JP2018045378A - System and method for processing computer-assisted name identification of anonymously processed information, server device, and program - Google Patents

Abstract

Personal information such as patient information is anonymized by name identification processing, and exchange of personal information with enhanced confidentiality is enabled in a plurality of medical facilities. The system includes a survey form input unit that accepts input of a survey form in which patient information is recorded, a half-width confirmation unit that performs a half-width check of the first and last names of patient information in the survey form, and different kanji for first and last names. Male / female judgment based on the first character of the last name in the last name, and the storage means that stores the standardized correspondence table that replaces the fonts with the standardized kanji, the standardization method that standardizes each kanji of the last name by matching with the standardized correspondence table, and conversion Gender determination means for performing, rarity determination means for determining the rarity of surnames in first and last names, information acquisition means for acquiring information from first and last names, hashing means for converting patient information into a hashed character string, A survey form collecting means for collecting survey form data composed of hashed character strings is provided. [Selection] Figure 1

Description

  The present invention relates to a name identification processing system, a name identification processing method, a server device, and a program for anonymous processing information used in various medical facilities, for example.

  Conventionally, for example, in academic research in which a plurality of medical facilities or the like participate, various survey forms in which names are anonymized are used to manage and collect user attribute information. By using such a survey form, it is possible to provide a survey form from a medical facility to each center in an anonymized state when, for example, information about multiple medical facilities is centrally managed by the center. It becomes. In recent years, with the introduction of the Personal Information Protection Law, etc., it has become extremely important to conceal personal information. However, by utilizing such a questionnaire, the personal information can be safely disclosed without being disclosed to the outside. It is possible to communicate with. Today, various methods are used for the name identification processing method in order to prevent such leakage between specific organizations.

  As this type of technology, for example, Patent Document 1 is a technology using a server, and for a record related to name identification, an attribute value related to the record is converted into a plurality of attribute values, and the plurality of attribute values are concealed. In addition, a name identification processing system that records and appropriately reads the record to generate name identification data is disclosed.

JP2013-149056A

  However, in general name identification processing, information that can identify a patient individual is converted into a sequence of random fixed-length alphanumeric characters using a hash function or the like, making it difficult to identify the patient individual outside the facility. That was a problem.

  The same applies to the technique disclosed in Patent Document 1 described above. Furthermore, in the conventional name identification process, for example, it was not possible to make a gender determination based on the first letter of the name or to determine the rarity of the last name.

  The present invention has been made in view of such a problem. For example, personal information such as patient information is anonymized in a plurality of medical facilities and the like, and the name identification accuracy is improved while further improving confidentiality. The present invention relates to a name identification processing system, a name identification processing method, a server device, and a program.

  In order to solve the above-described problem, a name identification processing system according to an aspect of the present invention includes a survey form input unit that receives input of a survey form in which patient information is recorded, and an input check of a first name and last name of the patient information in the survey form. Checking means for storing, storage means for storing at least a standardized correspondence table for replacing each kanji of the first and last names with standardized kanji characters, and matching each kanji of the first and last names with the standardized correspondence table for conversion Standardization means for standardization, gender determination means for determining gender with the last letter of the last name, rarity determination means for determining the rarity of the last name among the first and last names, and information for obtaining information from the first and last names An acquisition unit, a hashing unit that converts the patient information into a hashed character string, and a collection of survey form data including the hashed character string. And a vote recovery means.

  According to the present invention, for example, a name identification processing system and a name identification processing method for anonymizing personal information such as patient information and enabling exchange of anonymized information with higher confidentiality in a plurality of medical facilities and the like A server device and a program can be provided.

It is a block diagram of the name collation processing system which concerns on one Embodiment of this invention. It is a detailed block diagram of the Web server of a name collation processing system. It is a detailed block diagram of a name collation server. It is a detailed block diagram of a DB server device. It is a detailed block diagram of an application server. It is a detailed block diagram of an external terminal. It is a figure which shows the table structure of the same name collation processing system. It is a flowchart which shows the process sequence of the same name collation processing system.

  Hereinafter, an embodiment of the present invention will be described with reference to the drawings.

First, in the name collation system of the present invention, patient information (personal information) can be connected and anonymized in hospitals throughout the country, for example, and patient information ( Anonymous processing information) can be tracked (name identification).
Secondly, the name identification system of the present invention cannot encode the same name with the same name (kana), for example, depending on the environment (character code, etc.) of the personal computer, server, etc. The problem is solved.
Thirdly, the system can track (name identification) patient information (anonymized information) by generating IDs and variables that enable patient identification while ensuring, for example, 2 anonymity.

  FIG. 1 shows and describes the configuration of a name identification processing system according to an embodiment of the present invention.

  As shown in the figure, the name identification processing system 1 includes a Web server 2, a name identification server 3, an application server 4, a DB server 7, an information terminal 5 of a survey target facility, and an information terminal 6 of a dialysis society. It consists of The Web server 2 receives the mailing of the survey form, processes the input check entered at the information terminal 5 of the surveyed facility, standardization, gender determination with the first letter of the name, rarity of the last name, etc. Is transmitted to the information terminal 6 of the center. The information terminal 6 at the center uploads information related to the processing results to the Web server 2. From the Web server 2, the survey form data is registered in the DB server 7. This survey form data is stored in the patient table 8.

  In addition to the patient DB 8 described above, the DB server 7 includes a standardized DB 9, a name identification result DB 10, a transfer-in / out DB 11, and a name identification black list DB 12. The standardized DB 9 stores a standardized first and last name result correspondence table, and the correspondence table is transferred to the patient table 8. The name identification server 3 registers the name identification result in the name identification result DB 10. The name identification result DB 10 has a relational relationship with the transfer-in / out DB 11 and the name identification black list DB 12, and the name identification result is transferred to the application server 4. The patient data in the patient DB 8 is also transferred to the application server 4. The name identification server 3 receives the data transfer from the patient DB 8 and registers the result of name identification.

  In such a configuration, when receiving survey form data registration from the Web server 2, the DB server 7 updates the data in the patient DB 8 and transfers the result to the name identification server 3. The DB server 7 updates the data in each DB. The application server 4 generates a survey form based on the name identification result registered in the name identification result DB 10 and the patient information in the patient DB 8. The survey form generated in this way is output to the management department server 13 and managed.

  FIG. 2 shows and describes the detailed configuration of the Web server 2 of the name identification processing system.

  As shown in the figure, the server device 1 includes a control unit 11, a communication unit 12, and a storage unit 13. The control unit 11 functions as a main control unit 14a, a data management unit 14b, and a survey form data registration unit 14c by reading and storing the program 14 in the storage unit 13.

  In such a configuration, the main control unit 14a performs overall control. The data management unit 14b stores and manages the uploaded survey form data in the storage unit 13. The survey form data registration unit 14c registers the survey form data with respect to the DB server 7. That is, the Web server 2 receives and registers data related to the survey form, and registers with the DB server 7.

  FIG. 3 shows and describes the detailed configuration of the name identification server 4 of the name identification processing system.

  As shown in the figure, the name identification server device 4 includes a control unit 31, a communication unit 32, and a storage unit 33. The control part 31 reads the program 34 memorize | stored in the memory | storage part 33, the main control part 31a, the anonymization part 31b, the corresponding | compatible recovery part 31c, the input assistance part 31d, the reliability provision part 31e, the rarity acquisition part of a surname 31f, weighting unit 31g, information acquisition unit 31h, gender discrimination unit 31i, half-width check unit 31b, standardization unit 31c, transfer-in / out trace unit 31l, name identification processing unit 31m, and hashing unit 31m.

  In such a configuration, the main control unit 31a performs overall control. The anonymization unit 31b performs anonymization and realization. The correspondence table recovery unit 31c recovers the correspondence table. The input support unit 31d includes a manual input display control function, a writing error reduction function, and a draft patient survey form printing function. The reliability imparting unit 31e imparts reliability. The rarity acquisition unit 31f for surnames has a nationwide rarity grouping function and a rarity grouping function for each region. The name kanji and kana type number weighting unit 31h weights the kanji for the last name and the reading. The information acquisition unit 31h from the last name has an initial acquisition function from the last name and a function to acquire the number of characters from the last name. The gender discrimination unit 31i performs a gender discrimination function with the last character of the last name (kanji) and gender discrimination with the last letter of the name (kanji). The half-width check unit 31j performs a half-width check of the first and last names (kana) and a half-width check of the first and last names (kana). The standardization unit 31k performs standardization of a surname (kanji), a first name (kanji), a standardization function, a surname (kana), a first name (kana) standardization, and a standardized first name surname complement. The transfer-in / out trace unit 31i performs transfer-in / out determination in only one direction, appropriate transfer-in / out determination, and time-series consistency determination of transfer-in / out. The name identification blacklist 31m has a function of determining a different person based on the blacklist. Then, the hashing unit 31m converts the patient information into a hash value.

  FIG. 4 shows a detailed configuration of the DB server device 1 and will be described.

  As shown in the figure, the DB server device 1 includes a control unit 41, a communication unit 42, a storage unit 43, and a DB group 44. The control unit 41 reads and executes the program 45 in the storage unit 43, thereby functioning as a main control unit 41a, a DB registration unit 41b, a data transfer control unit 41c, a name identification result registration unit 41d, and a standardization unit 41e. The DB group 44 is illustrated as including the various DBs described above with reference to FIG. Since the configuration of each DB constituting the DB group 44 is as described above, it is omitted here.

  In such a configuration, the main control unit 41 a governs overall control of the DB server device 1. The DB registration unit 41 b registers and updates various data in the DB group 44. The data transfer control unit 41c performs data transfer between DBs. The name identification result registration unit 41d registers the name identification result in the name identification result DB 10. The standardization unit 41e performs standardization with reference to the standardization correspondence table.

  The move-in / out trace DB 11 manages the standard of possible movement (hospital transfer). The standard of possible movement (transfer) is used at the time of name identification processing. From the past name identification results, from the patient data that could be identified by the difference in the prefecture of residence, the combination of the resident prefecture code of possible movement (transfer) Define. If it is a combination of different residence prefectures at the time of name identification processing, if it is a combination that exists within the definition of possible movement (transfer), points are added. From the name identification results of the previous year, extract the patient data of the combinations identified by the different prefectures in which they live. Based on the results of name identification more than a certain number, define the combination of resident prefecture codes, and calculate the score from the movement results. In principle, new data will be created every year based on data from the previous year.

  The name identification blacklist stored in the name identification DB 12 is used every year, and is used at the time of name identification processing. Stores the facility code / first and last name information as a blacklist for combinations that have been scored higher than the previous results of name identification but have been manually changed. When writing the name identification result of the name identification process, if the combination exists in the black list, it is forcibly rewritten to another person's status. Based on the result of the previous year's name identification, the facility code of the patient data that was manually changed by another person, first name (kanji) / first name (kana) / date of birth will be entered into the blacklist. In principle, new data will be created every year based on data from the previous year.

  Next, various tables that are finally generated by the server device 13 in the management department based on output data from the application server 4 will be described.

  In the server device 13, a standardization correspondence table, a gender correspondence table, a surname national ranking table, and a surname regional version ranking table are generated.

  The standardization correspondence table is updated every year, and at the time of anonymization processing, a combination of feelings with a large amount of notation is defined as different characters and standardization, and each kanji of the first and last name is replaced from the different characters to the standardized one. As for the creation / update method, Kanji characters with many notations are defined, and they are defined as variants and standardization. The original data is acquired from any external data. Add definitions to the survey form every year.

  The gender name correspondence table defines a kanji and gender correspondence table as the last character of names frequently used by men and women every year. Get the last letter of the first and last name and check if there is no difference with the gender entered in the questionnaire.

  The last name ranking in the last name is hashed and divided into small score groups, and the score is defined for each group. Search from the hash value of the surname (kanji) and record the score according to the group. National ranking data of external data as of year x month x day is acquired, groups are created at regular intervals, and rarity scores are defined.

  In the surname ranking, the top 4000 rankings by prefecture in the surname ranking are hashed and divided into small score groups, and a score is defined for each group. Search from the hash value of the surname (kanji) and record the score according to the group National ranking data of external data as of year x month x day is acquired, groups are created at regular intervals, and rarity scores are defined.

  The kana alphabet correspondence table defines a correspondence table for each of the 50 sounds of “A = A” and “K = K”, acquires the first letter of the first name and the last name (Kana), and acquires the corresponding alphabet. Define a table of alphabets corresponding to every 50 tones of “A = A”, “I = I”, “C = U”.

  In the standardized DB 9 of the DB server device 7, a standardized first and last name result correspondence table is recorded. This standardized first and last name result correspondence table defines a correspondence table of hash values of first and last names (kanji) and standardized first and last names (kanji) based on patient data collected in the past. For a patient whose first and last names are null after standardization, the correspondence table is searched from the entered first and last names (kanji) and the obtained first and last names are complemented. In addition, the patient whose first name and last name after standardization are input matches the correspondence table, and an error is displayed when they do not match. Hash values of surnames (kanji) and standardized surnames of collected patients in the previous year, and hash values of surnames (kanji) and standardized first names are aggregated and defined as a standardized surname result correspondence table. Only when it does not exist in the existing standardized first and last name correspondence table, it is newly added and is not deleted. In addition, when the standardization correspondence table is updated, the hash value of the first and last name after standardization obtained from the latest collected patient data is updated.

  FIG. 5 is a detailed configuration diagram of the application server 4.

As shown in the figure, the application server 4 includes a control unit 51 that performs overall control. The control unit 51 includes a communication unit 52. The display unit 53 and the storage unit 54 are connected.
The control unit 51 functions as the main control unit 51a, the survey form creation unit 51b, and the display control unit 51c by reading and executing the program 55 stored in the storage unit 54. The communication unit 52 is a communication interface that communicates with the DB server 7 and the like. As the display unit 53, various types such as an LCD display can be adopted.

  In such a configuration, the main control unit 51a comprehensively controls various controls such as communication with an external server. The survey form creation unit 51 b creates a survey form based on the data transferred from the DB server 7. The display control unit 51 c controls various displays by the display unit 53. The generated survey form is sent to the server device 13.

  FIG. 6 shows and describes the detailed configuration of the information terminals 5 and 6 outside the system.

  As shown in the figure, the information terminals 5 and 6 include a control unit 61 that performs overall control, and includes a communication unit 62, an operation unit 63, a display unit 64, and a storage unit 65. Yes. The control unit 61 functions as a main control unit 61a, a survey form management unit 61b, a transmission / reception control unit 61c, and a display control unit 61d by reading and executing the program 26 in the storage unit 65. The communication unit 62 is a communication interface that communicates with an external server. The operation unit 63 is various operation members such as a mouse and a keyboard. The display unit 64 is a display device such as an LCD display. The operation unit 63 and the display unit 64 can be integrally configured as a touch panel.

  In such a configuration, the main control unit 61a is responsible for overall control. The communication unit 62 is a communication interface that communicates with an external server device or the like. The operation unit 63 is various operation members such as a mouse and a keyboard. The display unit 64 is a display device such as an LCD display. The storage unit 65 stores the program 26 executed by the control unit 61. And the control part 61 functions as the main control part 61a, the survey form management part 61b, the transmission / reception control part 61c, and the display control part 61d by reading and executing the program 26 of the memory | storage part 65. FIG.

  The main control unit 61a is responsible for overall control. The survey form management unit 61b generates a survey form based on the survey form data sent from the DB server device 7. The transmission / reception control unit 61 c controls transmission / reception of various data via the communication unit 62. The display control unit 61d controls display by the display unit 64. The generated survey form is output to the server device 13 in a unified data format. This questionnaire includes a standardization correspondence table, a gender name correspondence table, a surname national ranking table, and a surname regional version ranking.

  Here, FIG. 7 shows a table configuration of the same name identification processing system and will be described.

  As shown in the figure, a possible movement master stores a prefecture code. The score management master stores various data in association with the score ID. The scoring (current year) result master stores the current year name ID (FK) and the score ID. The scoring (year-round) result stores a year-round name identification result ID (FK), a score ID (FK), and a calculated score. The name identification result (current year) table stores a current year name identification result ID and a patient code name identification result. The name identification result (year-round) stores a year-round name identification result ID, a collection year, and a set of patient codes. The patient (anonymous) table stores a management serial number (new), a patient code, a collection year, a prefecture code, and an aggregation destination patient code. The standardized first and last name master stores a hash value before standardization and a hash value after standardization. Then, the transfer target patient table stores a management serial number (old), a final patient code, a full name (hash), youth date, and a transferred flag. The above is just a series of various masters.

  Hereinafter, with reference to the flowchart of FIG. 8, a processing procedure of the name identification processing system according to the present embodiment will be described and described.

  When the DB server device 7 receives the registration of survey form data from the Web server 2 and the survey form data is transferred to the name collation server 3 (S1), the name collation server device 3 first performs input check processing of first and last names (kana). Is performed (S2). In this half-width check process, a surname (kana) and a first name (kana) are acquired and replaced by an input conversion function.

  Subsequently, standardization processing is executed (S3). Specifically, in this standardization process, a surname (kanji) and a name (kanji) are acquired. 1 Match and convert with standardized correspondence table, acquire first and last names of all patients, match with and convert to Kana standardized correspondence table.

  Next, gender discrimination is performed with the first letter of the name (S4). In this process, one character of the name (kanji) is acquired, and No. 2 Match with gender name correspondence table, match with gender name correspondence table, perform gender discrimination, and perform error when gender and discrimination result do not match.

  Subsequently, the rarity of the surname is determined (S5). Here, the ranking of surnames in the national version and the ranking in each region are determined. More specifically, a surname (kanji) is acquired, a resident prefecture code is acquired, and it is determined whether a prefecture-specific ranking exists. And when it exists, it is No. Collate with the 3 surname national rankings and get a rarity score based on the ranking. On the other hand, if it does not exist, no. Collate with the 4 surname regional version ranking, and acquire a rarity score based on the ranking as above.

  Next, information is acquired from the first and last names (S6). More specifically, the surname (kana) and first name (kana) are acquired. Match the 5-kana alphabet correspondence table and get the number of characters. Then, a surname (kanji) and a first name (kanji) are acquired, and the number of characters is acquired.

  Subsequently, hashing is performed (S7). In this hashing, patient information is converted into a random fixed-length alphanumeric sequence using a hash function so that the patient cannot be identified outside the facility. For example, a process of converting patient information into a hash value can be performed, and an electronic survey form using a spreadsheet can be created.

  Thus, the survey form data is collected (S8), and the survey form data is uploaded to the Web server 2 (S9). The uploaded survey form data is registered in the DB server device 7, and first, standardized first name surname is complemented (S10). In this standardized first name surname complementation, the surname (kanji) and first name (kanji) of the patient whose standardized first name is Null are acquired. 6 Match with the standardized first and last name correspondence table.

  Subsequently, the transfer-in / out trace process is performed (S9). In this transfer-in / out determination process, the resident prefecture code is acquired, and No. 7 It matches with the standard of possible movement, and the score is added as an appropriate transfer-in / out determination.

  Next, a scoring process is performed (S10). In this scoring process, scoring is performed on the data after the transfer-in / out trace, and the result is stored in the scoring (current year) result master. The scoring (current year) result master stores the current year name ID (FK) and the score ID. The scoring (year-round) result stores a year-round name identification result ID (FK), a score ID (FK), and a calculated score.

  Subsequently, another person determination is performed based on the name identification black list (S11). In the other person determination by this name identification black list, No. The combination of another person existing in the 8-person blacklist is collated, and when it exists in the blacklist, the name-matching result is determined as another person.

  In this way, name identification is performed by the name identification server 3 (S12), the name identification result is stored in the name identification result table, and data is transmitted to the application server 4. In this way, a series of processes up to name identification is completed.

  Although one embodiment of the present invention has been described above, the present invention is not limited to this, and it is needless to say that various improvements and changes can be made without departing from the spirit of the present invention.

  For example, the size of the two rollers and the angle between the axes are not limited to those described above, and can be designed to have various sizes according to the user.

  DESCRIPTION OF SYMBOLS 1 ... Name identification processing system, 2 ... Web server, 3 ... Name identification server, 4 ... Application server, 5 ... Information terminal, 6 ... Information terminal, 7 ... DB server, 8 ... Patient DB8, 9 ... Standardization DB9, 10 ... Name identification DB , 11 ... transfer-in / out DB, 12 ... name identification DB, 13 ... server device.

Claims (5)

Survey form input means for receiving input of a survey form in which patient information is recorded;
Half-width confirmation means for performing input check of first and last name of the patient information in the survey form,
Storage means for storing at least a standardization correspondence table for replacing each kanji of the first and last name from a different character to a standardized kanji;
Standardization means for performing standardization by matching and converting the standardized correspondence table for each kanji of the first and last names,
A gender discrimination means for performing gender judgment with the last letter of the last name,
A rarity determination means for determining the rarity of a surname among the first and last names
Information acquisition means for acquiring information from the first and last names;
A hashing means for converting the patient information into a hashed character string;
Survey form collecting means for collecting the survey form data consisting of the hashed character string;
Name processing system with The storage means further stores a gender name correspondence table,
The name collation processing system according to claim 1, wherein the gender discrimination means performs gender discrimination for one character of a kanji character associated with a first name in a surname by referring to the gender name correspondence table. The name identification processing system according to claim 1, wherein the rarity determination unit acquires a resident prefecture code and, when there is a ranking by prefecture, refers to a surname ranking table and acquires a score relating to rarity. . Receiving an input of a questionnaire that records patient information;
Performing a half-width check of the first and last name of the patient information in the survey form;
A step of storing at least a standardization correspondence table for replacing each kanji of the first and last name from a different character to a standardized kanji;
Performing standardization by matching and converting the standardized correspondence table for each kanji of the first and last name;
Performing gender determination with the last letter of the surname,
Determining the rarity of surnames among the surnames;
Obtaining information from the first and last names;
Converting the patient information into a hashed character string;
Collecting the survey form data consisting of the hashed character string;
Name identification processing method with Server device
Receiving an input of a questionnaire that records patient information;
Performing a half-width check of the first and last name of the patient information in the survey form;
A step of storing at least a standardization correspondence table for replacing each kanji of the first and last name from a different character to a standardized kanji;
Performing standardization by matching and converting the standardized correspondence table for each kanji of the first and last name;
Performing gender determination with the last letter of the surname,
Determining the rarity of surnames among the surnames;
Obtaining information from the first and last names;
Converting the patient information into a hashed character string;
Collecting the survey form data consisting of the hashed character string;
A program for running.