk, and when x=3, it indicates low-level risk.

[0012] Furthermore, the process of solving the risk assessment value of drug interaction is as follows: According to the formula: , and obtain the risk assessment value HP of drug interaction, where W 总 It represents the total number of material component type matching groups of risk levels, μ7, μ8 and μ9 represent the set weight coefficients respectively, and μ7>μ8>μ9.

[0013] Furthermore, the specific process of controlling and analyzing the prescription is as follows: According to the captured prescription abnormality signal, the matching evaluation value of drug combination, the reasonable evaluation value of drug dosage and the risk evaluation value of drug interaction are retrieved in sequence, and compared and analyzed with the corresponding preset threshold value or reference comparison interval respectively. According to the comparison result, the corresponding abnormality signal is generated, and the matching deviation value of drug combination, the reasonable deviation value of drug dosage and the risk deviation value of drug interaction are calculated in sequence based on the abnormal signal; At the same time, each deviation value is matched and analyzed with the stored matching deviation status table, reasonable deviation status table and risk deviation status table to determine the corresponding matching deviation level, reasonable deviation level and risk deviation level, and based on each deviation level, the corresponding drug combination adjustment suggestion plan, drug dosage adjustment suggestion plan and drug interaction adjustment suggestion plan are generated, thereby integrating to form a prescription drug adjustment suggestion plan.

[0014] Compared with the known prior art, the technical solution provided by the present invention has the following beneficial effects: 1. The present invention ensures the accuracy and completeness of the data by identifying and analyzing the key fields of the prescription, detecting blank fields, parsing the layout structure, identifying the position and content of key fields, and performing text standardization and drug entity matching processing, and then converting the prescription data into a structured data format, thereby providing reliable data support for subsequent analysis.

[0015] 2. The present invention, through the set algorithm model, respectively evaluates and analyzes the matching degree of drug combination in the prescription, the rationality of drug dosage and the risk of drug interaction, obtains the matching evaluation value of drug combination, the rationality evaluation value of drug dosage and the risk evaluation value of drug interaction, and comprehensively analyzes the obtained evaluation values ​​to obtain the comprehensive evaluation value of the drugs in the prescription, and determines whether to generate a prescription abnormality signal based on this, and performs management and control analysis on the prescription to generate a prescription drug adjustment recommendation plan, thereby realizing accurate processing and comprehensive evaluation of drug combination, drug dosage and drug interaction, and can more accurately identify potential drug risks in prescriptions, and provide medical personnel with more accurate and reliable prescription quality evaluation results, which helps them make more informed medical decisions. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the prior art descriptions are briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention, and for ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0017] Figure 1 It is the overall module block diagram of the present invention. DETAILED DESCRIPTION

[0018] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0019] like Figure 1 As shown, the drug monitoring and management system based on clinical pharmacy includes: prescription identification module, drug combination analysis module, drug dosage analysis module, drug combination analysis module, comprehensive analysis module, control and processing module, display terminal and cloud database; The cloud database is used to store standardized prescription data information, drug target category table, WHO ATC code table, material ingredient category risk table, matching deviation status table, reasonable deviation status table and risk deviation status table.

[0020] The prescription recognition module is used to identify and analyze the key fields of the prescription. The specific analysis process is as follows: Use the CNN+LSTM model to detect whether there are any unfilled blank fields in the prescription. If it is detected that there are unfilled blank fields in the prescription, a prescription failure signal is generated. If it is detected that there are no unfilled blank fields in the prescription, a prescription pass signal is generated. According to the generated prescription pass signal, a deep learning-based target detection model (such as Faster R-CNN) is used to parse the layout structure of the prescription, identify the location and content of key fields, obtain prescription data information, and perform text standardization and drug entity matching on the prescription data information to obtain standardized prescription data information. The standardized prescription data information is converted into a structured data format (JSON / SQL) and transmitted to a cloud database for storage.

[0021] It should be noted that text standardization includes but is not limited to: Character conversion: convert full-width / half-width characters, and convert between traditional and simplified Chinese to ensure uniform text format; Spelling correction: Use OCR post-processing + NLP models (such as SpellChecker / BERT) to perform spelling correction on drug names and dosage units; Field alignment: Match common drug field formats (such as dosage mg / g / ml) through regular expressions; Drug entity matching includes but is not limited to: Drug name matching: Based on the National Medical Products Administration (NMPA) database, WHO ATC classification, Micromedex and other standard libraries, the drug names are proofread and identified through fuzzy matching + TF-IDF / BERT vector matching; Dosage unit conversion: Ensure that all dosage units comply with national standards, such as converting "0.5g" to "500mg"; Standardization of medication frequency: Unify different expressions into a standard format, such as: "three times a day" to "3 times a day", and once every 8 hours to "q8h" (clinical standard abbreviation).

[0022] The drug combination analysis module is used to detect the drug combination status information in the prescription, thereby evaluating and analyzing the matching degree of the drug combination in the prescription. The specific analysis process is as follows: By extracting the names of the drugs in the prescription and the characteristics of the patient's disease, the names of the drugs in the prescription and the characteristics of the patient's disease are obtained; According to the standardized drug indication database, the indications of each drug in the prescription are matched to obtain the corresponding indication list of each drug; By comparing and analyzing the indications corresponding to each drug with the patient's disease characteristics, according to the formula: , get the indication matching value S Indication , where D represents the patient's disease characteristics, I i represents the indication corresponding to the i-th drug, Sim(D, I i ) represents the similarity value between indications and diseases, and N represents the total number of drugs; The process of solving the similarity value is as follows: The description language of the patient's disease characteristics is standardized through NLP terminology (such as the UMLS vocabulary), converted into standard medical terms, and the description text of the disease is obtained; Extract key medical terms from the list of indications corresponding to the drug, remove stop words, modifiers (such as can be used for, applicable to) and other non-keywords, and obtain the description text of the drug indications; The description texts of diseases and drug indications are assigned keywords for overlapping to obtain similar values. The specific overlapping assignment rules are as follows: Complete overlap: The keywords of the disease are completely included in the keywords of the drug indications, and the value is Sim(D, I i ) = 1.0; Partial overlap: The keywords of the disease and the keywords of the drug indications partially overlap, and Sim (D, I) is weighted according to the number of overlapping words.i ) = 0.6; No overlap: The keywords of the disease and the drug indications do not overlap, and Sim(D, I i ) = 0; By matching the names of each drug with the drug target type table, the target type of each drug is obtained, and the target overlap value S is calculated. Target , the specific calculation formula is: , where T i and T i* represents the target set of the i-th and i*-th drugs, and i≠i*, 1≤i<i*≤N, |T i ∩T i* | represents the number of common targets of two drugs, |T i ∪T i* | represents the total number of targets of the two drugs; The drug duplication value S is calculated by matching the drug name with the WHO ATC code table to obtain the hierarchy of each drug. Duplication , the specific calculation formula is: , where C i and C i* represents the hierarchy of the i-th and i*-th drugs, δ(C i , C i* ) represents the hierarchical similarity value. If C i =C i* (indicated that the levels are exactly the same), then the level similarity value is 1. If C i =1 / 3C i* (indicated by the same level 1 / 3), the level similarity value is 0.75. If C i =1 / 2C i* (indicated as level 1 / 2 are the same), the level similarity value is 0.5, otherwise, the level similarity value is 0; Extract indication matching value S Indication , target overlap value S Target and drug repeat value S Duplication The value of is normalized according to the formula: , the matching evaluation value SP of the drug combination is obtained, wherein e represents the set natural constant, μ1, μ2 and μ3 represent the set weight coefficients respectively, and μ1>μ2>μ3, and the setting of the specific values ​​of μ1, μ2 and μ3 is made by those skilled in the art in the specific case, and the weight coefficient is used to balance the proportion of each data in the formula calculation, thereby promoting the accuracy of the calculation result.

[0023] The drug dosage analysis module is used to detect the dosage status of the drugs in the prescription, thereby evaluating and analyzing the rationality of the drug dosage in the prescription. The specific analysis process is as follows: The height, weight, age, serum creatinine and treatment frequency of the patient in the prescription were obtained by extracting them and calibrating them as Rcm, Rkg, Rag, Rck and Rdf, respectively, according to the formula: , the patient characteristic influence coefficient λ is obtained, where Ω represents the gender value. When the patient is female, Ω is 61.2, and when the patient is male, Ω is 72, indicating that p1, p2, and p3 represent the weight coefficients of body surface area, renal function value, and treatment frequency, respectively, and p3>p2>p1; By extracting the dosage, frequency and course of treatment of each drug in the prescription, we can obtain the dosage, frequency and course of treatment of each drug in the prescription and mark them as Y Dose i , Y Fre i and Y Treat i , according to the formula: , and obtain the reasonable evaluation value GP of drug dosage, where Q Dose i , Q Fre i and Q Treat i Represent the standard dose, standard frequency and standard course of treatment of the i-th drug, ΔY Dose i , ΔY Fre i and ΔY Treat i They respectively represent the set allowable dosage error value, allowable frequency error value and allowable treatment course error value of the ith drug, μ4, μ5 and μ6 respectively represent the set weight coefficients, and μ4>μ6>μ5.

[0024] The drug combination analysis module is used to detect the interaction status information of drugs in the prescription, thereby evaluating and analyzing the risk of drug interaction in the prescription. The specific analysis process is as follows: By obtaining the types of material components of each drug in the prescription, the types of material components of each drug in the prescription are obtained, and a set S is constructed i , the specific expression is: S i ={F ij}={F i1 , F i2 , …F iM}, where F ij represents the type of the jth substance component in the i-th drug; Combine the drugs in the prescription in pairs to form a drug combination set. The specific expression is: K = {(S i , S i* )|i≠i*,i,j∈[1,N]}; In each drug combination (S i , S i* ), and then the types of material components contained therein are combined in pairs to form a set of material component type matching groups. The specific expression is: G={(F iN , F jM )|F iN ∈S i , F jM ∈S i* , i≠j}, where G represents the set of all possible material component type matching groups, and each material component type matching group (F iN , F jM ) indicates a specific combination of substance types between two drugs; Example: Assume that the prescription contains drug A and drug B, and the types of their material ingredients are: Drug A:S A ={F A1 , F A2}; Drug B: S B ={F B1 , F B2}; Then the material component type matching group set is: G={(F A1 , F B1 ), (F A1 , F B2 ), (F A2 , F B1 ), (F A2 , F B2 )}.

[0025] Compare the material component type matching group set with the material component type risk table to determine the risk level of each material component type matching group. The risk level includes high risk, medium risk and low risk. According to the risk level of each material component type matching group, each material component type matching group falls into the corresponding risk level. The number of material component type matching groups at each risk level is counted and marked as W. zhing x , according to the formula: , the risk assessment value HP of drug interaction is obtained, where x represents the risk level, x=1,2,3. When x=1, it indicates high risk, when x=2, it indicates medium risk, and when x=3, it indicates low risk. 总It represents the total number of material component type matching groups of risk levels, μ7, μ8 and μ9 represent the set weight coefficients respectively, and μ7>μ8>μ9.

[0026] The comprehensive analysis module is used to extract the matching evaluation value SP of drug combination, the reasonable evaluation value GP of drug dosage and the risk evaluation value HP of drug interaction for normalization according to the formula: , the comprehensive evaluation value SGHP of the prescription is obtained, where θ1, θ2 and θ3 represent the set correction factor coefficients respectively; The prescription drug comprehensive evaluation threshold is set to SGHY, and the prescription drug comprehensive evaluation value SGHP is compared and analyzed with the drug comprehensive evaluation threshold SGHY. When the prescription drug comprehensive evaluation value SGHP is greater than or equal to the drug comprehensive evaluation threshold SGHY, a prescription normal signal is generated; when the prescription drug comprehensive evaluation value SGHP is less than the drug comprehensive evaluation threshold SGHY, a prescription abnormal signal is generated; The generated prescription exception signal is sent to the management and control processing module.

[0027] The control and processing module is used to receive prescription abnormality signals, and then perform control and analysis on the prescription. The specific analysis process is as follows: According to the captured prescription abnormality signal, the matching evaluation value of the drug combination is retrieved, and compared with the preset matching evaluation threshold of the drug combination for analysis. If the matching evaluation value of the drug combination is less than the preset matching evaluation threshold, a drug combination abnormality signal is generated. According to the generated drug combination abnormality signal, the matching evaluation value of the drug combination and the preset matching evaluation threshold are extracted for difference calculation to obtain the matching deviation value of the drug combination; Matching and analyzing the matching deviation value of the drug combination with the stored matching deviation state table to obtain the matching deviation level of the drug combination, and each matching deviation value of the drug combination corresponds to a matching deviation level, and matching it with the drug combination adjustment suggestion plan corresponding to the matching deviation level to obtain the drug combination adjustment suggestion plan of the prescription; If the matching evaluation value of the drug combination is greater than or equal to the preset matching evaluation threshold, the reasonable evaluation value of the drug dosage is retrieved and compared with the preset reasonable evaluation threshold of the drug dosage. If the reasonable evaluation value of the drug dosage is greater than the preset reasonable evaluation threshold, an abnormal drug dosage signal is generated. Based on the generated abnormal drug dosage signal, the reasonable evaluation value of the drug dosage and the preset reasonable evaluation threshold are extracted for difference calculation to obtain a reasonable deviation value of the drug dosage. The reasonable deviation value of the drug dosage is matched and analyzed with the stored reasonable deviation status table to obtain the reasonable deviation level of the drug dosage, and each reasonable deviation value of the drug dosage corresponds to a reasonable deviation level, and at the same time, it is matched with the drug dosage adjustment suggestion plan corresponding to the reasonable deviation level to obtain the drug dosage adjustment suggestion plan of the prescription; If the reasonable assessment value of the drug dosage is less than or equal to the preset reasonable assessment threshold, the risk assessment value of the drug interaction is retrieved and compared with the preset reference comparison interval for analysis; if the risk assessment value of the drug interaction is outside the preset reference comparison interval, a drug interaction abnormality signal is generated; based on the generated drug interaction abnormality signal, the risk assessment value of the drug interaction is extracted and the difference between the preset reference comparison median is calculated to obtain the risk deviation value of the drug interaction, wherein the reference comparison median is obtained by taking the average of the minimum and maximum values ​​of the reference comparison interval; The risk deviation value of the drug interaction is matched and analyzed with the stored risk deviation status table to obtain the risk deviation level of the drug interaction, and each risk deviation value of the drug interaction corresponds to a risk deviation level, and the risk deviation value is matched with the drug interaction adjustment suggestion plan corresponding to the risk deviation level to obtain the drug interaction adjustment suggestion plan of the prescription; The prescription drug adjustment suggestion plan is composed of the prescription drug combination adjustment suggestion plan, the drug dosage adjustment suggestion plan and the drug interaction adjustment suggestion plan, and is displayed on the display terminal.

[0028] The preferred embodiments of the present invention disclosed above are only used to help explain the present invention. The preferred embodiments do not describe all the details in detail, nor do they limit the invention to only specific implementation methods. Obviously, many modifications and changes can be made according to the content of this specification. This specification selects and specifically describes these embodiments in order to better explain the principles and practical applications of the present invention, so that those skilled in the art can understand and use the present invention well. The present invention is limited only by the claims and their full scope and equivalents.

Claims

1. A drug monitoring and management system based on clinical pharmacy, characterized in that: include: The drug combination analysis module is used to evaluate the matching degree of the drug combinations in the prescription and calculate the matching evaluation value of the drug combinations, where: The matching assessment value of drug pairing is determined by the indication matching value, target overlap value, and drug duplication value; The drug dosage analysis module is used to evaluate the rationality of the drug dosage in the prescription and calculate the reasonable evaluation value of the drug dosage, where: The reasonable evaluation value of drug dosage is determined by the dosage, frequency, course of treatment and patient characteristics of each drug; The drug combination analysis module is used to evaluate the risk of drug interactions in prescriptions and calculate the risk assessment value of drug interactions, where: The risk assessment value of drug interactions is jointly determined by the number of matching groups of substance ingredient types at each risk level; The comprehensive analysis module is used to extract the matching evaluation value of drug combination, the reasonable evaluation value of drug dosage and the risk evaluation value of drug interaction, and perform normalization processing to obtain the comprehensive evaluation value of the drug in the prescription, thereby determining whether to generate a prescription abnormality signal; The control and processing module is used to receive prescription abnormality signals, thereby performing control and analysis on the prescription and obtaining a recommended solution for prescription drug adjustment.

2. The drug monitoring and management system based on clinical pharmacy according to claim 1 is characterized in that: It also includes a prescription recognition module for identifying and analyzing key fields of a prescription, obtaining standardized prescription data information, and converting it into a structured data format and transmitting it to a cloud database for storage; The specific process of identifying and analyzing the key fields of the prescription is as follows: By detecting whether there is an unfilled blank field in the prescription form, if it is detected that there is no unfilled blank field in the prescription form, a prescription form passing signal is generated; Based on the generated prescription signal, the layout structure of the prescription is parsed, the position and content of key fields are identified, the prescription data information is obtained, and the prescription data information is subjected to text standardization and drug entity matching to obtain standardized prescription data information.

3. The drug monitoring and management system based on clinical pharmacy according to claim 1 is characterized in that: The process of solving the indication matching value is as follows: By extracting the names of the drugs in the prescription and the characteristics of the patient's disease, the names of the drugs in the prescription and the characteristics of the patient's disease are obtained; According to the standardized drug indication database, the indications of each drug in the prescription are matched to obtain the corresponding indication list of each drug; By comparing and analyzing the indications corresponding to each drug with the patient's disease characteristics, according to the formula: , get the indication matching value S Indication , where D represents the patient's disease characteristics, I i represents the indication corresponding to the i-th drug, Sim(D, I i ) represents the similarity value between indications and diseases, and N represents the total number of drugs; The process of solving the similarity value is as follows: The description language of the patient's disease characteristics is processed with standard medical terminology to obtain a description text of the disease; Extract key medical terms from the list of indications corresponding to the drug, remove stop words and modifiers that are not keywords, and obtain the description text of the drug indications; The description texts of diseases and drug indications are assigned overlapping keywords to obtain similar values.

4. The drug monitoring and management system based on clinical pharmacy according to claim 1 is characterized in that: The process of solving the target overlap value is as follows: By matching the names of each drug with the drug target type table, the target type of each drug is obtained, and the target overlap value S is calculated. Target , the specific calculation formula is: , where T i and T i* represents the target set of the i-th and i*-th drugs, and i≠i*, 1≤i<i*≤N, |T i ∩T i* | represents the number of common targets of two drugs, |T i ∪T i* | represents the total number of targets of the two drugs.

5. The drug monitoring and management system based on clinical pharmacy according to claim 1 is characterized in that: The process of solving the repeated values ​​of drugs is as follows: The drug duplication value S is calculated by matching the drug name with the WHO ATC code table to obtain the hierarchy of each drug. Duplication , the specific calculation formula is: , where C i and C i* represents the hierarchy of the i-th and i*-th drugs, δ(C i , C i* ) represents the hierarchical similarity value. If C i =C i* (indicated that the levels are exactly the same), then the level similarity value is 1. If C i =1 / 3C i* (indicated by the same level 1 / 3), the level similarity value is 0.

75. If C i =1 / 2C i* (indicated as level 1 / 2 are the same), the level similarity value is 0.5, otherwise, the level similarity value is 0.

6. The drug monitoring and management system based on clinical pharmacy according to claim 1 is characterized in that: The process of solving the matching evaluation value of drug combination is as follows: Extract indication matching value S Indication , target overlap value S Target and drug repeat value S Duplication The value of is normalized according to the formula: , the matching evaluation value SP of the drug combination is obtained, where e represents the set natural constant, μ1, μ2 and μ3 represent the set weight coefficients respectively, and μ1>μ2>μ3.

7. The drug monitoring and management system based on clinical pharmacy according to claim 1 is characterized in that: The process of solving the reasonable evaluation value of drug dosage is as follows: By extracting the dosage Y of each drug in the prescription Dose i , frequency Y Fre i 、Course Y Treat i And the patient characteristics influence coefficient λ for comprehensive analysis, according to the formula: , and obtain the reasonable evaluation value GP of drug dosage, where Q Dose i , Q Fre i and Q Treat i Represent the standard dose, standard frequency and standard course of treatment of the i-th drug, ΔY Dose i , ΔY Fre i and ΔY Treat i They represent the set allowable dose error value, allowable frequency error value and allowable treatment course error value of the i-th drug, μ4, μ5 and μ6 represent the set weight coefficients, and μ4>μ6>μ5; Among them, the patient characteristics influence coefficient is determined by the patient's height, weight, age, serum creatinine and treatment frequency in the prescription.

8. The drug monitoring and management system based on clinical pharmacy according to claim 1 is characterized in that: The process of solving the number of matching groups of material components at each risk level is as follows: By extracting the types of material components of each drug in the prescription and constructing a set, each drug in the prescription is combined in pairs to form a drug combination set; In each drug combination, the types of material components contained therein are then combined in pairs to form a set of material component type matching groups; Compare the set of material component type matching groups with the material component type risk table to determine the risk level of each material component type matching group. According to the risk level of each material component type matching group, each material component type matching group falls into the corresponding risk level. Count the number of material component type matching groups at each risk level to obtain the number of material component type matching groups at each risk level and mark it as W. zhing x , where x represents the risk level, x=1, 2, 3. When x=1, it indicates high-level risk, when x=2, it indicates medium-level risk, and when x=3, it indicates low-level risk.

9. The drug monitoring and management system based on clinical pharmacy according to claim 1, characterized in that: The process of solving the risk assessment value of drug interactions is as follows: According to the formula: , and obtain the risk assessment value HP of drug interaction, where W 总 It represents the total number of material component type matching groups of risk levels, μ7, μ8 and μ9 represent the set weight coefficients respectively, and μ7>μ8>μ9.

10. The drug monitoring and management system based on clinical pharmacy according to claim 1, characterized in that: The specific process of controlling and analyzing prescriptions is as follows: According to the captured prescription abnormality signal, the matching evaluation value of drug combination, the reasonable evaluation value of drug dosage and the risk evaluation value of drug interaction are retrieved in sequence, and compared and analyzed with the corresponding preset threshold value or reference comparison interval respectively. According to the comparison result, the corresponding abnormality signal is generated, and the matching deviation value of drug combination, the reasonable deviation value of drug dosage and the risk deviation value of drug interaction are calculated in sequence based on the abnormal signal; At the same time, each deviation value is matched and analyzed with the stored matching deviation status table, reasonable deviation status table and risk deviation status table to determine the corresponding matching deviation level, reasonable deviation level and risk deviation level, and based on each deviation level, the corresponding drug combination adjustment suggestion plan, drug dosage adjustment suggestion plan and drug interaction adjustment suggestion plan are generated, thereby integrating to form a prescription drug adjustment suggestion plan.

Images