WO2021181212A1

WO2021181212A1 - Method of recommending medical team through user information and query analysis

Info

Publication number: WO2021181212A1
Application number: PCT/IB2021/051790
Authority: WO
Inventors: 조현철
Original assignee: 조현철
Priority date: 2020-03-09
Filing date: 2021-03-04
Publication date: 2021-09-16
Also published as: KR102395921B1; KR20210113769A

Abstract

A method of recommending a medical team including at least one doctor to a visitor having accessed a server comprises the steps in which a server: maintains a database in which an academic score, a team score, and a review score are distinguished from each other, in relation to a medical team; selects a query by using a search word input from the visitor; searches for medical content stored in a database included in the server or a database positioned out of the server by using the query; rates a weight for academic scores, a weight for team scores, and a weight for review scores by using the query; calculates rating scores of medical teams by applying the weight for academic scores, the weight for team scores, and the weight for review scores to academic scores, team scores, and review scores, respectively; and provides at least one medical team to the visitor according to the rating scores.

Description

2021/181212 1»（：1'year2021/051790

【Specification】

【Title of the invention】 Method of recommending a medical object through user information and query analysis 【Technical Field】 The present invention relates to a method of recommending a medical object such as a doctor, medical staff or hospital. It relates to a method for quantitatively evaluating a doctor through computer network or big data analysis and recommending medical subjects according to objective standards.

[Background Art] With the spread of the Internet, people can find useful information through Internet searches. For example, for information on famous restaurants, comfortable hotels, and the latest home appliances, people access and search portal sites, search blogs, user reviews, etc., and find it useful by referring to the number of 'likes'. get the information However, information in a certain professional field contains elements that cannot be evaluated with information such as user reviews or 'likes', and when such information is evaluated, there are things that are disadvantageous or dangerous to consumers. One such example is information about a doctor or hospital. Because the evaluation of disease diagnosis, treatment, procedure, or surgery is in a professional area that the general public cannot advise, and sometimes these evaluations are not factors related to the competence or competence of actual doctors or hospitals, and are not at all It is also influenced by unrelated factors, such as the friendliness of the doctor, the temporary feeling after medical treatment, the gender of the doctor, the attitude of the nurse, the appointment time of the hospital, and the cleanliness of the hospital. According to some overseas surveys, insurance coverage is the criteria for selecting surgeons. There was also a report that the above and the location of the hospital were evaluated as more important than the number of surgeries and the reliability of the medical staff (jamasurgery.com, February 2014). Recently, there are websites (Physician Rating Websites: PRWs) or organizations that can rate doctors and surgeons, such as Yelp.com, Healthgrade.com, Insurance plan, and Angie's List. However, these websites and organizations may also have limitations in that they consider only the reviews, evaluation scores, and regional distances of experienced practitioners in evaluating doctors or their abilities. Looking at Chinese Laid-Open Patent CN 104680458 A, a method for recommending a hospital or a doctor is disclosed. The above Chinese Patent Publication collects all prescriptions for each doctor, acquires treatment effects through various methods such as telephone contact, network, and visits after a certain period of time, and integrates them into a database to recommend medical subjects to the public. However, such a method requires a lot of time and control, it is impossible to collect all prescriptions, and it can be said that it is impossible to realize a high possibility of personal information leakage.

【Detailed Description of the Invention】

【Technical Issues】 The present invention is not limited to simply collecting reviews and evaluation scores of experienced people, but also provides information as a doctor's qualification itself, such as a doctor's academic background, career, history, etc. It provides a recommendation method that can recommend a medical target based on objective and quantitative criteria that reflect different weights according to the latest information and information on symptoms and diagnosis or treatment identified from queries. The present invention is a medical subject weight that can provide a standard that can quantitatively evaluate a doctor based on objective and academic performance, rather than a doctor's evaluation standard that rides on popularity. 2021/181212 1»（：1' Year 2021/051790 provides thousand methods. Reflecting the rapidly changing era of medical technology and information, the present invention considers objective and academic performance, but by weighting reasonable criteria, it is possible to provide recommendations according to the most accurate evaluation results to patients or consumers. Provides recommendations for medical subjects.

[Technical Solution] According to an exemplary embodiment of the present invention for achieving the above-described objects of the present invention, the method of recommending a medical subject including at least one doctor to a visitor accessing the server includes: Maintaining a database that classifies academic scores, target scores, and review scores for medical subjects, selecting a query using a search term from a visitor, Step of retrieving the medical content stored in the database, evaluating the weight for the academic score, the weight for the target score, and the weight for the review score using a query) step, the weight for the academic score, the target score calculating the evaluation score of the medical object by applying the weight for the weight and the weight for the review score to the academic score, the target score, and the review score, respectively, and providing at least one medical target to the accessor according to the evaluation score includes . For reference, in this specification, a 'search word' may be a word or a combination of words or a character directly provided by a user accessing the server or another server to obtain a result. The server or database may be a word, term, or a combination thereof selected as a medical term that can be recognized by the medical and medical database. Depending on the server, the search term can be used directly as a query, but otherwise, the database You can add, change, or select other queries to use. In the step of evaluating the weight for the academic score, the weight for the target score, and the weight for the review score using the selected query, at least one of symptom keywords, treatment keywords and other keywords is extracted from the query and , it may include the process of evaluating the weight for the academic score, the weight for the target score, and the weight for the review score using symptom keywords, treatment keywords, or other keywords extracted from the query. The method for recommending a medical object according to the present embodiment may further include receiving patient information about the patient from the accessor, and the patient may be the person or a third party specified by the patient information provided by the accessor. Accessors can input patient information every time, but they can use patient information by logging into the server and calling stored member information. Of course, the patient here may be the accessor himself, but may also be a family member or a third party. The server may use stored information already recorded in response to the patient information. However, the present invention is not limited thereto, and it is preferable to use the newly searched new information based on the access and input of a new query. The server may compare the recorded information with the new information, and add or replace the patient information with new information. The new information may include information input by a visitor or information collected from a public data server, private data server, or block chain that can be viewed on the server. Therefore, depending on the scope of access, databases such as the National Medical Insurance Administration, Health Insurance Review and Assessment Service, private insurance companies, national/public hospitals, private hospitals, private or commercial applications, and cloud You can update the necessary information with the latest information by accessing the website or blockchain. Here, the patient information can be said to be useful information for recommending a specific patient, objective and rational doctor target. As an example, basically, the patient information may include the patient's name, age, gender, etc. for patient identification, occupation, medical history, family history, smoking history, drinking history, drug history, travel information, It may include at least one of accident information, physical examination results, health examination results, blood test results, and imaging results. Here, the medical history may include medical history, surgical history, medical history, and other medical history, and may further include other social history such as hobbies or sports, in addition to smoking, drinking, occupation, travel, and the like. For reference, since patient information such as medical history may be easier to grasp the current status of a new medical history than a past medical history, in the step of calculating the evaluation score of a medical subject, stored information in which the weight for the new information has already been recorded It may be desirable to be rated equal to or higher than the weight for . Evaluation using a query may be specifically performed by reflecting symptoms and diagnostic keywords and treatment keywords, and may also reflect other keywords including other contents. For example, when a visitor enters a search term such as 'shoulder pain' or 'shoulder injury surgery', the server responds from the search term to 'arthritis', 'rotator cuff tear', 'shoulder rotator cuff', 'shoulder' ， You can select a combination of various queries according to medical terms and roots, such as 'rotator cuff surgery'. When a query is selected in this way, from the query, the server can be classified into symptom keywords (Dx keyword) related to symptoms, diseases, diagnosis, and the like.

When a query such as 'artificial joint surgery' or 'liposuction' is selected, the query is It can be classified as a treatment keyword (Tx keyword). If a query such as 'arthritis artificial joint' is selected, it may be classified as including both symptom keywords and treatment keywords. The symptom keyword may be specified using the disease name or diagnosis name classified by referring to the International Classification of Diseases (ICD). And, in particular doctor's name or a specific paper Name, 'people', 'woman doctor', 'confidential ^one, "Facilities", "cheap", etc. If the criteria of a different series than the symptoms and treatment sequence, which other keywords can be classified as However, if the extracted symptom keyword relates to cancer, it is also possible to evaluate the weight for the academic score higher than the weight for the target score and the weight for the review score. If the extracted keyword is a symptom keyword, the rarity, severity, incidence, frequency, acute/chronicity, emergency, cosmetic surgery of the corresponding symptom according to the symptom keyword At least one of beauty, insurance, and contagiousness is used as an evaluation factor, and the weight for the academic score, the weight for the target score, or the weight for the review score can be evaluated as a default value or a modified value. In addition, if the extracted keyword is a treatment keyword, at least one of difficulty (intractability) of corresponding treatment, hospitalization / outpatient (inpatient / outpatient), surgery / procedure / administration process (surgery) according to the extracted treatment keyword As an evaluation factor, the weight for the academic score, the weight for the target score, or the weight for the review score can be evaluated as a default value or a modified value. In addition, if the extracted keyword is other keywords, the patient according to the extracted other keywords At least one of mobility, privacy, social issue, and user tendency is an evaluation factor, and the weight for the academic score, the weight for the target score, or the review score The weights can be evaluated as default or modified values. The result and weight adjustment interface for the weight for the academic score, the weight for the target score or the weight for the review score can be provided to the accessor, and the accessor can adjust the three presented weights. If a specific weight is readjusted by the visitor, the evaluation score may be recalculated according to the readjusted weight, and at least one medical object may be provided to the visitor again according to the recalculated evaluation score. When maintaining a database that stores academic scores for medical subjects, academic scores reflect the thesis data of medical subjects, but the thesis data includes the number of papers, the number of citations and publication time data for each paper, information on the first author, and correspondence or journal information and evaluation grade, including responsible author information, co-author information, journal information included, journal impact factor, etc.; It may include at least one of the information and evaluation grade of the thesis itself, and the doctor's information and evaluation grade, such as the doctor's h-index. In addition to general abstract and thesis data, academic scores are presented in seminars, symposiums, and academic conferences, as well as abstracts, opinions, letters, research plans, editorials, and other forms of literature, patents, news, television, newspapers, etc. ， It can target various data including media materials such as magazines and SNS (social network service). In addition, various metadata services such as Altmetric, Medeley, and PlumX metrics that aggregate these data can be used. 2021/181212 1»（：1' Year 2021/051790 The academic score is calculated by reflecting the data at the time of publication. can be useful In addition, it may be useful not only for the academic score, but also for the target score and the review score to have the latest score calculated high. The distance between the user and the medical object can be an important consideration in recommending the medical object. When calculating the evaluation score of a medical subject reflecting the academic score, subject score, and review score, it may be common sense to recommend a medical subject close to the user first. However, in certain circumstances, it may be useful to recommend a medical target even if you are at a great distance. For example, when it is judged by user information, when it is determined that the patient's condition has recently deteriorated, or when it is determined that higher-level diagnosis and treatment may be necessary, such as when the patient's condition has recently deteriorated or does not respond to treatment in the same disease or hospital; When judging by a query, in cases where there is a high possibility of a diagnosis and symptoms such as cancer or serious disease, and in situations such as a query in which a higher-level special diagnosis and treatment is required, it is not bound by common sense recommendation standards. It can be useful to recommend medical subjects that are far away but have particularly high academic and subject scores. In this case, the weighting associated with the calculation algorithm of the evaluation score of the medical object may adjust the user and third party weights. In the present specification, the medical object is not limited to a doctor, and may be a medical team to which the doctor belongs or a hospital to which the doctor belongs.

[Effects of the Invention] According to the treatment method for a medical object of the present invention, it is not limited to simply collecting reviews or evaluation scores of experienced people, but it is possible to recommend a medical object on an objective basis. In particular, the information entered by the user, the newly updated information, Even non-experts can obtain the evaluated results from an expert point of view by applying the identification that belongs to them. In particular, unlike providing a uniform recommendation result irrespective of the patient's situation or type of disease, it provides practically helpful results by considering the patient's situation or the rarity or difficulty of the disease. can provide results. In addition, reflecting the trend of the times in which medical technology and information are rapidly changing, objective and academic performance according to real-time trends is considered, but by weighting the criteria close to time, it is possible to provide recommendations according to the most accurate evaluation results to consumers. have. In addition, under the same conditions, the nearest medical target is usually recommended first, but by judging the user's information and query, when a severe condition or disease or difficult diagnosis, procedure, and treatment are needed, even if it is far away, medical care with a high academic score and target score The target can be recommended first. For example, if the rarity, severity, incidence, and acute/chronicity of the corresponding symptom according to the extracted symptom keyword is above a predetermined threshold, academic score and A medical subject having a relatively high subject score compared to the review score may be preferentially recommended. In addition, other criteria may be applied. For example, if the contagiousness of the corresponding symptom according to the extracted symptom keyword is above a predetermined threshold value, a medical target close to the patient specified by the visitor may be recommended first.

[Brief Description of Drawings] FIG. 1 is a diagram for explaining a process of a method for recommending a medical object according to an embodiment of the present invention. 2021/181212 1» (: 1' year 2021/051790 FIG. 2 is a diagram for explaining a weight evaluation process according to a query according to a method for recommending a medical object according to an embodiment of the present invention. 3 is a diagram for explaining a process of calculating an academic (show score) according to a method for recommending a medical subject according to an embodiment of the present invention. FIG. 4 is a method for recommending a medical subject according to an embodiment of the present invention. It is a diagram for explaining the server and network relationship for

[Mode for Implementation of the Invention] Hereinafter, preferred embodiments of the present invention will be described in detail with reference to the accompanying drawings, but the present invention is not limited or limited by the embodiments. For reference, in this description, the same numbers refer to substantially the same elements, and the contents described in other drawings under the above rules may be cited and described, and the contents determined to be obvious to those skilled in the art or repeated may be omitted. 1 is a diagram for explaining a process of a method for recommending a medical object according to an embodiment of the present invention, and FIG. 2 is a view for explaining a weight evaluation process according to a query according to the method for recommending a medical object according to an embodiment of the present invention 3 is a view for explaining a process of calculating a scientific (show score) according to a medical subject recommendation method according to an embodiment of the present invention, and FIG. 4 is a medical subject according to an embodiment of the present invention. It is a diagram for explaining the relationship between the server and the network for performing the recommendation method. Referring to Figs. 1 to 4, the accessor

It is possible to access the server 110 through login or non-login through the computer 20, etc. (比120), the accessor can input a search term while connected to the server 110 and the network, and the server 110 140) can select a query using the search word input from the accessor (比140). The server 110 may store the selected query in the database 114, etc., and the search unit 113 of the server may inquire about patient information through an external database based on the query or search and extract thesis data, etc. The calculation unit 112 of the server may calculate an academic score in advance by reflecting the weight of the dictation using the patient information or taking the thesis data into consideration. Then, the evaluation score is calculated in consideration of the academic score, the target score, the review score and their weights stored in the database 114 (S160), and at least one medical object is recommended to the visitor according to the final or temporary evaluation score. It can be (S170). To this end, the server 110 may include an operation unit 112 , a search unit 113 , and a database 114 . The server 110 may further include a communication unit 115 for transmitting and receiving data with the outside, and the communication unit 115 may receive a query from a visitor or a search result from the external database 30 . In addition, the communication unit 115 may transmit a search keyword according to the query to the external database 30 , and may provide a medical target recommendation result according to the evaluation score to the visitor in real time. The external database 30 may be a public data server, a private data server, or a block chain, etc., and in order to read patient information, the target is the National Medical Insurance Service, Health Insurance Review and Assessment Service, private insurance company, and the National Health Insurance Service. / It can be a public hospital, a private hospital, a private or commercial application, a database such as a cloud, or a blockchain, etc., and in order to search for medical-related papers, it is a database that can be accessed in real time, such as Pubmed, Scopus, Google Scholar ( Google scholar) , SCOPUS, etc. can be targeted, and as a domestic search site, Naver full-text search, NDSL, KCI, RISS, etc. can be used, Altmetr ic, Mendeley meta Data services are also available. In addition In order to calculate academic scores, data such as citation, use, capture, mention, and social media utilization may be used in newspapers, magazines, radio, TV, internet portals, social media, social network services, etc., and Altmetric and PlumX metrics You can also use metadata services such as As described above, the medical subject may be understood as a concept including a doctor, physician, surgeon, medical staff or hospital, and the recommendation result according to the condition desired by the accessor is a specific doctor, medical team, or hospital name. can recommend As will be described later, evaluation scores can be collected based on the academic performance of individual doctors. It is possible to collect the academic achievements of doctors in hospitals. According to the medical subject recommendation method according to the present embodiment, the server 110 generates a database 114 that classifies the academic score (AP score), the subject score (PP score), and the review score (RP score) for the medical subject. can keep Here, academic scores (academic properties or academic scores) are evaluation indexes determined by the academic activities of medical subjects, and the material to be evaluated is not limited to papers. In addition to general paper materials, seminars, symposiums, and academic conferences Various types of documents such as presentation materials and abstracts, opinions, letters, research plans, editorials, and contributions, patents, news, media materials such as television, newspapers, magazines, and various data including SNS (social network service) can be targeted. In addition, various metadata services such as Altmetric, Medeley, and PlumX metrics that synthesize these data may be used. Paper data may include various types of information. For example, depending on the medical subject 2021/181212 1^(:1^2021/051790 The number of searched doctor's articles, the number of citations for each article, and publication date data, first author information, corresponding or responsible author information, co-author information, journal information listed, journal Journal information and evaluation grade, including the impact factor of It may include at least one of information and evaluation grades of doctors, such as index, etc. In addition, the thesis data may include the university to which the doctor to be searched belongs, the current status of the doctor, and the years of service after the major.

In the case of SNS (social network service) data, various information may be included. Citations (citation indexes) , patent citations , clinical citations , policy citations, etc.) , usage: clicks , downloads ( downloads) , views , library creation ( 1 ibrary holdings ) , video plays etc.), captures; bookmarks , code forks , favorites , notes Seeds (fol lowers) , readers (readers) , viewers (watchers), etc.) , lines and saves (exports/ saves) , subscribers (subscribers, etc.) , comments (blog posts) , comments (comments) , reviews , Wikipedia references , news media, etc.) , and social media; shares , likes , comments , tweets , data of the first (chat, etc.), but is not limited thereto. Although it is possible to search for a specific doctor, it is also possible to separate the results by doctor after collecting all the thesis data or other academic achievements that are searched based on a specific symptom, treatment, disease name, etc. Physician properties or physician scores (Ex. 八: 比!!), a quantitative summary of history, career, and medical and medical activities. It may include the presence and type of qualifications, the presence and type of specialists, and other medical-related qualifications and licenses. History and experience are the age, gender, training, major and type and name of the medical institution where they worked, the length of experience according to the qualifications they have, their field of expertise and experience revealed by the medical subject, meetings, meetings, conferences, and research groups they participated in. , seminars, task force, various medical and medical-related organizational activities such as committees, and awards and commissioning experience. Medical activity refers to the medical and medical experience of a medical subject, and may include the number of patients treated and treated, the number of procedures and surgeries, the number of prescriptions, the number of insurance claims, the number of referrals and referrals, and related periods. have. Qualifications, history and experience of the above target score, and examples of medical and medical activities are merely examples and include, but are not limited to. Review properties or review scores are evaluation indices determined by the evaluation and reputation of patients, their families, related workers, etc. on the medical or social activities of a medical subject, The data can be an index reflecting reviews of prior experience, word of mouth, recommendations of fellow doctors, data and evaluations on portals, PRW websites and social network services, medical-related claims and lawsuits, criminal records, etc. The database for academic scores, target scores, review scores, etc. can be upgraded in real time, the target can be added according to the query of the visitor, and can be built internally on the server based on the members registered in the medical association. . In a situation where the database 114 is maintained, the server 110 may recognize the access of the visitor (S120). Here, the access of the accessor is the accessor's login (log-in) or sign-in. It can be recognized through (sign-in), and even if you are not registered as a member, you can recognize the access of the visitor in the process of specifying a patient or entering a query. If the server 110 can recognize the information about the accessor or the patient, before or after receiving the query, the server 110 uses the patient information through the internal database 114 or the external database 30, etc. can be updated (S130). Upon receiving the query from the accessor (S140), the server 110 may perform weight evaluation for calculating an evaluation score using the received query (S150). In order to perform weight evaluation through a query, the operation unit 112 of the server 110 may extract at least one of a symptom keyword (Dx keyword), a treatment keyword (Tx keyword) and other keywords (ETC keyword) from the query. have. Then, using the extracted symptom keyword, treatment keyword, or other keywords, the query entered by the patient is divided into disease/symptom/diagnosis, treatment method, and other matters, and the rarity, severity, _{The weight for the academic score (a ap} ), the weight for the target score (a _pp ), and the weight for the review score (a _rp ) are calculated by comprehensively considering the incidence rate, the difficulty of treatment, the operation status, and the patient’s behavior. It can be calculated and evaluated. When these weights are calculated, the academic score (X _ap ), the target score, which is first calculated by the weight for the academic score (a _ap ), the weight for the target score (a _pp ), and the weight for the review score (a _{rp )} (X _pp ), and a review score (X _rp ), respectively, to yield a secondly weighted academic score (X _ap *), subject score (X _PP *), and review score (X _rp *), or Using these, it is possible to calculate the evaluation score of the medical object (S160). And according to the evaluation score, at least one medical object may be provided to the accessor (S170). When the visitor accesses the server 110, the server 110 enters the personal information of the visitor. You can call the patient information or refer to the patient information entered by the accessor. As described above, the 'patient' here may be the accessor or a third party. The server 110 may use stored information already recorded in the database in response to the specified patient information. However, the present invention is not limited thereto, and it is desirable to collect the latest information based on the access and input a new query and use the newly searched new information. When new information is input, the server 110 can compare the stored information already recorded in the database 114 with the new information, and adds new patient information to the database 114 . Or you can replace the old content with the new one. The new information may include information input by a visitor or information collected from a public data server, private data server, or block chain that can be viewed on the server. Therefore, depending on the scope of reading, external databases (30), for example, the National Medical Insurance Administration, Health Insurance Review and Assessment Service, private insurance companies, national/public hospitals, private hospitals, private or commercial applications, databases such as cloud, Alternatively, the necessary information can be updated with the latest information by accessing the blockchain, etc. In this embodiment, the patient information may include the patient's name, age, gender, etc. for patient identification, in addition to the patient's occupation, medical history, family history, smoking history, drinking history, drug history, travel information, It may include accident information, physical examination results, health examination results, blood test results, imaging results, and the like. Here, the medical history may include medical history, surgical history, medical history, and other medical history, and may further include other social history such as hobbies or sports, in addition to smoking, drinking, occupation, travel, and the like. For reference, since patient information such as medical history may be easier to grasp the current status of a new medical history than a past medical history, in the step of calculating the evaluation score of a medical subject, stored information in which the weight for the new information has already been recorded It may be desirable to be rated equal to or higher than the weight for . In the above, it is assumed that the patient information is specified as one person, but in some cases, the accessor may input only diseases, symptoms, diagnosis, treatment, and other matters without patient information. Even in the present invention, the accessor does not necessarily need to specify a patient, and it is also possible to input a query for a general patient without specifying the patient. When the visitor provides a query, the operation unit 112 of the server 110 may readjust the weight for correcting and supplementing the evaluation score through analysis of the input query. In general, the evaluation using a query may be performed by specifically reflecting symptom keywords and treatment keywords, and may also reflect other keywords including other contents. For example, if a visitor inputs a query such as 'arthritis' or 'shoulder pain', it can be classified as symptom keyword (Dx keyword) related to symptoms, disease, diagnosis, etc., and 'arthroplasty' or 'liposuction' Enter a query such as , and it can be classified as a treatment keyword (Tx keyword) related to treatment. If a query of a complex concept such as 'arthritis artificial joint' is input, it may be classified as including both symptom keyword and treatment keyword. The symptom keyword may be specified using a disease name or diagnosis name classified by referring to the International Classification of Diseases (ICD), and may be specified using a conventionally used disease name or diagnosis name, and other inconvenient parts or actions. And, if a keyword from a category other than the symptom or treatment category is entered, such as the name of a specific doctor or a specific area, 'female doctor', 'confidentiality', 'facility ^{1', 'cheap',} It can be classified by other keywords. However, if the extracted symptom keyword relates to cancer, it is also possible to evaluate the weight for the academic score higher than the weight for the target score and the weight for the review score. This is because, in the case of cancer or other fatal diseases, it may be more efficient to judge based on academic experience or cases. If the extracted keyword is a symptom keyword, the rarity of the corresponding symptom according to the symptom keyword, severity, incidence, frequency, acute/chronicity, emergency, cosmetic surgery At least one of beauty, insurance, and contagiousness is used as an evaluation factor, and the weight for the academic score, the weight for the target score, or the weight for the review score can be evaluated as a default value or a modified value. For example, when the severity is high, the academic score is reflected by about 10-20% addition, when the intractability is high, the academic score and the subject score are reflected by about 20-30% addition, and when the rarity is high, the academic score is about 40 It is possible to reflect an additional -50%. In addition, if the extracted keyword is a treatment keyword, at least one of difficulty (intractability) of corresponding treatment, hospitalization / outpatient (inpatient / outpatient), surgery / procedure / administration process (surgery) according to the extracted treatment keyword As an evaluation factor, the weight for the academic score, the weight for the target score, or the weight for the review score can be evaluated as a default value or a modified value. For example, if surgery is necessary for treatment, about 30-50% addition to the target score may be reflected, and if hospitalization is necessary, about 10-30% addition to the review score may be reflected. In addition, depending on the length of hospitalization, the review score may be reflected as important. In addition, if the extracted keyword is other keywords, at least one of patient mobility, privacy, social issue, and user tendency according to the extracted other keywords As an evaluation factor, the weight for the academic score, the weight for the target score, or the weight for the review score can be evaluated as a default value or a modified value. The result and weight adjustment interface for the weight for the academic score, the weight for the target score or the weight for the review score can be provided to the accessor, and the accessor can adjust the three presented weights. If a specific weight is readjusted by the visitor, the evaluation score may be recalculated according to the readjusted weight, and at least one medical object may be provided to the visitor again according to the recalculated evaluation score. _{When maintaining a database that stores academic scores (X ap} ) for medical subjects, the academic score reflects the thesis data of medical subjects, but the thesis data includes the number of thesis, the number of citations for each thesis, and data at the time of publication. 1 Journal information and ratings, including author information, correspondence or lead author information, co-author information, journal information listed, journal impact factor, etc., and article citation, use, capture, mention, and social media It may include at least one of information and evaluation grade of the thesis itself such as utilization and exposure frequency, and doctor's information and evaluation grade, such as the doctor's h-index. In addition to general abstract and thesis data, academic scores are presented in seminars, symposiums, and conferences, as well as various types of documents such as abstracts, opinions, letters, research plans, editorials, contributions, patents, news, television, newspapers, etc. ， It can target various data including media materials such as magazines and SNS (social network service). In addition, Altmetric, which aggregates these data, Various metadata services such as Mendeley and PlumX metrics can be used. The academic score (X _ap ) stored in the database is calculated by reflecting the data at the time of publication. Under the same conditions, it may be useful for the academic score of the latest paper to be calculated equal to or higher than that of the older paper. The server 110 may provide a medical object recommendation result according to the evaluation score to the visitor in real time. To this end, academic scores can also be calculated in real time or updated in units of a certain period to form a predetermined default value. As the external database 30 used to calculate academic scores, etc., medical-related thesis search is possible, and as a database that can be accessed in real time, Pubmed , Scopus , Google scholar , Scopus, etc. can be targeted. As a domestic search site, Naver full-text search, NDSL, KCI, RISS, etc. can be used, and Altmetr ic and Mendeley metadata services can be used. In addition, data such as citation, use, capture, mention, and social media utilization can be used in newspapers, magazines, radio, TV, internet portals, social media, social network services, etc. to calculate academic scores, Altmetr ic and metadata services such as PlumX metrics. As described above, the medical subject may be understood as a concept including a doctor, physician, surgeon, medical staff or hospital, and the recommendation result according to the condition desired by the accessor is a specific doctor, medical team, or hospital name. can recommend Academic scores can be collected based on the academic performance of individual doctors. In this case, it is possible to collect the academic achievements of doctors in the hospital. When a query is entered, the search unit 113 may request a search result from the external database 30 as it is the entered query, but in some cases, the input query is standardized to facilitate searching or similar search terms are added. Thus, academic materials such as papers can be searched through the external database 30 . The searched materials are not limited to papers, and in addition to general abstracts and paper data, various types of documents such as seminars, symposiums, and conferences, abstracts, opinions, letters, research proposals, editorials, contributions, etc. ， It can target various data including media materials such as news, television, newspapers, magazines, and SNS (social network service). In addition, various metadata services such as Altmetric, Medeley, and PlumX metrics that synthesize these data can be used. Paper data may include various types of information. For example, the number of doctor's papers searched according to medical subject, the number of citations and publication time data for each paper, first author information, corresponding or responsible author information, co-author information, journal information included, journal impact factor ( Journal information and evaluation grade, including impact factor), and information and evaluation grade of the thesis itself, such as citation, use, capture, mention, and number of social media usage and exposure, and doctor's information such as the doctor's h-index And it may include at least one of the evaluation grade. In addition, the thesis data may include the university to which the doctor to be searched belongs, the current status of the doctor, and the number of years of service after the major. When the thesis data is collected for each doctor, the computing unit 112 of the server may calculate the academic score of the medical subject according to various methods. Here, the evaluation score may be calculated using only the thesis data, but is not limited thereto, and the target score and the review score 2021/181212 1^(:1^2021/051790 It is possible to calculate the medical object evaluation score by reflecting.

The above equation is an equation for calculating the medical subject evaluation score 0, and in this embodiment, it can be calculated including the review score (/) of the prior experience, the target score (¾/) and the academic score (¾/). Score according to regional proximity according to rarity, severity, acuteness, chronicity, etc. judged by symptom keywords, etc.

factors can be taken into account. In order to calculate an evaluation score of 0°, a weight may be varied according to its importance. In the above equation, ¾· _!) is the weight of the review, ¾ (1 _{) is the weight of the thesis data, ᄆ !)} is the weight of the target score based on the qualification of the medical subject, history, etc., 31。^^ is the social relationship between the patient and the medical subject Alternatively, it may be weighted by geographic proximity. If all are determined to be the same factor, the weights are all 1, but may be specified as a number less than 1 or greater than 1 depending on the weight. However, the reviews of prior experience alone cannot reflect these recent trends, and the papers can be evaluated differently depending on the level of the journal in which they are published as well as how recently the data is published. Therefore, in relation to the academic score (¾/), the thesis data extracted through the external database may include the publication time data of the thesis, and the operation unit 112 of the server 110 calculates the evaluation score. It is possible to evaluate the academic score differently according to the data at the time of publication included in the data. 2021/181212 1»（：1' Year 2021/051790 In calculating the academic score (¾/), the academic score of the paper with the latest publication date is the academic score of the older paper under the condition that all other conditions are the same and the publication date is different. It is desirable to be calculated equal to or higher than the score. Here, the basic academic score 0 ) may have a different academic score depending on its own level, the level of the journal, whether it is the first author, whether it is a co-author, and the like. For example, if the author of the thesis is the lead author or the first author, or the corresponding author (0)0'631)011 (1111 yo) Score (¾) can be obtained In addition, the grade or classification of published journals may vary, such as XI, 然比, 1, ^10, or non-listed journals in the index, for example, In this case, the journal score (¾) is 2 points, in the case of a ^1 grade journal or academic journal, the journal score (¾) is 1 point, in the case of a ^10 grade journal or academic journal, the journal score (¾) is 0.5 points, and the In the case of other journals, 0.2 points can be set. In addition, score 0 ) according to the impact factor of the journal can be additionally reflected. Considering these, academic score 0 ) can be set as follows.

In addition to this, various conditions can be set, as shown below, by further strengthening the influence of the score () ) in the academic score () ) (1+ ¾ can be considered. =¾ ^· Three ¾ · ¾ Also, as shown below, Impact By using the factor reflection value (0 << 1), it is also possible to adjust the degree to which the impact factor is considered according to the reflection value.

2021/181212 1»（：1' year2021/051790 Also, if there are multiple papers, the sum of academic scores using paper data

(¾ _! 0 ₁ ) may be the sum of the individual scores 0 (1 ) ^* ) of each paper, which can be expressed as follows.

Here, ⁾ (1) is the individual score of each thesis, and ¾(1) may be the weight of the publication time of each thesis. The academic score 0 ) calculated in this way can be reflected in the evaluation score 0 . An evaluation score of 0° may be performed for all or some of the doctors according to the query, and a predetermined number of doctors, such as 5, 10, etc., among the doctors who received a high evaluation score may be reflected in the list. The review score (¾/) and the target score (¾/) can also be determined by various methods, but weights appropriate for the times can be given to each variable. For example, it is possible to calculate a score for a certain period of time, and give more weight to the recently added review score and target score element. For this purpose, factors that can be reflected in the target score and review score for 10 years are calculated on a yearly basis, and weights can be given to the recently added ones. _{The target score (¾/) reflects the weight of the target score ( ¾1)} ) in the qualifications, history and career of medical subjects, medical and medical activities, etc. Specifically, the degree of the medical subject, the presence or absence of a doctor's license, the number of professional qualifications, details The number of professional qualifications, other medical qualifications and licenses, and university rankings to which medical subjects belong can be reflected. For example, the degree of a medical subject may be awarded in the order of bachelor's degree, master's degree, and doctoral degree. 2021/181212 1»（：1' Year 2021/051790 Materials published through one media are available. The method of calculating the target score may have different weights depending on the values of the society at the time. In calculating the target score, history or experience may also be reflected. To this end, the age, gender, hospital of trainees and specialists working in hospitals, length of medical experience, period of professional experience, period of specialized specialist experience, period of other medical-related qualifications and licenses, number of specialties and fields of experience, The number and ranking of participating meetings, the number of awards, the number of appointments, and other medical and medical-related appointments and awards can be reflected. Specifically, age generally has a small stake reflected in the score, but some users may prefer a specific age group, and may be reflected relatively significantly in majors such as neurosurgery. Gender is also mostly not scored, but in a specific field that prefers a specific gender (approximately . Obstetrics and Gynecology) may decide otherwise. With respect to specialized fields and fields of experience, the more the number of specialized fields that can be objectively selected from the keywords of the thesis, the higher the score can be reflected. It can be highly reflected . Participation meetings may be reflected in scores differently depending on the number of various medical and medical-related organizational activities such as meetings, meetings, conferences, research groups, seminars, task forces, committees, etc., and content related to awards and appointments in the relevant field can also be reflected in the target score. In calculating the target score, medical and medical activities may also be reflected. These factors can take into account the number of patients recorded in medical activities, the number of tests, the number of readings and judgments, the number of diagnoses, the number of procedures, the number of surgeries, the number of prescriptions, the number of insurance claims, the number of requested patients, the number of referred patients, etc. . Basically, the higher the score for medicine and medical activity, the higher the score can be. The number of diagnoses can be weighted according to the severity of the diagnosis, in which case the severity is 2021/181212 1» (：1' year 2021/051790 You can refer to the classification of plains, etc.). The more the number of procedures, the number of surgeries, the number of prescriptions, the number of insurance claims, etc., the higher the insurance claim cost, the more weight can be given. may vary. The review score（ / ） means reviews, word-of-mouth, recommendations, comments,

Data, claims, civil complaints, lawsuits, criminal records, etc. and weights accordingly may be reflected. In general, reviews about medical subjects, word of mouth, recommendations, comments,

The data may have different scores depending on the number and positive/negative, but it is necessary to select or adjust the weight, etc., because non-objective, malicious reviews or advertisement reviews may be included. Data can be largely classified into citation, use, capture, mention, and distribution of social media, and scores can be reflected differently according to the number. In addition,

Different weights can be assigned to additional ripples such as likes/shares/retweets in the data. The score or weight may increase as there are fewer claims, civil complaints, lawsuits, criminal records, etc. during the review. Of course, the criteria for reflecting the above-described academic score, target score, and review score are merely examples, and the present invention may not be limited to the above criteria. Reflecting the trend of the times in which medical technology and information are rapidly changing, objective and academic performance according to real-time trends is considered, but by weighting the criteria close to time, it is possible to provide recommendations according to the most accurate evaluation results to consumers. Under the same conditions, a nearby medical subject is usually recommended first, but by judging the user's information and query, when a severe condition or disease or difficult diagnosis, procedure, and treatment is required, medical subjects with high academic scores and target scores are selected even if they are far away. can recommend first 2021/181212 1» (：1' year2021/051790 There is. For example, if the rarity, severity, incidence, and acute/chronicity of the corresponding symptom according to the extracted symptom keyword is above a predetermined threshold, academic score and By giving a relatively high weight to the target score, it can be adjusted so that medical subjects with relatively high academic and target scores compared to the review score are recommended with priority. In addition, if rapid treatment is required, on the contrary, it is also possible to reflect the ^{score for proximity (¾。 _ * ) high.} For example, according to the extracted symptom keyword, if the contagiousness of the corresponding symptom is greater than or equal to a predetermined threshold, a weight (31. You may. Through this, the accessor can obtain a list of doctors or medical subjects according to the objective criteria, and by using the results reflecting academic scores, target scores, review scores, etc. You can check where you belong. Of course, the above embodiment relates to recommending one or a plurality of medical objects according to a query. Alternatively, it is also possible to provide only an evaluation score for a specific medical object. As described above, although described with reference to preferred embodiments of the present invention, those skilled in the art can variously modify and modify the present invention within the scope without departing from the spirit and scope of the present invention as set forth in the following claims. You will understand that it can be changed.

Claims

2021/181212 1»（：1' year2021/051790【Scope of claim】

[Claim 1] In the method of recommending a medical subject including at least one doctor to a visitor connected to a server, the server comprising the steps of maintaining a database for classifying academic scores, subject scores, and review scores for the medical subjects ; selecting a query using the search word from the accessor; searching for medical content stored in a database included in the server or a database external to the server by using the query; evaluating (vs. 1) a weight for the academic score, a weight for the target score, and a weight for the review score by using the query; calculating an evaluation score of a medical subject by applying a weight for the academic score, a weight for the target score, and a weight for the review score to the academic score, the target score, and the review score, respectively; and providing at least one medical object to the accessor according to the evaluation score.

[Claim 2] The method of claim 1, wherein the step of evaluating the weight for the academic score, the weight for the target score, and the weight for the review score (vs. 11) using the query comprises: Extracting at least one of symptom keywords, treatment keywords, and other keywords from the query, and extracting the symptom keyword, the treatment keyword, or the other key from the query A method for recommending medical subjects, characterized in that the weight for the academic score, the weight for the target score, and the weight for the review score are evaluated using the word 2021/181212 1» (: 1' year 2021/051790 .

[Claim 3] The method of claim 1, further comprising: receiving patient information about the patient from the accessor, wherein the patient is the person or a third party specified by the patient information provided by the accessor How to recommend a medical target.

[Claim 4] The medical subject according to claim 3, wherein the server updates the stored information for the patient information using stored information previously recorded in response to the patient information or newly searched new information. Recommended method.

[Claim 5] The medical subject according to claim 4, wherein the new information is information input by the visitor or includes information collected from a public data server, a private data server, or a block chain that the server can read. Recommended method.

[Claim 6] The patient information according to claim 4, wherein the patient information includes the patient's name, age, gender, occupation, medical history, family history, smoking history, drinking history, drug history, travel information, accident information, physical examination results, and medical examination results. ， blood test A method for recommending a medical subject, characterized in that it includes at least one of an examination result and an imaging result.

[Claim 7] The method of claim 4, wherein in the step of calculating the evaluation score of the medical subject, the weight for the new information is rated equal to or higher than the weight for the previously recorded stored information. A method of recommending a medical target using

[Claim 8] The method of claim 2, wherein the symptom keyword is specified using a disease name or diagnosis name classified with reference to International Classification of Diseases (ICD).

[Claim 9] The method of claim 2, wherein if the extracted symptom keyword relates to cancer, evaluating the weight for the academic score higher than the weight for the target score and the weight for the review score A method of recommending a medical target characterized by its characteristics.

[Claim 10] According to claim 2, according to the extracted symptom keyword, rarity, severity, incidence, frequency, acute / chronicity, emergency ) , cosmetic plastic surgery purpose (beauty), insurance coverage (insurance), and at least one of infectiousness as an evaluation factor, the weight for the academic score, the weight for the target score or a method for recommending a medical subject, characterized in that evaluating a weight for the review score.

[Claim 11] The evaluation of claim 2, at least one of difficulty (intractability) of corresponding treatment, hospitalization/outpatient (inpatient/outpatient), and surgery/operation/medication process (surgery) according to the extracted treatment keyword A method for recommending a medical subject, characterized in that the weight for the academic score, the weight for the target score, or the weight for the review score is evaluated with is as a factor.

[Claim 12] The evaluation factor according to claim 2, wherein at least one of patient mobility, privacy, socialissue, and usertendency is used according to the extracted other keywords. and, a weight for the academic score, a weight for the target score, or a weight for the review score.

[Claim 13] The method of claim 1, wherein a weight for the academic score, a weight for the target score or a weight for the review score, and a weight adjustment interface are provided to the visitor, and the weight is adjusted by the visitor. When readjusted, the above rating is based on the readjusted weight. A method for recommending a medical object, characterized in that the score is recalculated, and at least one medical object is provided back to the accessor according to the recalculated evaluation score.

[Claim 14] The method of claim 1, wherein when maintaining the database for storing academic scores for the medical subject, the academic score reflects the thesis data of the medical subject, wherein the thesis data is the number of thesis, each thesis A method of recommending a medical target, characterized by including at least one of data at the time of publication, first author information, corresponding or responsible author information, co-author information, journal information included, journal impact factor, and doctor's h-index .

[Claim 15] The academic score of claim 14, wherein the academic score is calculated by reflecting the data at the time of publication, and under the same conditions, the academic score of the latest paper is calculated to be equal to or higher than that of the older paper. A method of recommending a medical target using

[Claim 16] The method according to claim 2, wherein, according to the extracted symptom keyword, the rarity, severity, incidence, frequency, and acute/chronicity of the corresponding symptom are set at predetermined boundaries. If the value is greater than or equal to the value, the medical subject recommendation method, characterized in that it is possible to preferentially recommend a medical subject having a relatively high academic score and the subject score compared to the review score.

[Claim 17] 2021/181212 1» (：1' year 2021/051790 According to the extracted symptom keyword, if the infectiousness of the corresponding symptom is above a predetermined threshold, priority is given to a medical subject close to the patient specified by the accessor. A method of recommending a medical target, characterized in that it can be recommended.

[Claim 18] The method of claim 1, wherein the medical subject is a doctor, a medical team to which the doctor belongs, or a cause of a disease to which the doctor belongs.