Method and device for processing medicine document text data, equipment and medium

Abstract

The invention relates to the field of data processing, and discloses a processing method and device of medicine document text data, equipment and a medium. The method for processing the medicine document text data comprises the following steps: firstly, obtaining medicine document text data related to a to-be-evaluated medicine; then performing structured processing on the medicine literature text data to obtain structured background data, structured method data, structured result data and structured conclusion data; performing pattern recognition and context analysis on the structured result data to obtain key data related to economic evaluation indexes of the to-be-evaluated medicine; performing standardized conversion on the to-be-converted index data according to the research annual time data to obtain standard economic index data of the to-be-evaluated drug; and finally, performing consistency check on the standard economic index data by using the structured conclusion data to obtain an anomaly identification result of the standard economic index data. By automatically processing the medicine literature text data, the data processing efficiency is greatly improved.

Description

Technical Field

[0001] The present application relates to the field of data processing, and in particular to a method, device, equipment and medium for processing pharmaceutical literature text data. Background Art

[0002] In the process of new drug development and marketing, pharmaceutical companies need to comprehensively evaluate the economic and clinical value of drugs. In related technologies, the processing method of drug literature text data mainly adopts manual methods, such as manually retrieving relevant literature from professional databases and manually reading literature abstracts.

[0003] It can be seen that the processing process of drug literature text data in the related art relies heavily on manual work, resulting in low processing efficiency. Therefore, it is urgent to propose a new drug literature text data processing method. Summary of the invention

[0004] The present application provides a method, device, equipment and medium for processing drug literature text data, which solves the technical problem of low processing efficiency of drug literature text data in related technologies and achieves the technical effect of quickly and efficiently processing drug literature text data.

[0005] In order to achieve the above objectives, the main technical solutions adopted in this application include:

[0006] In a first aspect, an embodiment of the present application provides a method for processing pharmaceutical literature text data, the method comprising:

[0007] Obtaining drug literature text data related to the drug to be evaluated;

[0008] Structuring the drug literature text data to obtain structured background data, structured method data, structured result data and structured conclusion data;

[0009] Performing pattern recognition and context analysis on the structured result data to obtain key data related to the economic evaluation indicators of the drug to be evaluated; wherein the key data includes indicator data to be converted and research year time data; the indicator data to be converted is used to describe the economic value and clinical value of the drug to be evaluated;

[0010] Performing standardization conversion on the indicator data to be converted according to the research annual time data to obtain standard economic indicator data of the drug to be evaluated;

[0011] The structured conclusion data is used to perform a consistency check on the standard economic indicator data to obtain an abnormality identification result of the standard economic indicator data.

[0012] Optionally, the obtaining of drug literature text data related to the drug to be evaluated includes:

[0013] Obtaining abstract text data of drug literature related to the drug to be evaluated;

[0014] If the economic evaluation keywords of the drug to be evaluated are detected in the summary text data, the summary text data is determined as the drug literature text data; or

[0015] If the economic evaluation keywords of the drug to be evaluated are not detected in the abstract text data, the entire text data of the drug document is determined as the drug document text data.

[0016] Optionally, the drug literature text data is structured to obtain structured background data, structured method data, structured result data and structured conclusion data, including:

[0017] Inputting the drug literature text data into a large language model for recognition to obtain initial structured text data;

[0018] Based on the initial structured text data, paragraph recognition and structured extraction are performed to obtain the structured background data, the structured method data, the structured result data and the structured conclusion data in JSON format.

[0019] Optionally, the pattern recognition and context analysis of the structured result data are performed to obtain key data related to the economic evaluation indicators of the drug to be evaluated, including:

[0020] Recognize the numerical value and the unit in the structured result data to obtain a preliminary numerical value recognition result;

[0021] Based on the preliminary numerical recognition results, indicator-related modifiers are extracted and associated analyzed to obtain the key data.

[0022] Optionally, the indicator data to be converted corresponds to a target currency type; the standardization conversion of the indicator data to be converted according to the research year time data to obtain the standard economic indicator data of the drug to be evaluated includes:

[0023] Searching in a pre-built exchange rate database according to the research year time data and the target currency type to obtain target exchange rate data;

[0024] Using the target exchange rate data to convert the to-be-converted indicator data into currency units, to obtain intermediate economic indicator data;

[0025] The intermediate economic indicator data is standardized according to preset standard requirements to obtain the standard economic indicator data.

[0026] Optionally, the target currency type corresponding to the indicator data to be converted is determined in the following manner:

[0027] Matching the to-be-converted indicator data in pre-constructed currency type relationship data to obtain a currency type matching result;

[0028] The structured background data is used to perform semantic judgment on the currency type matching result to obtain the target currency type corresponding to the indicator data to be converted.

[0029] Optionally, matching the to-be-converted indicator data in pre-constructed currency type relationship data to obtain a currency type matching result includes:

[0030] According to the indicator data to be converted, exact matching and fuzzy matching are performed in the pre-constructed currency type relationship data to obtain the currency type matching result.

[0031] In a second aspect, an embodiment of the present application provides a device for processing pharmaceutical literature text data, the device comprising:

[0032] A literature text acquisition module is used to acquire drug literature text data related to the drug to be evaluated;

[0033] A text structured processing module, used for performing structured processing on the drug literature text data to obtain structured background data, structured method data, structured result data and structured conclusion data;

[0034] A key data extraction module, used for performing pattern recognition and context analysis on the structured result data to obtain key data related to the economic evaluation index of the drug to be evaluated; wherein the key data includes the index data to be converted and the research year time data; the index data to be converted is used to describe the economic value and clinical value of the drug to be evaluated;

[0035] A standardized conversion module, used for performing standardized conversion on the indicator data to be converted according to the research year time data, so as to obtain the standard economic indicator data of the drug to be evaluated;

[0036] The consistency check module is used to use the structured conclusion data to perform a consistency check on the standard economic indicator data to obtain an abnormality identification result of the standard economic indicator data.

[0037] In a third aspect, an embodiment of the present application provides a computer device, including:

[0038] A memory and a processor, wherein the memory and the processor are communicatively connected to each other, the memory stores computer instructions, and the processor executes the method described in any of the above embodiments by executing the computer instructions.

[0039] In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium, on which computer instructions are stored, and the computer instructions are used to enable a computer to execute the method described in any of the above embodiments.

[0040] The method for processing drug literature text data in this application first obtains drug literature text data related to the drug to be evaluated; then performs structured processing on the drug literature text data to obtain structured background data, structured method data, structured result data and structured conclusion data; then performs pattern recognition and context analysis on the structured result data to obtain key data related to the economic evaluation indicators of the drug to be evaluated; wherein the key data includes the indicator data to be converted and the research year time data; the indicator data to be converted is used to describe the economic value and clinical value of the drug to be evaluated; then the indicator data to be converted is standardized according to the research year time data to obtain the standard economic indicator data of the drug to be evaluated; finally, the standard economic indicator data is checked for consistency using the structured conclusion data to obtain the abnormal identification results of the standard economic indicator data. By automatically processing drug literature text data, the processing efficiency of drug literature text data is greatly improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] In order to more clearly illustrate the specific implementation methods of the present application or the technical solutions in the prior art, the drawings required for use in the specific implementation methods or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are some implementation methods of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0042] Figure 1 A flow chart of a method for processing pharmaceutical literature text data provided in the embodiments of this specification;

[0043] Figure 2 A flow chart of a method for processing pharmaceutical literature text data provided in the embodiments of this specification;

[0044] Figure 3 A flow chart of a method for processing pharmaceutical literature text data provided in the embodiments of this specification;

[0045] Figure 4 A flow chart of a method for processing pharmaceutical literature text data provided in the embodiments of this specification;

[0046] Figure 5 A flow chart of a method for processing pharmaceutical literature text data provided in the embodiments of this specification;

[0047] Figure 6 A flow chart of a method for processing pharmaceutical literature text data provided in the embodiments of this specification;

[0048] Figure 7 A schematic diagram of a device for processing pharmaceutical literature text data provided in an embodiment of this specification;

[0049] Figure 8 A schematic diagram of the structure of a computer device provided in an embodiment of this specification. DETAILED DESCRIPTION

[0050] In order to make the purpose, technical solution and advantages of the embodiments of the present application clearer, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of the present application.

[0051] In the process of new drug development and marketing, it is required to analyze a large number of pharmacoeconomic evaluation data in published drug literature. However, drug literature text data faces the problem of heterogeneity from multiple data sources. First, drug literature data are usually scattered in different databases and lack a unified standardized interface, which makes the retrieval and aggregation process cumbersome. Secondly, the representation of monetary units is not uniform, and different documents may use different currency symbols, which increases the complexity of data integration. The sources of exchange rate data are also very diverse, which may come from different countries or institutions, and different data sources may use different formats. Finally, the format of GDP data is also inconsistent. GDP data from different countries, regions and years may differ in expression, making unified processing more difficult.

[0052] On this basis, data standardization faces high difficulties, which are mainly reflected in the following aspects: First, different documents use different monetary units, which need to be converted and unified. Second, historical exchange rate data is difficult to obtain, which increases the difficulty of processing. Furthermore, GDP data needs to be matched according to different countries and years, and the format may also be different, which increases the complexity of adjustment. Finally, data quality control is complicated, and errors in any link may affect the accuracy of the final result.

[0053] In related technologies, the drug literature data processing process still relies on a lot of manual operations. Researchers need to read the abstracts of the literature one by one and manually extract relevant data (such as ICER, incremental QALY, etc.). This process is time-consuming and easy to miss important information. In addition, the currency unit conversion process involves the calculation of historical exchange rates and the comparison of data from multiple countries, which is cumbersome and error-prone. Another limitation of manual processing is that it is difficult to cope with large-scale data sets, especially when the amount of data is huge, the efficiency and accuracy of manual methods are limited.

[0054] Based on this, the present application provides a new method for processing drug literature text data, firstly obtaining drug literature text data related to the drug to be evaluated; then performing structured processing on the drug literature text data to obtain structured background data, structured method data, structured result data and structured conclusion data; then performing pattern recognition and context analysis on the structured result data to obtain key data related to the economic evaluation indicators of the drug to be evaluated; wherein the key data includes the indicator data to be converted and the research year time data; the indicator data to be converted is used to describe the economic value and clinical value of the drug to be evaluated; then the indicator data to be converted is standardized and converted according to the research year time data to obtain the standard economic indicator data of the drug to be evaluated; finally, the standard economic indicator data is checked for consistency using the structured conclusion data to obtain the abnormal identification result of the standard economic indicator data. By automatically processing drug literature text data, not only the efficiency of data processing is greatly improved, but also human errors and subjective biases can be effectively reduced, providing more accurate and reliable support for pharmacoeconomic evaluation.

[0055] The method provided in this application includes the following modules:

[0056] 1) Text structuring processing module: used to automatically identify and structure information in drug literature text data; extract key information such as study type and intervention measures; identify the study country / region and study year.

[0057] 2) Key indicator extraction module: used to automatically identify ICER values ​​and incremental QALY values; extract relevant monetary unit information; and establish a standardized data structure.

[0058] 3) Currency unit standardization module: used to establish a standardized mapping table for currency symbols; automatically identify and convert currency symbols in different formats; and handle special currency representation methods.

[0059] 4) Exchange rate data integration module: used to integrate historical exchange rate data from multiple sources; establish a unified exchange rate query interface; and handle missing exchange rate data.

[0060] 5) ICER standardization processing module: used to obtain GDP data of various countries over the years; automatically convert currency units; calculate the ratio of ICER to per capita GDP; and generate standardized analysis results.

[0061] 6) Data quality control module: used to automatically detect duplicate data; verify data integrity; and ensure data consistency.

[0062] According to an embodiment of the present application, an embodiment of a method for processing pharmaceutical literature text data is provided. It should be noted that the steps shown in the flowchart of the accompanying drawings can be executed in a computer system such as a set of computer executable instructions, and although a logical order is shown in the flowchart, in some cases, the steps shown or described can be executed in an order different from that shown here.

[0063] See also Figure 1 The method for processing drug literature text data provided in the embodiment of the present application comprises the following steps:

[0064] S110. Obtain drug literature text data related to the drug to be evaluated.

[0065] S120, structurally processing the drug literature text data to obtain structured background data, structured method data, structured result data and structured conclusion data.

[0066] S130. Perform pattern recognition and context analysis on the structured result data to obtain key data related to the economic evaluation indicators of the drug to be evaluated.

[0067] S140. Standardize and convert the indicator data to be converted according to the research annual time data to obtain standard economic indicator data of the drug to be evaluated.

[0068] S150. Use the structured conclusion data to perform consistency check on the standard economic indicator data to obtain anomaly identification results of the standard economic indicator data.

[0069] Among them, the drugs to be evaluated may be drugs that are undergoing economic evaluation, clinical effect analysis or other research. Usually, these drugs have not yet been widely promoted or formally approved, and need to be evaluated through a series of standardized economic indicators, clinical data, etc. to assess their cost-effectiveness, clinical effects or other related values. The analysis of the drugs to be evaluated helps decision makers understand the economic rationality of the drug in different scenarios and provides a basis for whether it should be widely used or promoted.

[0070] Drug literature text data can be literature text data on economic evaluation of drugs, clinical research results, drug efficacy, etc., and can come from databases (such as PubMed, Embase), journal articles, clinical trial reports and other resources.

[0071] Structural processing can be to convert the original, unstructured drug literature text data into structured data that is easy to analyze, such as converting drug literature text data into structured background data, structured method data, structured result data and structured conclusion data.

[0072] Pattern recognition can be used to identify pharmacoeconomic-related numerical information and unit information in structured result data, including numerical pattern recognition and unit recognition. Numerical pattern recognition can be used to identify segments containing numbers in structured result data, such as the extraction of data related to costs and benefits. Unit recognition can be used to identify units related to numerical values, such as monetary units (e.g., U.S. dollars, RMB), time units (e.g., years, months), etc.

[0073] Context analysis can be used to understand the semantics of modifiers related to pharmacoeconomic indicators (such as increase, decrease) and identify the relationship between indicators.

[0074] Key data may include indicator data to be converted and research year time data.

[0075] The indicator data to be converted can be the initial data recorded in the text data of drug literature, which is used to describe the economic value and clinical value of the drug to be evaluated. These data need to be processed through a standardized conversion process to obtain standard economic indicator data that meet unified standards. The indicator data to be converted usually includes information such as cost and benefit, and will then be adjusted based on relevant data such as exchange rates and GDP to ensure its comparability and accuracy in different times, regions or economic backgrounds.

[0076] The research year time data may be data such as the year and month in which the economic performance of a certain drug is studied as recorded in the drug literature text data.

[0077] The anomaly identification result can be the unexpected, abnormal or inconsistent parts of the data found after the consistency check of the standard economic indicator data through the structured conclusion data. These anomalies may appear as errors, deviations or unreasonable values ​​in the data, which need further review or correction to ensure the accuracy and reliability of the data.

[0078] Specifically, firstly, economic evaluation literature resources or clinical trial reports on the drug to be evaluated are obtained from professional databases (such as PubMed and Embase) as drug literature text data.

[0079] Secondly, the text structured processing module can be used to perform structured processing on the drug literature text data to obtain structured background data (such as research year, research country / region), structured method data (such as intervention measures), structured result data (such as incremental cost, incremental effect, incremental cost-effectiveness ratio, etc.) and structured conclusion data (such as summary of research findings, policy recommendations, etc.).

[0080] Furthermore, the key indicator extraction module is used to identify pharmacoeconomic-related numerical information and unit information in the structured result data, where the numerical information can be fragments containing numbers in the structured result data, such as data involving costs, benefits, etc., and the unit information can be units related to these numerical values, such as monetary units (such as US dollars, RMB) or time units (such as years, months), etc., to establish a standardized data structure.

[0081] Furthermore, the currency unit standardization module is used to standardize the indicator data to be converted according to the study year time data, and the exchange rate conversion is performed through the exchange rate data integration module. For example, the unit of the incremental cost is obtained in pounds / year, but the unit of the incremental cost-effectiveness ratio (ICER) used uniformly is US dollars / year, so the unit of the incremental cost needs to be converted into US dollars according to the exchange rate in the study year time data as the standard economic indicator data of the drug to be evaluated.

[0082] Finally, the data quality control module uses structured conclusion data to check the consistency of standard economic indicator data. For example, if the research conclusion is that a certain drug is economical, but the incremental QALY obtained by reasoning is negative, it indicates that there is an abnormality in data processing and further investigation is required.

[0083] In the above embodiment, firstly, the drug literature text data related to the drug to be evaluated is obtained; then the drug literature text data is structured to obtain structured background data, structured method data, structured result data and structured conclusion data; then the structured result data is pattern recognized and context analyzed to obtain key data related to the economic evaluation index of the drug to be evaluated; wherein the key data includes the index data to be converted and the research year time data; the index data to be converted is used to describe the economic value and clinical value of the drug to be evaluated; then the index data to be converted is standardized according to the research year time data to obtain the standard economic index data of the drug to be evaluated; finally, the standard economic index data is checked for consistency using the structured conclusion data to obtain the abnormal identification result of the standard economic index data. By automatically processing the drug literature text data, not only the efficiency of data processing is greatly improved, but also human errors and subjective biases can be effectively reduced, providing more accurate and reliable support for drug economic evaluation.

[0084] See also Figure 2 In some embodiments, obtaining drug literature text data related to the drug to be evaluated includes the following steps:

[0085] S210: Obtain summary text data of drug literature related to the drug to be evaluated.

[0086] S220: If the economic evaluation keywords of the drug to be evaluated are detected in the summary text data, the summary text data is determined as drug literature text data.

[0087] S230: If the economic evaluation keywords of the drug to be evaluated are not detected in the abstract text data, the entire text data of the drug literature is determined as the drug literature text data.

[0088] Among them, the keywords of economic evaluation can be the core terms and concepts used in evaluating the cost-effectiveness of drugs, including incremental cost-effectiveness ratio (ICER), incremental cost, incremental effect, quality-adjusted life year (QALY), incremental quality-adjusted life year, etc.

[0089] Specifically, after obtaining a drug document related to the drug to be evaluated, the abstract text data of the drug document can be first processed through the text structured processing module. If economic evaluation keywords related to the drug to be evaluated are detected in the abstract text data, such as incremental cost-effectiveness ratio (ICER) and incremental quality-adjusted life year (incremental QALY), the standard economic indicator data of the drug to be evaluated can be obtained from the abstract text data without processing the entire document. It is understandable that if the indicator data to be converted is obtained, it is also necessary to standardize the indicator data to be converted according to the research year time data.

[0090] If no economic evaluation keywords of the drug to be evaluated are detected in the abstract text data, the entire text data of the drug literature is determined as the drug literature text data, and then it is structured to obtain structured background data, structured method data, structured result data and structured conclusion data; then the structured result data is pattern recognized and context analyzed to obtain key data related to the economic evaluation indicators of the drug to be evaluated. Then, the indicator data to be converted is standardized according to the research year time data to obtain the standard economic indicator data of the drug to be evaluated. Finally, the standard economic indicator data is checked for consistency using the structured conclusion data to obtain the abnormal identification results of the standard economic indicator data.

[0091] See also Figure 3 In some embodiments, the drug literature text data is structured to obtain structured background data, structured method data, structured result data and structured conclusion data, and the steps include:

[0092] S310: Input the drug literature text data into the large language model for recognition to obtain initial structured text data.

[0093] S320, performing paragraph recognition and structured extraction based on the initial structured text data to obtain structured background data, structured method data, structured result data and structured conclusion data in JSON format.

[0094] Among them, the initial structured text data refers to the extraction of the basic structural elements in the literature after the preliminary processing and recognition of the drug literature text data, including the abstract, method, result, conclusion and other parts of the literature. At this stage, each part of the literature has been disassembled and marked as an independent structural unit, but the original information of the drug literature text data is still maintained.

[0095] Specifically, the drug literature text data can be input into the pre-trained Transformers large model, and the large language model can be used to perform structured analysis and recognition of the drug literature. The large language model uses contextual relationships to identify the various parts of the literature for primary classification and divides them into structures such as abstract, method, results and conclusion.

[0096] Then, the Transformer model is used to perform paragraph recognition on the initial structured text data based on pattern matching. The model uses deep semantic understanding to perform structured extraction of key information of the abstract, method, results, and conclusions, such as the research year and research country / region in the background section, the research subjects, intervention measures, and control group in the method section, the incremental cost, incremental effect, and incremental cost-effectiveness ratio in the results section, and the summary of research findings in the conclusion section as secondary classifications, and output in JSON format to obtain structured background data, structured method data, structured result data, and structured conclusion data. It should be noted that the data format must be unified and the field names must be standardized during the extraction process.

[0097] See also Figure 4 In some embodiments, pattern recognition and context analysis are performed on the structured result data to obtain key data related to the economic evaluation indicators of the drug to be evaluated, and the steps include:

[0098] S410 , identifying the numerical values ​​and units in the structured result data to obtain a preliminary numerical value recognition result.

[0099] S420. Based on the preliminary numerical recognition results, perform indicator-related modifier extraction and association analysis to obtain key data.

[0100] Among them, numerical pattern recognition can be to identify segments containing numbers in structured result data, such as the extraction of data related to costs, benefits, etc. Unit recognition can be to identify units related to numerical values, such as monetary units (such as US dollars, RMB), time units (such as years, months), etc.

[0101] Modifiers can be words that describe changes in data indicators, such as increase, decrease, or multiple.

[0102] Association analysis can be used to understand the semantics of data by analyzing the associations between data.

[0103] Specifically, the structured result data is first obtained through the text structured processing module, and then the text fragments containing numbers are identified through the key indicator extraction module, and the numerical values ​​and their units (such as monetary units, time units, etc.) are identified to obtain preliminary numerical recognition results. In some examples, the relevant modifiers of the preliminary numerical recognition results can also be extracted, such as increase, decrease, etc., and the key data can be inferred through the modifiers. In some other examples, the semantics of some numerical values ​​can also be determined by association analysis to obtain key data. For example, the structured result data records that the incremental cost of drug A is 2000 US dollars compared to drug B, and the incremental effect is 2 years, then it can be inferred that the incremental cost-effectiveness ratio (ICER) is 1000 US dollars / year. If the value of 1000 is also recorded in the structured result data, it can be known that this data refers to ICER. In addition, in the process of obtaining key data, it is also necessary to consider the standardization of key data, involving unified numerical format, identification of numerical intervals, extraction of confidence intervals, etc.

[0104] See also Figure 5 In some embodiments, the indicator data to be converted corresponds to the target currency type; the indicator data to be converted is standardized and converted according to the research year time data to obtain the standard economic indicator data of the drug to be evaluated, and the steps include:

[0105] S510: Search in a pre-built exchange rate database according to the research year time data and the target currency type to obtain target exchange rate data.

[0106] S520 , convert the currency unit of the indicator data to be converted using the target exchange rate data to obtain intermediate economic indicator data.

[0107] S530. Standardize the intermediate economic indicator data according to preset standard requirements to obtain standard economic indicator data.

[0108] The exchange rate database may be a database recording the exchange rates between various types of currencies in different years and months. The intermediate economic indicator data may be data obtained by converting the indicator data to be converted by performing exchange rate conversion. The preset standard requirements may be standardization requirements for key data set in the ICER standardization processing module, and the ratio of ICER to per capita GDP is usually used as the standardized ICER data.

[0109] Specifically, since different countries and regions have different monetary units, if cross-national or cross-regional economic comparisons are to be made, all data must be converted into the same monetary unit for analysis. This is an important prerequisite for ensuring the effectiveness and comparability of comparisons between economic indicators. Therefore, establishing a unified and accurate exchange rate database plays an indispensable role in processing exchange rate conversions and ensuring data consistency.

[0110] The steps for building an exchange rate database include: first, collecting exchange rate information from various data sources. Different data sources may use different exchange rate recording methods and data formats. For example, the U.S. dollar to Hong Kong dollar may be recorded as "June 2020: 1 U.S. dollar = 8.2 Hong Kong dollars" in one data set, and the U.S. dollar to the British pound may be recorded in another data set as "June 2020: 1 British pound = 1.2 U.S. dollars". Therefore, it is necessary to comprehensively collect multi-source exchange rate data to ensure that the exchange rate information of all relevant currency pairs is covered; secondly, identify the meaning of the exchange rate and unify the data format. The format and expression of exchange rate data are usually diverse, and it is necessary to collect exchange rate data based on the text information of the exchange rate file. To determine the meaning of the exchange rate and perform standardization for subsequent calculation and analysis; then, establish a time series index. Since the exchange rate changes over time, it is crucial to ensure the time accuracy of the exchange rate data. Each exchange rate data record should be clearly associated with its corresponding date to ensure that the exchange rate used in the conversion meets the time requirements; finally, output a unified format exchange rate database. All exchange rate information must be sorted and standardized, and then output into a unified format exchange rate database, which can support efficient conversion between different currency units and ensure that when processing economic evaluations, all indicators involving currency can be effectively compared under the same currency unit. After the exchange rate database is established, the exchange rate query interface of the exchange rate integration module can be used to query the exchange rate.

[0111] In some embodiments, assuming that the target currency type corresponding to the indicator data to be converted is British Pound, and the research year time data is April 2020, and the currency type needs to be converted into US Dollar, the exchange rate query interface of the exchange rate integration module can be used to first search the exchange rate between British Pound and US Dollar in April 2020 in the pre-built exchange rate database, and then convert the indicator data to be converted into the intermediate economic indicator data in US Dollar. Further, the ICER standardization processing module obtains the GDP data of the country in the research year time data according to the country name recorded in the research year time data and the drug literature text data, calculates the ratio of ICER in the intermediate economic indicator data to the per capita GDP data, and obtains the standard economic indicator data.

[0112] See also Figure 6 In some embodiments, the target currency type corresponding to the indicator data to be converted is determined by:

[0113] S610: Match the to-be-converted indicator data in the pre-constructed currency type relationship data to obtain a currency type matching result.

[0114] S620: Use the structured background data to perform semantic judgment on the currency type matching result to obtain the target currency type corresponding to the indicator data to be converted.

[0115] The currency type relationship data may be the correspondence between the currencies and currency symbols of various countries. It should be noted that the currencies and currency symbols of various countries are not necessarily one-to-one. For example, RMB can be represented by two symbols, ￥ or ￥, and these two symbols can also represent Japanese yen. The currency type matching result may be the result of matching the currency symbol in the indicator data to be converted in the currency type relationship data, which may be one currency type or multiple currency types. The target currency type may be the only currency type that the indicator data to be converted finally matches.

[0116] Specifically, if there is a currency symbol after the numerical value in the indicator data to be converted, it can be matched in the pre-constructed currency type relationship data first to obtain the currency type matching result. For example, if the indicator data to be converted is "1000￥", it is matched in the pre-constructed currency type relationship data to obtain the currency type matching results of RMB and Japanese Yen. When the currency type matching result is a variety of cases, the structured background data can be used to further judge the currency type matching result through semantics to obtain the target currency type corresponding to the indicator data to be converted. For example, the structured background data records "Study the economic feasibility of xx drug in China", then it can be judged that ￥ represents RMB.

[0117] In some embodiments, matching is performed in pre-constructed currency type relationship data according to the indicator data to be converted to obtain a currency type matching result, including: performing exact matching and fuzzy matching in pre-constructed currency type relationship data according to the indicator data to be converted to obtain a currency type matching result.

[0118] Specifically, in the indicator data to be converted, if there is a currency symbol after the value, firstly, an exact match can be performed in the pre-built currency type relationship data. If one or more currency types are matched, it is the currency type matching result. If no exact match is found, the currency type matching result can be obtained through fuzzy matching, such as fuzzy matching through currency symbols similar to the currency symbol, and then the target currency type can be further determined by combining the information recorded in the structured background data.

[0119] See also Figure 7 The embodiment of the present application further provides a device 700 for processing drug literature text data. The device 700 for processing drug literature text data includes:

[0120] The document text acquisition module 710 is used to acquire drug document text data related to the drug to be evaluated;

[0121] The text structure processing module 730 is used to perform structure processing on the drug literature text data to obtain structured background data, structured method data, structured result data and structured conclusion data;

[0122] The key data extraction module 750 is used to perform pattern recognition and context analysis on the structured result data to obtain key data related to the economic evaluation indicators of the drug to be evaluated; wherein the key data includes the indicator data to be converted and the research year time data; the indicator data to be converted is used to describe the economic value and clinical value of the drug to be evaluated;

[0123] A standardized conversion module 770 is used to perform standardized conversion on the indicator data to be converted according to the research year time data to obtain the standard economic indicator data of the drug to be evaluated;

[0124] The consistency check module 790 is used to perform consistency check on the standard economic indicator data using the structured conclusion data to obtain anomaly identification results of the standard economic indicator data.

[0125] In some embodiments, the document text acquisition module 710 is also used to acquire summary text data of drug literature related to the drug to be evaluated; if economic evaluation keywords of the drug to be evaluated are detected in the summary text data, the summary text data is determined as drug literature text data; or if economic evaluation keywords of the drug to be evaluated are not detected in the summary text data, the entire text data of the drug literature is determined as drug literature text data.

[0126] In some implementations, the text structuring processing module 730 further includes:

[0127] A data recognition unit is used to input the drug literature text data into the large language model for recognition to obtain initial structured text data;

[0128] The recognition and extraction unit is used to perform paragraph recognition and structured extraction based on the initial structured text data to obtain structured background data, structured method data, structured result data and structured conclusion data in JSON format.

[0129] In some implementations, the key data extraction module 750 further includes:

[0130] The recognition unit is used to recognize the numerical value and unit in the structured result data to obtain a preliminary numerical recognition result;

[0131] The key data acquisition unit is used to extract and analyze the index-related modifiers based on the preliminary numerical recognition results to obtain the key data.

[0132] In some embodiments, the indicator data to be converted corresponds to the target currency type; the indicator data to be converted is standardized and converted according to the research year time data to obtain the standard economic indicator data of the drug to be evaluated, and the key data extraction module 750 includes:

[0133] An exchange rate search unit is used to search in a pre-built exchange rate database according to the research year time data and the target currency type to obtain the target exchange rate data;

[0134] A currency conversion unit, used for converting the currency unit of the indicator data to be converted by using the target exchange rate data to obtain the intermediate economic indicator data;

[0135] The standardization unit is used to standardize the intermediate economic indicator data according to the preset standard requirements to obtain standard economic indicator data.

[0136] In some implementations, the key data extraction module 750 further includes:

[0137] A currency type matching unit, used to match the to-be-converted indicator data in the pre-built currency type relationship data to obtain a currency type matching result;

[0138] The semantic judgment unit is used to perform semantic judgment on the currency type matching result by using the structured background data to obtain the target currency type corresponding to the indicator data to be converted.

[0139] In some implementations, the key data extraction module 750 further includes:

[0140] The currency type matching unit is used to perform precise matching and fuzzy matching in pre-built currency type relationship data according to the indicator data to be converted, and obtain a currency type matching result.

[0141] The further functional description of each of the above modules and units is the same as that of the above corresponding embodiments and will not be repeated here.

[0142] The processing device for drug literature text data in this embodiment is presented in the form of a functional unit, where the unit refers to an ASIC (Application Specific Integrated Circuit) circuit, a processor and memory that executes one or more software or fixed programs, and / or other devices that can provide the above functions.

[0143] See also Figure 8 , Figure 8 is a schematic diagram of the structure of a computer device provided in an embodiment of the present application, such as Figure 8 As shown, the computer device includes: one or more processors 10, a memory 20, and interfaces for connecting various components, including high-speed interfaces and low-speed interfaces. Various components are connected to each other using different buses for communication, and can be installed on a common mainboard or installed in other ways as needed. The processor can process the instructions executed in the computer device, including instructions stored in or on the memory to display the graphical information of the GUI on an external input / output device (such as, a display device coupled to the interface). In some optional embodiments, if necessary, multiple processors and / or multiple buses can be used together with multiple memories and multiple memories. Similarly, multiple computer devices can be connected, and each device provides some necessary operations (for example, as a server array, a group of blade servers, or a multi-processor system). Figure 8 A processor 10 is taken as an example.

[0144] The processor 10 may be a central processing unit, a network processor or a combination thereof. The processor 10 may further include a hardware chip. The hardware chip may be a dedicated integrated circuit, a programmable logic device or a combination thereof. The programmable logic device may be a complex programmable logic device, a field programmable gate array, a general purpose array logic or any combination thereof.

[0145] The memory 20 stores instructions executable by at least one processor 10, so that the at least one processor 10 executes the method shown in the above embodiment.

[0146] The memory 20 may include a program storage area and a data storage area, wherein the program storage area may store an operating system, an application required for at least one function; the data storage area may store data created according to the use of the computer device, etc. In addition, the memory 20 may include a high-speed random access memory, and may also include a non-transient memory, such as at least one disk storage device, a flash memory device, or other non-transient solid-state storage device. In some optional embodiments, the memory 20 may optionally include a memory remotely arranged relative to the processor 10, and these remote memories may be connected to the computer device via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0147] The memory 20 may include a volatile memory, such as a random access memory; the memory may also include a non-volatile memory, such as a flash memory, a hard disk or a solid state drive; the memory 20 may also include a combination of the above types of memory.

[0148] The computer device also includes an input device 30 and an output device 40. The processor 10, the memory 20, the input device 30 and the output device 40 may be connected via a bus or other means. Figure 8 The example of connecting through bus is taken in the following.

[0149] The input device 30 can receive input digital or character information, and generate key signal input related to the user settings and function control of the computer device, such as a touch screen, a keypad, a mouse, a track pad, a touch pad, an indicator bar, one or more mouse buttons, a trackball, a joystick, etc. The output device 40 may include a display device, an auxiliary lighting device (e.g., an LED) and a tactile feedback device (e.g., a vibration motor), etc. The above-mentioned display device includes but is not limited to a liquid crystal display, a light emitting diode, a display and a plasma display. In some optional embodiments, the display device can be a touch screen.

[0150] The embodiment of the present application also provides a computer-readable storage medium. The above method according to the embodiment of the present application can be implemented in hardware, firmware, or can be implemented as a computer code that can be recorded in a storage medium, or can be implemented as a computer code that is originally stored in a remote storage medium or a non-temporary machine-readable storage medium and will be stored in a local storage medium through a network download, so that the method described herein can be stored in such software processing on a storage medium using a general-purpose computer, a dedicated processor, or programmable or dedicated hardware. Among them, the storage medium can be a magnetic disk, an optical disk, a read-only storage memory, a random access memory, a flash memory, a hard disk or a solid-state hard disk, etc.; further, the storage medium can also include a combination of the above types of memories. It can be understood that a computer, a processor, a microprocessor controller, or programmable hardware includes a storage component that can store or receive software or computer code. When the software or computer code is accessed and executed by a computer, a processor, or hardware, the method shown in the above embodiment is implemented.

[0151] The embodiment of the present application provides a computer program product, which includes computer instructions, which are stored in a computer-readable storage medium. A processor of a computer device reads the computer instructions from the computer-readable storage medium, and the processor executes the computer instructions, so that the computer device executes the method of any embodiment of the present application.

[0152] Although the embodiments of the present application are described in conjunction with the accompanying drawings, those skilled in the art may make various modifications and variations without departing from the spirit and scope of the present application, and such modifications and variations are all within the scope defined by the appended claims.

[0153] The systems, devices, modules or units described in the above embodiments may be implemented by computer chips or entities, or by products with certain functions. A typical implementation device is a computer. Specifically, the computer may be, for example, a personal computer, a laptop computer, a cellular phone, a camera phone, a smart phone, a personal digital assistant, a media player, a navigation device, an email device, a game console, a tablet computer, a wearable device, or a combination of any of these devices.

[0154] For the convenience of description, the above device is described in various units according to their functions. Of course, when implementing the present application, the functions of each unit can be implemented in the same or multiple software and / or hardware.

[0155] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems, or computer program products. Therefore, the present application may adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment in combination with software and hardware. Moreover, the present application may adopt the form of a computer program product implemented in one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) that include computer-usable program code.

[0156] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0157] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0158] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0159] It should also be noted that the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, commodity or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, commodity or device. In the absence of more restrictions, the elements defined by the sentence "comprises a ..." do not exclude the existence of other identical elements in the process, method, commodity or device including the elements.

[0160] Each embodiment in this specification is described in a progressive manner, and the same or similar parts between the embodiments can be referred to each other, and each embodiment focuses on the differences from other embodiments. In particular, for the system embodiment, since it is basically similar to the method embodiment, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiment.

[0161] The above is only an embodiment of the present application and is not intended to limit the present application. For those skilled in the art, the present application may have various changes and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application should be included in the scope of the claims of the present application.

[0162] Although the embodiments of the present application have been described in conjunction with the accompanying drawings, those skilled in the art may make various modifications and variations without departing from the spirit and scope of the present application, and such modifications and variations are all within the scope defined by the appended claims.

Claims

1. A method for processing drug literature text data, characterized in that: The method comprises: Obtaining drug literature text data related to the drug to be evaluated; Structuring the drug literature text data to obtain structured background data, structured method data, structured result data and structured conclusion data; Performing pattern recognition and context analysis on the structured result data to obtain key data related to the economic evaluation indicators of the drug to be evaluated; wherein the key data includes indicator data to be converted and research year time data; the indicator data to be converted is used to describe the economic value and clinical value of the drug to be evaluated; Performing standardization conversion on the indicator data to be converted according to the research annual time data to obtain standard economic indicator data of the drug to be evaluated; The structured conclusion data is used to perform a consistency check on the standard economic indicator data to obtain an abnormality identification result of the standard economic indicator data.

2. The method according to claim 1, characterized in that The step of obtaining drug literature text data related to the drug to be evaluated includes: Obtaining abstract text data of drug literature related to the drug to be evaluated; If the economic evaluation keywords of the drug to be evaluated are detected in the summary text data, the summary text data is determined as the drug literature text data; or If the economic evaluation keywords of the drug to be evaluated are not detected in the abstract text data, the entire text data of the drug document is determined as the drug document text data.

3. The method according to claim 1, characterized in that The structural processing of the drug literature text data to obtain structured background data, structured method data, structured result data and structured conclusion data includes: Inputting the drug literature text data into a large language model for recognition to obtain initial structured text data; Based on the initial structured text data, paragraph recognition and structured extraction are performed to obtain the structured background data, the structured method data, the structured result data and the structured conclusion data in JSON format.

4. The method according to claim 1, characterized in that The pattern recognition and context analysis of the structured result data are performed to obtain key data related to the economic evaluation indicators of the drug to be evaluated, including: Recognize the numerical value and the unit in the structured result data to obtain a preliminary numerical value recognition result; Based on the preliminary numerical recognition results, indicator-related modifiers are extracted and associated analyzed to obtain the key data.

5. The method according to claim 1, characterized in that The indicator data to be converted corresponds to the target currency type; The method of performing standardization conversion on the indicator data to be converted according to the research year time data to obtain the standard economic indicator data of the drug to be evaluated includes: Searching in a pre-built exchange rate database according to the research year time data and the target currency type to obtain target exchange rate data; Using the target exchange rate data to convert the to-be-converted indicator data into currency units, to obtain intermediate economic indicator data; The intermediate economic indicator data is standardized according to preset standard requirements to obtain the standard economic indicator data.

6. The method according to claim 5, characterized in that The target currency type corresponding to the indicator data to be converted is determined by: Matching the to-be-converted indicator data in pre-constructed currency type relationship data to obtain a currency type matching result; The structured background data is used to perform semantic judgment on the currency type matching result to obtain the target currency type corresponding to the indicator data to be converted.

7. The method according to claim 6, characterized in that The matching is performed in the pre-constructed currency type relationship data according to the indicator data to be converted to obtain a currency type matching result, including: According to the indicator data to be converted, exact matching and fuzzy matching are performed in the pre-constructed currency type relationship data to obtain the currency type matching result.

8. A device for processing drug literature text data, characterized in that: The device comprises: A literature text acquisition module is used to acquire drug literature text data related to the drug to be evaluated; A text structured processing module, used for performing structured processing on the drug literature text data to obtain structured background data, structured method data, structured result data and structured conclusion data; A key data extraction module, used for performing pattern recognition and context analysis on the structured result data to obtain key data related to the economic evaluation index of the drug to be evaluated; wherein the key data includes the index data to be converted and the research year time data; the index data to be converted is used to describe the economic value and clinical value of the drug to be evaluated; A standardized conversion module, used for performing standardized conversion on the indicator data to be converted according to the research year time data, so as to obtain the standard economic indicator data of the drug to be evaluated; The consistency check module is used to use the structured conclusion data to perform a consistency check on the standard economic indicator data to obtain an abnormality identification result of the standard economic indicator data.

9. A computer device, characterized in that: include: A memory and a processor, wherein the memory and the processor are communicatively connected to each other, the memory stores computer instructions, and the processor executes the method according to any one of claims 1 to 7 by executing the computer instructions.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a computer to execute the method according to any one of claims 1 to 7.

Images