KR20230020608A - Apparatus and method for building a pipelines to explore adverse drug reaction - Google Patents

Abstract

The present invention relates to a method and an apparatus for building a pipeline to explore the side effects of a drug. The method for building a pipeline to explore the side effects of a drug according to one embodiment of the present invention comprises the steps of: accessing a social channel based on social network services (SNS) to collect at least one piece of social data related to a target drug, based on a first term set; pre-processing the collected social data; extracting data related to side effects from the preprocessed social data, based on the second term set; analyzing drug side effect patterns for the target drug according to analysis results to classify the same according to preset categories; and building or learning a drug side effect detection prediction model by using the classification results, wherein the first term set may include data sets consisting of at least one term representing at least one drug and the second term set may include a data set consisting of terms representing drug side effects. Accordingly, the side effects of the target drug can be more accurately predicted.

Description

Pipeline construction method and apparatus for drug side effect detection {APPARATUS AND METHOD FOR BUILDING A PIPELINES TO EXPLORE ADVERSE DRUG REACTION}

The present invention relates to a method and apparatus for constructing a pipeline for detecting side effects of a drug, and more particularly, to a method and apparatus for constructing a pipeline for detecting side effects of a drug that enables search and update of side effects of a drug through social data analysis. It is about.

In recent years, the frequency and quantity of drug use are rapidly increasing due to the aging of the world. As a result, drug side effects are also rapidly increasing, threatening the safety of patients. Accordingly, it is emerging as an important issue to analyze whether there are new side effects or serious side effects of drugs on the market for early detection.

In addition to publicly known side effects of the target drug, new side effects may occur, and side effects may additionally occur due to long-term administration.

However, it is cumbersome and inconvenient for a drug taker who takes the target drug in daily life to collect or monitor the symptoms expressed by taking the drug by receiving the data of the drug and the drug intake review, so there is a limit to obtaining actual information about the drug. there is

Meanwhile, as the use of social networking is rapidly increasing in the recent medical industry, many people share their feelings and experiences in social network services (SNS). As one of the acts of sharing valuable information, they respond by posting a review of a particular drug they have taken (or are taking) as a post or by writing comments.

As such, information posted or registered through social network services corresponds to the individual's direct testimonials, so it will be meaningful as it includes data related to new side effects or indications in addition to the side effects already known about the specific drug.

Therefore, it is necessary to develop a technology that can more accurately predict drug side effects caused by a target drug by analyzing patterns of drug side effects by collecting social data related to a target drug based on a social network service.

Korean Patent Publication No. 10-2015-0049937 (published on May 08, 2015)

The present invention has been proposed to solve the above problems, and collects social data related to a target drug based on a social network service and analyzes the pattern of drug side effects to more accurately predict drug side effects caused by the target drug. An object of the present invention is to provide a pipeline construction method and apparatus for detecting side effects of a drug.

The problems to be solved by the present invention are not limited to the problems mentioned above, and other problems not mentioned will be clearly understood by those skilled in the art from the description below.

In order to solve the above problems, a pipeline construction method for drug side effect detection according to an embodiment of the present invention accesses a social channel based on Social Network Services (SNS) to obtain a first term set. collecting at least one piece of social data related to a target drug based on; pre-processing the collected social data; performing exploratory data analysis by extracting data related to side effects from the preprocessed social data based on a second term set; Analyzing a drug side effect pattern for the target drug according to the analysis result and classifying it according to a preset category; and constructing or learning a drug side effect detection prediction model using the classification result, wherein the first term set includes data sets composed of at least one term representing each of at least one drug, and wherein the second A term set may include a data set composed of terms representing adverse drug reactions.

On the other hand, an apparatus for constructing a pipeline for detecting side effects of drugs according to an embodiment of the present invention includes a communication module; a storage module for storing at least one piece of information or data for constructing a pipeline for detecting the side effect of the drug; At least one social data related to a target drug is collected and pre-processed based on a first term set by accessing a social channel based on Social Network Services (SNS), and pre-processed based on a second term set. an analysis module for performing exploratory data analysis by extracting data related to side effects from social data, analyzing patterns of drug side effects for the target drug according to the analysis results, and classifying them according to preset categories; a learning module for constructing or learning a drug side effect detection prediction model using the classification result; and collecting and pre-processing at least one social data related to a target drug based on the first term set, and performing exploratory data analysis by extracting data related to side effects from the pre-processed social data based on the second term set. and a control module for controlling to construct or learn a drug side effect detection prediction model using the classification result after analyzing the drug side effect pattern for the target drug according to the analysis result and classifying it according to a preset category, , The first term set may include data sets composed of at least one term representing each of at least one drug, and the second term set may include a data set composed of terms representing drug side effects.

Other specific details of the invention are included in the detailed description and drawings.

According to the present invention, by collecting social data related to a target drug based on a social network service and analyzing a pattern of drug side effects, it is possible to more accurately predict drug side effects caused by the target drug.

The effects of the present invention are not limited to the effects mentioned above, and other effects not mentioned will be clearly understood by those skilled in the art from the description below.

1 is a block diagram showing the configuration of a pipeline construction apparatus for detecting side effects of drugs according to an embodiment of the present invention.
2 is a flowchart illustrating a method for constructing a pipeline for detecting side effects of drugs according to an embodiment of the present invention.
3 is a diagram schematically illustrating a pipeline for detecting side effects of drugs according to an embodiment of the present invention.
4 is a diagram illustrating a procedure for generating a second term set used for constructing a pipeline for detecting side effects of drugs according to an embodiment of the present invention.
5A is a diagram showing a first embodiment of analyzing a drug side effect pattern when constructing a pipeline for detecting side effect drug according to an embodiment of the present invention.
5B is a diagram illustrating a second embodiment of analyzing a drug side effect pattern when constructing a pipeline for detecting side effect drug according to an embodiment of the present invention.
5C is a diagram showing a third embodiment of analyzing a drug side effect pattern when constructing a pipeline for detecting side effect drug according to an embodiment of the present invention.

Advantages and features of the present invention, and methods of achieving them, will become clear with reference to the detailed description of the following embodiments taken in conjunction with the accompanying drawings. However, the present invention is not limited to the embodiments disclosed below, but may be implemented in various different forms, only these embodiments are intended to complete the disclosure of the present invention, and are common in the art to which the present invention belongs. It is provided to fully inform the person skilled in the art of the scope of the invention, and the invention is only defined by the scope of the claims.

Terminology used herein is for describing the embodiments and is not intended to limit the present invention. In this specification, singular forms also include plural forms unless specifically stated otherwise in a phrase. As used herein, "comprises" and/or "comprising" does not exclude the presence or addition of one or more other elements other than the recited elements. Like reference numerals throughout the specification refer to like elements, and “and/or” includes each and every combination of one or more of the recited elements. Although "first", "second", etc. are used to describe various components, these components are not limited by these terms, of course. These terms are only used to distinguish one component from another. Accordingly, it goes without saying that the first element mentioned below may also be the second element within the technical spirit of the present invention.

Unless otherwise defined, all terms (including technical and scientific terms) used in this specification may be used with meanings commonly understood by those skilled in the art to which the present invention belongs. In addition, terms defined in commonly used dictionaries are not interpreted ideally or excessively unless explicitly specifically defined.

The spatially relative terms "below", "beneath", "lower", "above", "upper", etc. It can be used to easily describe a component's correlation with other components. Spatially relative terms should be understood as including different orientations of elements in use or operation in addition to the orientations shown in the drawings. For example, if you flip a component that is shown in a drawing, a component described as "below" or "beneath" another component will be placed "above" the other component. can Thus, the exemplary term “below” may include directions of both below and above. Components may also be oriented in other orientations, and thus spatially relative terms may be interpreted according to orientation.

The term "unit" or "module" used in the specification means a hardware component such as software, FPGA or ASIC, and "unit" or "module" performs certain roles. However, "unit" or "module" is not meant to be limited to software or hardware. A "unit" or "module" may be configured to reside in an addressable storage medium and may be configured to reproduce one or more processors. Thus, as an example, a “unit” or “module” may refer to components such as software components, object-oriented software components, class components, and task components, processes, functions, properties, procedures, subroutines, segments of program code, drivers, firmware, microcode, circuitry, data, databases, data structures, tables, arrays and variables. Functions provided within components and "units" or "modules" may be combined into fewer components and "units" or "modules" or may be combined into additional components and "units" or "modules". can be further separated.

Unless otherwise defined, all terms (including technical and scientific terms) used in this specification may be used with meanings commonly understood by those skilled in the art to which the present invention belongs. In addition, terms defined in commonly used dictionaries are not interpreted ideally or excessively unless explicitly specifically defined.

Hereinafter, embodiments of the present invention will be described in detail with reference to the accompanying drawings.

1 is a block diagram showing the configuration of a pipeline construction apparatus for detecting side effects of drugs according to an embodiment of the present invention.

Referring to FIG. 1, an apparatus for constructing a pipeline for predicting adverse drug reactions (hereinafter referred to as 'construction apparatus') 100 includes a communication module 110, a storage module 130, an analysis module 150, and a learning module 170 ) and a control module 190.

The communication module 110 transmits and receives various types of information or data necessary for constructing a pipeline for detecting side effects of drugs with an external device. Specifically, social data may be collected by accessing (accessing) a social network service (portal site, messenger service, etc.) through the communication module 110 . That is, the communication module 110 is for performing communication with other terminals, servers, devices, etc., and transmits and receives wireless signals in a communication network according to wireless Internet technologies.

Wireless Internet technologies include, for example, WLAN (Wireless LAN), Wi-Fi (Wireless-Fidelity), Wi-Fi (Wireless Fidelity) Direct, DLNA (Digital Living Network Alliance), WiBro (Wireless Broadband), WiMAX (World Interoperability for Microwave Access), HSDPA (High Speed Downlink Packet Access), HSUPA (High Speed Uplink Packet Access), LTE (Long Term Evolution), LTE-A (Long Term Evolution-Advanced), etc. ) transmits and receives data according to at least one wireless Internet technology within a range including Internet technologies not listed above.

As for short range communication, Bluetooth™ RFID (Radio Frequency Identification), infrared communication (Infrared Data Association; IrDA), UWB (Ultra Wideband), ZigBee, NFC (Near Field Communication), Wi-Fi (Wireless-Fidelity), Wi-Fi Direct, and wireless USB (Wireless Universal Serial Bus) technology, and support for short-range communication using at least one of these, the apparatus 100 for establishing such a wireless local area network (Wireless Area Networks) Wireless communication between the device and an external device may be supported In this case, the local area wireless communication network may be a local area wireless personal area network.

The storage module 130 stores at least one term set as drug-related information and stores at least one process necessary for constructing a pipeline for detecting side effects of a drug.

The analysis module 150 collects social data on a target drug based on a social network service, constructs at least one term set, and analyzes and classifies a drug side effect pattern for the target drug based on this. Here, the social data may be a portal site based on a social network service, a messenger service, and the like.

To this end, the analysis module 150 may include a collection unit 151, a pre-processing unit 153, a term generation unit 155, a pattern analysis unit 157, and a classification unit 159.

Specifically, when the collecting unit 151 accesses a social network, collects at least one social data related to a target drug, and the preprocessing unit 153 preprocesses (refines) the collected at least one social data, a term is generated. A section 155 builds at least one term set for the target drug. Thereafter, the pattern analysis unit 157 analyzes the side effect pattern by extracting side effect-related data from the collected social data based on the at least one constructed term set, and the classification unit 159 analyzes the side effect pattern in advance based on the analysis result. Classify side effect information according to the set category. At this time, the preset category may be divided into known side effects, unknown side effects, and indications, but is not limited thereto, and may be changed or other categories may be added.

Here, the at least one term set is a first term set that is a data set for terms representing (referring to) the target drug (eg, ingredient, product name, brand, alias, abbreviation, etc.) and is standardized related to drug side effects. A second term set, which is a data set for terms representing drug side effects related to a target drug, constructed based on an open database, may be included. In this case, the first term set and the second term set may separately include data sets for each of a plurality of drugs. However, when a pipeline is first constructed, it is input by a manager or a worker, the pipeline is driven, and the first term set and the second term set may be respectively updated.

Meanwhile, when the analysis module 150 accesses a social network and collects at least one piece of social data related to a target drug, it may use the pre-constructed first term set. That is, the analysis module 150 collects social data by using keywords included in the data set for the target drug in the first term set.

The learning module 170 performs drug side effect labeling on the data classified by the analysis module 150, and then builds and/or learns a drug side effect detection prediction model by using the labeled data as training data.

At this time, the learning module 170 may build a drug side effect detection prediction model based on a recurrent neural network (RNN) learning method. As a structure constituting a recurrent neural network (RNN), a long short term memory network ) method can be used.

On the other hand, the learning module 170 uses a drug side effect detection prediction model based on data classified as known side effects among the classified data for learning, and when the data classified as unknown side effects is accumulated more than a preset number of times, it is known. It can be reclassified as a side effect and used to learn a drug side effect detection prediction model.

The control module 190 accesses a social network, collects at least one social data related to a target drug, and pre-processes (refines) the collected at least one social data, and the term generator 155 determines the target drug. Control to build at least one term set for Thereafter, the control module 190 extracts side effect-related data from the collected social data based on at least one constructed term set, analyzes the side effect pattern, and based on the analysis result, provides side effect information according to preset categories. control to classify.

2 is a flowchart illustrating a method for constructing a pipeline for detecting side effects of drugs according to an embodiment of the present invention.

Referring to FIG. 2 , in order to build a pipeline for detecting adverse drug reactions, the analysis module 150 accesses a social network and uses a data set related to a target drug in the first term set to generate at least one social network related to the target drug. Data is collected and preprocessed (S201), and side effect patterns are analyzed by extracting side effect related data from the preprocessed social data based on the second term set (S203).

Thereafter, labeling of drug side effects is performed based on the data extracted according to the analysis result (S205). At this time, the labeling may be performed automatically by the analysis module 150 or manually by a manager or operator. can

Thereafter, the learning module 170 uses the labeled data as learning data to learn the drug side effect prediction module (S207).

3 is a diagram schematically illustrating a pipeline for detecting side effects of drugs according to an embodiment of the present invention. This pipeline enables continuous or periodic prediction (detection) of side effects for a target drug and updates of the results. Here, the first term set and the second term set are described on the assumption that pre-constructed (generated) ones are used.

Referring to FIG. 3 , the construction device 100 accesses a social channel based on a social network service (S301), and generates at least one social data for a target drug based on a first term set for the target drug. Collect (S303).

At least one social data collected in this way is unstructured and unexpected unstructured data, and must be pre-processed (refined) in order to use it. At this time, unstructured data can be refined through natural language processing or text mining (S305). .

Here, text mining is a mining process for unstructured data, in which concepts or characteristics having statistical significance are extracted from data, and high-quality information such as patterns or trends among them is extracted.

On the other hand, at least one preprocessed social data is analyzed based on exploratory data analysis based on the second term set (S307), and a drug side effect pattern is analyzed using the analysis result (S309). Here, the exploratory data analysis may be to analyze valid data from big data according to user settings, and reduce and visualize the analyzed valid data according to preset setting options.

Here, exploratory data analysis may be performed based on correlation analysis or word embedding (word2vec) model. Specifically, correlation analysis is an analysis that finds interrelationships or dependent relationships between data existing within data, and enables simple but clear interpretation of results. Through this, it can be usefully used to discover unexpected knowledge from data by creating easy-to-understand rules for data. In addition, word embedding is an inference-based technique that vectorizes the meaning of words in order to consider similarities between words (keywords), and quantifies text to discover knowledge from a new perspective. To this end, a vector containing the maximum meaning of a word is created, and the similarity or relevance of a word pair is examined and interpreted.

According to the analysis result in step S309, the drug side effect pattern is classified into one category, for example, it can be classified into any one of known side effects, unknown side effects, and indications for the target drug.

Thereafter, labeling is performed based on data classified as known side effects among the classified data, and a drug side effect detection prediction model is learned using the labeled data as training data.

The trained drug side effect detection prediction model is used to detect side effects of social data collected later, and as the performance improves each time social data is collected and analyzed, more accurate prediction and detection becomes possible.

4 is a diagram illustrating a procedure for generating a second term set used for constructing a pipeline for detecting side effects of drugs according to an embodiment of the present invention.

Referring to FIG. 4 , the second term set may be generated using a list of adverse drug reactions generated based on WHO-ART and SIDER, which are standardized and open databases, and a pre-generated consumer dictionary. Here, WHO-ART represents an international classification system for terminology of adverse drug reactions, and is already most widely used in reporting ADR (Adverse Drug Reaction) in Korea. In addition, SIDER is a drug-side-effects database that provides a Drug-ADR relationship, and includes information on drug side effects on commercially available drugs and the drugs. Also, the consumer term dictionary may be defined by a manager or an operator. That is, the second term set is generated using standardized sets such as WHO-ART and SIDER.

Specifically, after generating a drug side effect list based on WHO-ART (S401), supplementing the drug side effect list based on SIDER (S403), and generating a drug side effect list based on consumer terms (S405), these By mutual mapping (S407), a second term set is generated as a result (S409).

However, the order of performing steps S401 to S405 is not defined, and may be performed independently (individually), and updates may be individually performed automatically or manually as needed.

However, WHO-ART and SIDER are only examples of standardized and open databases, and are not limited thereto, and other drug systems and drug data may be used.

Meanwhile, various methods may be applied to exploratory data analysis for analyzing drug side effect patterns from at least one piece of social data collected by the construction device 100 based on a social network service. Examples of each method will be described using FIGS. 5A to 5C.

5A is a diagram showing a first embodiment of analyzing a drug side effect pattern when constructing a pipeline for detecting side effect drug according to an embodiment of the present invention.

According to FIG. 5A, the analysis module 150 of the construction device 100 extracts words (keywords) appearing in the collected at least one piece of social data, analyzes the frequency of each of the extracted words (a ) to generate a frequency table as shown in

In addition, by using the terms to create a word cloud as shown in (b), it is possible to visually check which words and how frequently they are used.

5B is a diagram illustrating a second embodiment of analyzing a drug side effect pattern when constructing a pipeline for detecting side effect drug according to an embodiment of the present invention.

According to FIG. 5B, the analysis module 150 of the construction device 100 extracts words (keywords) appearing in the collected at least one piece of social data, and generates a table of top n related words among the extracted words. and create a visualization graph to visually confirm by analyzing the association between the words.

5C is a diagram showing a third embodiment of analyzing a drug side effect pattern when constructing a pipeline for detecting side effect drug according to an embodiment of the present invention.

According to FIG. 5C, the analysis module 150 of the construction device 100 extracts words (keywords) appearing in at least one social data collected from the word embedding, and matches the extracted words with a side effect dictionary after embedding them. Thus, it is possible to identify the pattern of side effects by organ system (SOC) of the human body, and to identify the top words in the order of specific side effect cosine distance to confirm the pattern of side effects of drugs.

Here, word2vec can be used as one of the word embedding methods.

The construction device 100 according to an embodiment of the present invention may be a server that builds a pipeline for detecting side effects of drugs, and the construction device 100 runs a plurality of applications (ie, applications) desired by a manager or operator. Computers that can be installed and run, Ultra Mobile PCs (UMPCs), workstations, net-books, Personal Digital Assistants (PDAs), portable computers, web tablets, wireless phones ), mobile phone, smart phone, pad, smart watch, wearable terminal, e-book, portable multimedia player (PMP), It may be a portable game device, a navigation device, a black box or digital camera, or other mobile communication terminals. Thus, the building device 100 may need to install a separate program or application to build a pipeline. However, this is only one example, and a pipeline may be constructed by accessing a web page.

Steps of a method or algorithm described in connection with an embodiment of the present invention may be implemented directly in hardware, implemented in a software module executed by hardware, or implemented by a combination thereof. A software module may include random access memory (RAM), read only memory (ROM), erasable programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), flash memory, hard disk, removable disk, CD-ROM, or It may reside in any form of computer readable recording medium well known in the art to which the present invention pertains.

Although the embodiments of the present invention have been described with reference to the accompanying drawings, those skilled in the art to which the present invention pertains can be implemented in other specific forms without changing the technical spirit or essential features of the present invention. you will be able to understand Therefore, it should be understood that the embodiments described above are illustrative in all respects and not restrictive.

100: pipeline building device 110: communication module
130: storage module 150: analysis module
170: learning module 190: control module
151: collection unit 153: pre-processing unit
155: term generation unit 157: pattern analysis unit
159: classification unit

Claims (5)

In the pipeline construction method for drug side effect detection,
accessing a social channel based on Social Network Services (SNS) and collecting at least one piece of social data related to a target drug based on a first term set;
pre-processing the collected social data;
performing exploratory data analysis by extracting data related to side effects from the preprocessed social data based on a second term set;
Analyzing a drug side effect pattern for the target drug according to the analysis result and classifying it according to a preset category; and
Building or learning a drug side effect detection prediction model using the classification result,
Characterized in that the first term set comprises data sets consisting of at least one term representing each of at least one drug, and the second term set comprises a data set consisting of terms representing drug side effects.
A method for constructing a pipeline for drug side effect detection.
According to claim 1,
The preset category is set by dividing into known side effects, unknown side effects and indications for the target drug,
The drug side effect detection prediction model learns using data classified as known side effects as learning data, and among the data classified as unknown side effects, data accumulated more than a preset number of times is reclassified as known side effects, and then Characterized in that the drug side effect detection prediction model is used as learning data for learning,
A method for constructing a pipeline for drug side effect detection.
According to claim 1,
The exploratory data analysis,
This is achieved by extracting keywords appearing in the at least one collected social data and creating a visualized graph using the extracted keywords,
Characterized in that the visualized graph is generated using any one of a frequency analysis technique, an association analysis technique, and a word embedding technique,
A method for constructing a pipeline for drug side effect detection.
According to claim 1,
The second term set,
Characterized in that it is generated by mapping a drug side effect list for the target drug obtained based on SIDER and a pre-generated consumer terminology dictionary to WHO-ART,
A method for constructing a pipeline for drug side effect detection.
In the pipeline construction device for detecting side effects of drugs,
communication module;
a storage module for storing at least one piece of information or data for constructing a pipeline for detecting the side effect of the drug;
At least one social data related to a target drug is collected and pre-processed based on a first term set by accessing a social channel based on Social Network Services (SNS), and pre-processed based on a second term set. an analysis module for performing exploratory data analysis by extracting data related to side effects from social data, analyzing patterns of drug side effects for the target drug according to the analysis results, and classifying them according to preset categories;
a learning module for constructing or learning a drug side effect detection prediction model using the classification result; and
Collecting and pre-processing at least one social data related to a target drug based on the first term set, extracting side effect-related data from the pre-processed social data based on the second term set, and performing exploratory data analysis; , A control module for controlling to construct or learn a drug side effect detection prediction model using the classification result after analyzing the drug side effect pattern for the target drug according to the analysis result and classifying it according to a preset category,
Characterized in that the first term set comprises data sets consisting of at least one term representing each of at least one drug, and the second term set comprises a data set consisting of terms representing drug side effects.
Pipeline construction device for drug side effect detection.

Images