JP6541353B2 - INFORMATION PROCESSING SYSTEM, INFORMATION PROCESSING METHOD, AND PROGRAM - Google Patents

Description

  The disclosed technology relates to an information processing system, an information processing method, and a program.

  Information transmission using forms such as slips or medical records is performed in banks or hospitals. When entering information in this form, there may be a case where information appearing continuously is replaced by a repetition symbol indicating repetition of information such as “〃” in order to reduce the workload. However, when a symbol such as "〃" is read by a scanner, even if it is tried to extract the contents of the form from the scan result, the content can not be extracted because "内容" itself has no meaning of words. In order to solve this problem, in Patent Document 1, the display means simultaneously displays the area including the repetition symbol and the area located above the area, and the repetition symbol refers to the area where the person is located above the area. It is disclosed to input the information referred to.

Japanese Patent Laid-Open No. 3-54672

  However, in the prior art, it is assumed that there is only one type of repetition symbol, and it is considered in the case where there are a plurality of repetition symbols (for example, “↑”, “↑”, etc.) referring to information at different positions. Not.

  The technology disclosed herein has been made in view of such circumstances, and one object is to replace repetitive symbols with appropriate information according to the type of repetitive symbols.

  It is to be noted that the operation and effect are not limited to the above object but are derived from the respective configurations shown in the embodiments for carrying out the invention described later, and it is another object of the present invention to exert the operation and effect not obtained by the prior art. It can be positioned as one.

  The disclosed information processing system comprises: extraction means for extracting a symbol or character meaning repetition of information from a document image which is an imaged document; determination means for determining the type of the extracted symbol or character; Determining means for determining information to which the symbol or character refers based on the determination result by the determining means, and replacing means for replacing the symbol or character with information to which the symbol or character determined by the determining means refers And

  According to the disclosed technology, it is possible to replace repetitive symbols with appropriate information depending on the type of repetitive symbols.

(A) is a block diagram which shows an example of the hardware constitutions in 1st Embodiment. (B) is a figure which shows typically an example of the function of CPU in 1st Embodiment. It is a figure which shows the example of the document of extraction object in 1st Embodiment. It is a figure which shows the example of table structure extraction in 1st Embodiment. It is a figure which shows the example of the reference destination address of the repetition symbol in the table structure in 1st Embodiment. It is a figure which shows the example of the reference destination candidate table of the repetition symbol in 1st Embodiment. It is a figure which shows the example of the drug prescription specification table in 1st Embodiment. It is a figure which shows the example of the chemical | medical agent name normalization table in 1st Embodiment. It is a figure which shows the example of the repetition symbol reconstruction result in 1st Embodiment. (A)-(c) is a figure which shows the example of the screen display of extraction result display * correction in 1st Embodiment. It is a figure which shows the example of the extraction process result output format in 1st Embodiment. It is a flowchart which shows an example of the process in 1st Embodiment. It is a flowchart which shows an example of the repetition symbol reconstruction process in 1st Embodiment.

  Hereinafter, preferred embodiments to which the present invention is applied will be described in detail with reference to the attached drawings. However, the components described in this embodiment are merely examples, and the scope of the technology of the present invention is determined by the scope of the claims, and is limited by the following individual embodiments. is not.

First Embodiment
FIG. 1A is a diagram showing an example of the configuration of an information processing system to which the present embodiment is applied.

  The information processing system illustrated in FIG. 1 includes a CPU 101, a ROM 102, a RAM 103, a document input unit 104, a display unit 105, and a communication unit 106.

  The CPU 101 controls the system by executing a control program stored in the ROM 102. That is, when the CPU 101 executes the control program, the information processing system functions as various means for executing each process shown in the flowchart to be described later. The functions realized by execution of the control program by the CPU 101 will be described later with reference to FIG.

  The ROM 102 stores control programs and the like executed by the CPU 101.

  The RAM 103 temporarily stores various data from each component. Further, the CPU 101 loads the control program read from the ROM 102 on the RAM 103 and executes the program.

  The document input unit 104 includes an optical device, an imaging device, and the like, and captures an image of an object in accordance with an instruction from the CPU 101. For example, the document input unit 104 captures a document such as a form to obtain a document image obtained by imaging the document. For example, the document input unit 104 is configured by a scanner. The document input unit 104 may be configured to acquire a document image obtained by a scanner provided outside the information processing system via a wired or wireless connection.

  The display unit 105 is configured of a liquid crystal panel or the like, and displays a document image acquired by the document input unit 104 and various information such as a result processed by the CPU 101.

  The communication unit 106 is a network interface (NIC), and is an interface that enables communication with other devices on the network via a LAN cable or the like.

  A bus 107 is a bus used to mutually transmit digital signals between units included in the information processing system.

  FIG. 1B is a diagram showing an example of the function of the CPU in the present embodiment. The CPU 101 executes a control program stored in the RAM 103 to function as an extraction unit 11, a determination unit 12, a determination unit 13, a substitution unit 14, and a display control unit 15, as shown in FIG. 1B. The CPU 101 and the RAM 103 included in the information processing system may be one or more. That is, when at least one or more processing units (CPU) and at least one storage unit (RAM) are connected, and at least one or more processing units execute a program stored in at least one or more storage units. The CPU 101 functions as each means described above. Although each means shown in FIG. 1B is realized by software using the CPU 101, part or all of the processing may be realized by hardware such as an electronic circuit.

  The extraction unit 11 extracts the repetition symbol (repetition symbol) from the document image acquired by the document input unit 104. Here, the repetition symbol includes a symbol or a character that means repetition of information, and is used to avoid repeated description of similar words when information such as the same word repeatedly appears. Repetitive symbols include characters, but in the present specification, for simplicity, they are simply described as repetitive symbols without describing repetitive symbols and characters. The repetition symbol refers to information such as a predetermined word, since the repetition symbol is used to avoid describing the same word repeatedly. The extraction means 11 corresponds to an example of an extraction means for extracting a symbol or a character meaning repetition of information from a document image which is an imaged document. The repetition symbol includes nono-dots, characters, and other symbols or characters (words).

  The determination means 12 determines the type of repetition symbol. That is, the determination means 12 corresponds to an example of the determination means for determining the type of the extracted symbol or character.

  The determination means 13 determines the information to which the repetition symbol refers in the document based on the determination result of the determination means 12. That is, the determination means 13 corresponds to an example of the determination means for determining the information to which the symbol or the character refers based on the determination result by the determination means. More specifically, the determination means determines the information to which the repetition symbol refers based on the table in which the type of repetition symbol is associated with the position of the information to which the repetition symbol refers and the determination result of the determination means 12.

  The substitution means 14 substitutes the repetition symbol with the information determined by the determination means. That is, the substitution means 14 corresponds to an example of substitution means for replacing a symbol or a character with information to which the symbol or the character determined by the determination means refers. Here, the substitution means 14 substitutes the repetition symbol with the information determined by the determination means 13, but the correspondence between the substituted information and the information of the reference destination may be maintained.

  The display control unit 15 causes the display unit 105 to display various types of information. For example, the display control unit 15 causes the display unit 105 to display the information replaced by the replacement unit 14. That is, the display control means 15 corresponds to an example of the display control means for causing the display means to display information to which the symbol or the character replaced by the replacement means refers. When the display control unit 15 causes the display unit 105 to display the information to which the repetition symbol determined by the determination unit 13 refers, the display control unit 15 may not display the repetition symbol on the display unit 105. Further, the display control means 15 may cause the display unit 105 to display the information to which the repetition symbol determined by the determination means 13 refers, together with the repetition symbol. In this case, the operator can understand at a glance which repetitive symbols are replaced with what information.

  Next, an example of a document image acquired by the document input unit 104 will be described with reference to FIG. FIG. 2 is an example of a document image to be subjected to document information extraction processing in the first embodiment, and describes a medicine list brought by a patient at the time of admission to a hospital. A document 201 is composed of document information 202 to 208. Reference numeral 202 denotes a title of the document 201, reference numeral 203 denotes a creation date of the document 201, and reference numeral 204 denotes a character string indicating the name of a subject patient described in the document 201. Reference numerals 205 to 208 summarize the brought-in drug in a tabular form, 205 is a note on the brought-in drug, 206 is a drug name, and 207 is a character string representing the drug dosage form. As the dosage form 207, terms representing external features of medicine such as "tablet", "tablet", "tablet", and "fine particle" appear. Reference numeral 208 is a character string representing a prescribed volume of medicine. As shown in FIG. 2, symbols such as “↑” and “出現” may also appear in the tabulated 205 to 208. Here, “〃” and “↑” are examples of repetition symbols.

  FIG. 3 shows an example of the result of extracting information as a table structure from the tabular input data shown in 205 to 208 of the document image shown in FIG. 2 in the embodiment. The processing of the same type of table structure extraction can be realized, for example, using the method disclosed in Japanese Patent Application Laid-Open No. 2006-99480.

  The information shown in FIG. 3 is configured as two-dimensional array data, generated by the extraction unit 11 and stored in the RAM 103. T (e (m, n)) indicated by 301 is a generalization of element data e of the table structure T, and element data of m rows and n columns is denoted as e (m, n). For example, element data represented by e (2, 1) is "anticoagulant". Data in the first column indicated by 302 is data relating to "attention" indicated by e (1, 1). Data in the second column indicated by 303 is data relating to the “drug name” indicated by e (1, 2). The data in the third column indicated by 304 is data on the “dosage form” indicated by e (1, 3). The data in the fourth column indicated by 305 is data relating to the “standard” indicated by e (1, 4). e (3, 3), e (4, 2), e (4, 3) stores “〃” as element data, and e (6, 2) stores “↑” as element data. There is. In addition, e (3, 1), e (4, 1), e (5, 1), e (6, 1), e (6, 3) have no extracted character string and no element data is blank. It has become.

  FIG. 4 is address information showing the relative positional relationship between each element data CP and other element data in the table structure shown in FIG. 3 in the first embodiment. For example, when CP is element data e (m, n), the element data indicated by the address R2 is e (m-1, n), and the element data indicated by the address R4 is e (m, n-1). Become.

  FIG. 5 is an example of a table showing candidates for reference destinations of repetition symbols in the first embodiment. The table is stored in the ROM 102 or the RAM 103. Here, 501 shown in FIG. 5 may be a single character or character string (generally referred to as a repetition symbol) that expresses being identical to adjacent element data in order to avoid repeated notation of the same meaning and content. Store).

  Reference numerals 502 and 503 denote reference destination address information (reference destination position information) indicating which element data indicates that the repetition symbol 501 indicates the same. Up to two pieces of reference address information can be set for one repetition character, and the reference address 502 has a priority relative to the reference address 503. For example, the reference destination R2 is applied prior to the reference destination R4 to the repetitive character “〃”. If there is no reference destination 2, "-" is stored. In the present embodiment, the reference destination address information is at most two per repetitive character, but the present invention is not limited to this, and three or more pieces of reference address information are associated with one repetitive character. You may do it.

  FIG. 6 shows an example of a drug prescription standard table in the first embodiment. The drug prescription standard table is a database storing drug information described in a document issued as an attached document by a pharmaceutical company at the time of being marketed as a licensed drug. The database is stored in the ROM 102 or the RAM 103. The drug prescription standard table shown in FIG. 6 corresponds to an example of a drug prescription standard.

  Although the drug name may take a plurality of notations such as a general drug name, an ingredient name, and a trade name of a pharmaceutical company, 601 stores a drug name normalization notation described later in FIG. Reference numeral 602 denotes a medicine usage category, which stores "internal medicine" or "external medicine". 603 stores information on dosage forms, such as "tablet", "granule", "fine particle", "sticking", "injection", "eye drop", etc., which are classifications of appearance shape features of medicines. 604 and 605 together represent the standard volume of the drug, and 605 stores the standard volume unit 604 and the standard volume value. 606 and 607 represent the usage standard volume of the drug, and 607 stores the standard dosage unit 606 and the value of the usage volume.

  FIG. 7 shows a drug name normalization table in the first embodiment. As mentioned above, in a document referring to a drug name, it is also common to use a trade name of a pharmaceutical company as a drug designation. Therefore, even the same medicine may have different names, so the names shown in FIG. 7 are used to unify (normalize) the names. The table is stored in the ROM 102 or the RAM 103.

  A field 701 stores a plurality of drug name variants, such as a drug component name, a general trade name, and a trade name, meaning the same drug. “Alephalin”, “Ormetech”, “Daiart”, “Nerbon”, “Benzarine”, “Eurodin”, “Warrin” used for the example of 701 are trademarks. The variant data is encoded in UTF-8 according to the ISO / IEC 10646 Character Set Standard (Unicode), and arranged in ascending order of code.

  702 stores a drug name normalization notation that normalizes drug names stored in 701 as a single notation. For example, since the drug name "azosemide" is normalized as the notation "Dieart", it can be referred to in 601 of the drug prescription standard table shown in FIG. In addition, it is possible to select and store the normalization notation by selecting any one of the different notation groups which mean the same medicine.

  FIG. 8 shows an example of the result of the recovery process of the repetition symbol in the first embodiment. When the table structure extraction result of FIG. 3 extracted from the document image shown in FIG. 2 includes the repetition symbol, the determination unit 11 and the determination unit 13 determine which character string in the document image the repetition symbol repeats. Do. Then, the substitution means 14 replaces the repetition symbol with the repetition character string of the reference destination and updates the table structure extraction result of FIG.

  At the end of element data e (3, 2) indicated by 801, "! (4, 2)" is substituted means. At the end of element data e (4, 2) indicated by 802, the symbol "! (3, 2)" is substituted means It is granted by. The symbol “!” Makes it possible to identify that the element data may contain an error in the content of the element data extracted with respect to the input document of FIG. Furthermore, (m, n) following “!” Stores the information determined as the reference destination of the repetition symbol. In the example in FIG. 8, not only e (4, 2) in which the repetition symbol is present, but also information e (3, 2) to which the repetition symbol refers is displayed information indicating the reference destination. That is, the correlation between the repetition symbol and the reference destination of the repetition symbol can be known. In e (3, 2), information indicating the position of the repetition symbol (4, 2) may not be displayed, and it is clarified that (4, 2) indicates the position of the repetition symbol. In order to reduce the display size, the display mode may be different from (3, 2).

  Further, a symbol string “!! (5, 3)” is given by the substitution unit 14 at the end of the element data e (6, 3) indicated by 803. The symbol string “!!” makes it possible to identify that element data that has been blank without a repetition symbol explicitly in the table structure extraction result of FIG. 3 is determined to be element data repetition. Furthermore, (m, n) following “!!” stores the information determined as the reference destination of the repetitive expression. As will be described later in the process flowcharts of FIGS. 11 to 12, when the drug prescription standard shown in FIG. 6 and the recognition result by the information processing system shown in FIG. 8 do not match, the symbol “!” Is added to the end. That is, the substitution means 14 compares the recognition result by the information processing system with the drug prescription standard shown in FIG. Then, the substitution means 14 adds the symbol “!” To the end of the unmatched element data when the recognition result by the information processing system does not match the drug name, the drug system, etc. shown in FIG. 6 as a result of comparison. Furthermore, the end symbol string “!!” is added when it is determined that element data that is blank without a repetition symbol explicitly is determined to be a repetition of the element data. The symbol added to the element data may be a symbol or a character other than the exclamation mark.

  FIG. 9 is a view showing an example of an operation screen for display of extraction results of document information in the first embodiment and correction editing by the user of the extraction results.

  When the result of the restoration process of the repetition symbol shown in FIG. 8 is obtained, the display control means 15 causes the display unit 105 to display the display screen of FIG. 9 (a). That is, the display control means 15 presents the display screen of FIG. 9A to the user. In FIG. 9A, 901 and 902 are display examples corresponding to 801 and 802 in FIG. 8, respectively. As can be seen from the comparison with the input document image of FIG. 2, although 901 must be extracted and displayed as "Dieart", it is displayed as "Nuiart". This may be caused by erroneous recognition in the character recognition process in the extraction process from the document image.

  This situation can be identified by whether the symbol "!" Is added to the end of the element data in the extraction result of FIG. That is, when the symbol “!” Is added to the end, the display control unit 15 causes the display unit 105 to display the element data and, as indicated by 903, displays a warning message for content confirmation / correction awakening to the user. Let

  In 903, an example in which a warning message is displayed by a balloon is shown, but other methods may be used such as changing the display color of display data, blinking, etc. without departing from the scope for calling for content confirmation.

  A warning message 904 is displayed when the symbol string "!!" is added to the end of the element data in the extraction result of FIG. A message 904 prompts the user to confirm the content of the result of determining the situation and extracting the element data as the processing object of the iterative restoration even though there is no information indicating an explicit repetition in the input document. The display control unit 15 can determine whether to display the message shown in 904 based on whether or not the symbol “!!” is added to the end of the element data.

  FIG. 9B shows a display example in which the user changes the extraction result "Nuiart" to "Daiart" by editing means (not shown). Here, the editing means is configured by a known means such as designating a target to be edited by pointing operation such as mouse or cursor movement, screen touch, etc., and performing data rewriting by a keyboard or voice utterance. Note that the characters highlighted in reverse white at 905 are corrected and edited characters.

  In FIG. 9C, after completion of editing in FIG. 9B, the element data e (3, 2) in the extraction result in FIG. Replaced with the correction result. Substantially at the same time, the substitution means 14 also replaces the element data of the repetition character 906 which refers to 905 with the same data as 905, and becomes the displays 907, 908.

  Although the same editing result is applied to the reference source 906 when the reference destination 905 is edited in the above description, the present invention is not limited to this. For example, when the reference source 906 is edited in the reverse order, the element data of the reference destination 905 may be similarly updated and displayed.

  FIG. 10 shows an example of the output format of the document information extraction result in the first embodiment. The display control means 15 outputs the two-dimensional array data of the extraction result shown in FIG. 8 as a format in a description format in a structured language. In the example in FIG. 10, the elements of the extraction data are defined by the structured language XML (extensible Markup Language). The <prescription> element is a container element that describes drug prescription information. There is only one <prescription> element paired with the document input in FIG. 2 and has multiple <medicine> elements. The <medicine> element is an element describing a set of extracted medicine information, and corresponds to one row of data in the table of the extraction result of FIG. The <medicine> element can be identified by the id attribute, and although not shown in FIG. 10, the relationship between the reference source and reference destination of the repetition character can also be implemented by this attribute description. The <name> attribute is the extracted drug name, and <type> is an element describing the dosage form of the drug. The <dosage> element is an element that describes a prescription volume, and is composed of a <quantitiy> element that describes a quantity and a <unit> that describes a unit of quantity. The <notice> element is an element that describes a disclaimer, and can be omitted if there is no disclaimer.

  FIG. 11 is a flowchart showing an example of processing in the first embodiment.

  When information extraction processing by the information processing system is started, first, in step 1101, the document input unit 104 records a document image to be extracted shown in FIG.

  In step 1102, the extraction unit 11 extracts the information of the table structure shown in FIG. 3 by using the unit of Japanese Patent Laid-Open No. 2006-99480, for example, and stores the document image in the RAM 103.

  At step 1103, the extraction means 11 detects the repetition symbol (repeated expression) appearing in the extracted table structure. Furthermore, the determination means 12, the determination means 13 and the substitution means 14 determine the reference destination of the repetition symbol and restore the repetition symbol to the information of the reference destination. That is, the substitution means 14 substitutes the repetition symbol with the information to which the repetition symbol refers. The details of this process will be described later with reference to FIG.

  In step 1104, the display control means 15 displays the information in which the repetition symbol is replaced with the information of the reference destination on the display unit 105 as shown in FIG. 9A. Note that if the symbol "!" Is added to the end of element data as shown in Fig. 8, a warning message indicating the element is "Please check and correct it." indicate. Further, when the symbol string “!!” is attached to the end of the element data of FIG. 8, a warning message “automatically inserted and confirm” indicating the element is displayed. The displayed message is an example and may be another message.

  In step 1105, the user performs a correction editing operation on display data. The user uses a user interface device such as a keyboard, a mouse or a display screen touch to point to a portion requiring correction and editing, and performs data deletion / insertion / change. The information is updated by the editing operation as shown in FIG. At this time, if correction is made to the data to which the repetition symbol subjected to the editing operation refers, the element data in which the repetition symbol is present is similarly corrected as described with reference to FIG. 9C. Be done.

  The display control unit 15 converts the information updated in the above step in step 1106 into the output format of FIG. 10 and causes the display unit 105 to display the information. Alternatively, the display control unit 15 records in an external storage device (not shown). Further, the display control unit 15 may transmit to another system (not shown) via the communication unit 106.

  12 is a flowchart showing in detail 1103 of FIG. 11 in the first embodiment.

  The extraction means 11 acquires the information of the table structure shown in FIG. 3 extracted in step 1102 of FIG. 11 in step 1201.

  At step 1202, the extraction means 11 sets an initial value (m = 1) for scanning element data in the row direction with respect to the information of the table structure of FIG.

  In step 1203, the extraction unit 11 sets an initial value (n = 1) for scanning element data in the column direction with respect to the information of the table structure of FIG. 3.

  At step 1204, the extraction means 11 acquires element data included in the table structure. Element data e (1, 1) is acquired as an initial value.

  At step 1205, the extraction means 11 determines whether the acquired element data is blank (no element). If it is blank, the process proceeds to step 1206. If it is not blank, the process proceeds to step 1207.

  If the element data is blank at step 1206, the extraction means 11 determines whether this blank is to be interpreted as a repetition symbol. This determination is made based on the data attribute described in the first row of the column data to which the element data belongs. That is, element data of e (1, n) is acquired, and the acquired element data is collated with the item attribute names 601 to 607 of the medicine prescription standard table of FIG. 6 as the column attribute. If the element data of e (1, n) matches the item attribute name of the drug prescription standard table, the extraction means 11 determines that the repetition symbol may be omitted. Then, the extraction means 11 sets a predetermined repetition symbol as blank element data as new element data. Although the predetermined repetition symbol is “記号” in this embodiment, the predetermined value may be changed according to the conditions. In this embodiment, it is judged that the “dosage form” shown in FIG. 3 matches the item 603 of the drug prescription standard table shown in FIG. Complement with "〃".

  If none of the item attribute names 601 to 607 matches, the extraction unit 11 determines that the omission possibility of the repetition symbol is none. For example, e (6,1) can omit the repetition symbol because the element data e (6,1) has a space where e (1,1) is "attention" and "attention" does not match the item attribute name. Is determined to be none. On the other hand, since e (1, 3) is a "dosage form" and the blank of e (6, 3) matches the item attribute name 603 of the drug prescription standard table, the omission possibility of the repetition symbol is determined to be present. , "E" is set as element data of e (6, 3). Here, in step 1223, it is determined in step 1206 whether or not complementation has been performed. If complementation has been performed, the process proceeds to step S1208. On the other hand, if the interpolation can not be performed, the process proceeds to step S1216.

  At step 1207, the extraction means 11 determines whether the element data is a repetition symbol. Specifically, it is determined by searching whether or not the repetition symbol / character string 501 of FIG. 5 matching the element data is present. If there is no repetition symbol / string matching the string 501 in the element data to be processed, the process proceeds to step 1216. If the repetition symbol / string matching the string 501 exists in the element data to be processed, the process proceeds to step 1208. It is to be noted that, in step 1207, extraction of the repetition symbol by the extraction unit 11 and determination of the type of repetition symbol by the determination unit 12 are performed by determining which repetition symbol of the character string 501 corresponds.

  The determination means 13 acquires the reference destination information 502 of the reference destination 1 corresponding to the repetition symbol and character string matched in step 1208. Specifically, the determination means 13 determines the reference destination 1 using the repetition symbol extracted and determined in step S1207 and the table shown in FIG.

  At step 1209, the substitution means 15 specifies element data of the reference destination indicated by the reference destination 1 and acquires element data. For example, in the case of the repetition symbol “〃” of e (3, 3), the reference destination 1 of the repetition symbol “〃” is acquired from FIG. 5 as “R2”, and the reference destination indicated by “R2” is FIG. If 3, 3) is the current location CP, it can be seen that the columns are the same and the rows are −1 (e (2, 3)). The substitution means 15 acquires element data "lock" of e (2, 3).

  In step 1210, the substitution means 15 determines whether or not the reference destination object such as "lock" has been acquired. It is determined whether or not acquisition has been possible if, in the case where some element data or an iteration symbol has been acquired, the object of the reference destination has been acquired. On the other hand, if the reference destination is blank, it is determined that the reference destination object could not be acquired. If it can be acquired, the processing proceeds to step 1211. If it can not be acquired, the processing proceeds to step 1213.

  If the substitution unit 15 can not acquire element data due to the reference destination 1, the reference destination 2 is acquired in step 1213. Similarly to 1208, the information 503 in FIG. 5 which is the information on the reference destination 2 is acquired, and the element data indicated by the reference destination is acquired as in step 1209. If the value of reference destination 2 is "-", acquisition is not possible.

  In step 1214, the substitution unit 15 determines whether the reference destination element data of the reference destination 2 can be acquired by the same determination method as that in step 1210. If the element data of the reference destination can be acquired, the process advances to step 1211; if the element data can not be acquired, the process advances to step 1215.

  In step 1215, the substitution means 15 determines that the reference destination of the repetition symbol can not be identified, and sets "NIL" indicating the meaning of the content indefinite character string as element data.

  In step 1211, the substitution means 15 determines whether or not the element data acquired from the reference destination of the repetition symbol is the repetition symbol. If it is an iteration symbol, the process branches to step 1208 to recursively determine the reference destination, and if not, it branches to step 1212.

  In step 1212, the replacement unit 15 replaces the element data acquired in step 1204 with the element data of the reference destination acquired in step 1209 or step 1213.

  In step 1216, the substitution means 15 determines whether or not data in the column direction remains, and if not, the process proceeds to step 1217, and if unprocessed column data remains, the process proceeds to step 1218.

  At step 1218, the index n indicating the column data is counted up to process the next column data element, and the process loops to step 1204.

  When repetition of repetition processing (replacement processing) for one row data is completed, a set of drug prescription information extracted from predetermined row data of the information shown in FIG. 8 in step 1217 and the drug prescription in FIG. Check content consistency with the standard table. The data sets e (m, 2), e (m, 3) and e (m, 4) of the m rows in FIG. 8 are the 601 drug name in FIG. The substitution means 14 determines whether or not they match. However, by using a conversion table (not shown) for matching determination regarding 603 "dosage forms", it is determined whether "tablets" in FIG. 8 are converted into "tablets" and match with "tablets" in FIG. Do. Further, the “standard” in FIG. 8 is separated into a numerical part and a unit part so as to be collated with 604 and 605, and collated with the drug prescription standard in FIG.

  In step 1219, the substitution means 14 determines whether or not the drug prescription standard and the data in which the repetition symbol shown in FIG. 8 is substituted coincide. If they match, the process proceeds to step 1220. If not, the process proceeds to step 1221.

  In step 1221, if the result of replacing the repetition symbol with the reference information according to the procedure does not match the drug prescription standard, the substitution means appends the symbol “!” To the end of the element data not matching, and the symbol “! At the end of ", store reference destination information in the format of (m, n).

  In step 1220, it is determined whether all the line data in FIG. 3 has been processed, and if it is ended, the restoration processing of the repetitive expression of all data is ended, and if there is unprocessed line data, step 1222 is performed. move on.

  In order to process the element data of the next row in step 1222, the subscript m indicating the row data is counted up, the subscript n indicating the column data is initialized to 1, and the process loops to step 1204.

  As described above, it is possible to extract the contents of the repetitive expression as data, even for a document in which a character string or a specific symbol such as “same symbol” means repetition of the descriptive content. Further, according to the present embodiment, even when there are a plurality of types of repetition symbols different in the position of the information to be referred to, it is possible to accurately specify the information of the reference destination and replace the repetition symbol with the information of the reference destination. It becomes possible. Therefore, it is possible to save the operator the trouble of manually replacing the repetition symbol. Furthermore, it is possible to prevent recognition by referring to the standard of drug prescription including information on dosage form and amount of medicine and displaying a warning.

(Other embodiments)
In the above embodiment, the extraction of information from the input document for extracting drug prescription information is used as an example, but the extraction of information from the document including the repetitive expression, ie, the purpose of determining the reference destination of the repetitive expression is impaired It is also applicable to other document cases that do not occur.

  In the embodiment of the present invention, although the repetition symbol and the reference destination of the repetition symbol have been described as predetermined, the repetition symbol reference destination candidate table of FIG. 5 may be defined by the user of the document information extraction system.

  The object of the present invention can also be achieved by the following configuration. That is, a recording medium (or storage medium) storing a program code (computer program) of software for realizing the functions of the above-described embodiments is supplied to a system or apparatus. Such a storage medium is a computer readable storage medium. Then, the computer (or CPU or MPU) of the system or apparatus reads out and executes the program code stored in the storage medium. In this case, the program code itself read out from the storage medium implements the functions of the above-described embodiments, and the storage medium storing the program code constitutes the present invention.

  In addition, an operating system (OS) or the like operating on the computer performs part or all of the actual processing based on an instruction of the program code read out by the computer. The case where the functions of the above-described embodiment are realized by the process is also included.

  Further, it is assumed that the program code read out from the storage medium is written to a memory provided in a function expansion board inserted in the computer or a function expansion unit connected to the computer. Thereafter, based on the instruction of the program code, the CPU provided in the function expansion board or the function expansion unit performs part or all of the actual processing, and the processing of the above-described embodiment is realized by the processing. Be

  When the present invention is applied to the storage medium, the storage medium stores program codes corresponding to the flowcharts described above.

11 Extraction means 12 Determination means 13 Determination means 14 Replacement means 15 Display control means 101 CPU
102 ROM
103 RAM
104 document input unit 105 display unit 106 communication unit 107 bus

Claims (9)

An extraction unit for extracting a symbol or a character meaning repetition of information from a document image which is an imaged document;
Determining means for determining the type of the extracted symbol or character;
A determination unit configured to determine information to which the symbol or the character refers based on the determination result by the determination unit;
Replacement means for replacing the symbol or the character with the information to which the symbol or the character determined by the determination means refers ;
An information processing system characterized in that, when the information to which the symbol or the character refers is changed, the replacing means changes the replaced information in accordance with the change of the information to which the symbol or the character refers .   2. The information processing system according to claim 1, further comprising display control means for causing display means to display information to which the symbol or the character replaced by the replacement means refers. 3. The display control means according to claim 2 , wherein when the information referred to by the symbol or the character determined by the determination means is displayed on the display means, the symbol or the character is not displayed on the display means. Information processing system as described.   The determination means determines information to be referred to by the symbol or character based on a table in which the type of the symbol or character is associated with the position in the document indicated by the symbol or character and the determination result by the determination means. The information processing system according to any one of claims 1 to 3, characterized in that:   The information processing system according to any one of claims 1 to 4, wherein the symbol or character includes a nono point. The information processing according to claim 1 , wherein, when the replaced information is changed, the replacing means changes the information to which the symbol or the character refers in accordance with the change of the replaced information. system. The document contains information about drug prescription,
A determination unit that determines whether the information on the prescription of the medicine included in the information replaced by the substitution means satisfies the prescription criteria of the medicine;
The information processing system according to any one of claims 1 to 6 , further comprising: warning means for giving a warning when the standard is not satisfied. An information processing method executed by the information processing system,
An extraction step of extracting a symbol or a character meaning repetition of information from a document image which is an imaged document;
A determination step of determining the type of the extracted symbol or character;
A determination step of determining information to which the symbol or character refers based on the determination result in the determination step;
Replacing the symbol or character with information referred to by the symbol or character determined in the determining step ;
An information processing method comprising: changing information referred to by the symbol or the character, in response to a change of information referred to by the symbol or the character . A program causing a computer to execute the information processing method according to claim 8 .

Images