Medical information processing device, operating method thereof, and program

Abstract

To provide a medical information processing device, operating method thereof, and program, which may provide assistance in generating new medical texts.SOLUTION: A medical information processing device 10 is provided, comprising a medical text acquisition unit 12 for acquiring sentences input by a user according to the input, a break monitoring unit 20 configured to determine a change in attribute of information described in a sentence to be a sentence break, a structuring processing unit 24 for performing structuring processing on each unit separated by the breaks, and a structuring result output unit 28 configured to terminate the determination upon completion of generation of a medical text comprising one or more sentences and output a result of the structuring processing.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The present invention relates to a medical information processing device, an operating method for the medical information processing device, and a program. [Background technology]

[0002] A radiology report, one of the medical documents prepared in the medical field, is a document in which a doctor observes medical images taken using medical equipment such as CT and MRI devices and records the doctor's findings regarding the location, size, shape, internal structure, and other characteristics of each disease.

[0003] Medical documents such as radiology reports are text-based data freely written by doctors and other medical professionals, and are unstructured data that is difficult to use for secondary purposes such as statistical analysis and content analysis.There is a growing need to structure medical documents in order to utilize the information contained in them.

[0004] Patent document 1 describes an information processing device that analyzes free-form text, acquires terminology and expressions such as medical terms, classifies the terminology and expressions based on how they are used, and presents the results of the classification in a unified presentation order.

[0005] The device described in this document analyzes sentences by performing morphological analysis to extract terms used in the sentence on a word-by-word basis and syntactic analysis to analyze word dependency relationships.The device also uses statistical information and co-occurrence relationships to classify term expressions. [Prior art documents] [Patent documents]

[0006] [Patent Document 1] Japanese Patent Application Laid-Open No. 2016-151827 Summary of the Invention [Problem to be solved by the invention]

[0007] However, although Patent Document 1 contains a description regarding the structuring of radiology reports, it does not contain any specific description regarding support for the creation of radiology reports, and it is difficult to understand from the description in Patent Document 1 what kind of support for the creation of radiology reports will be implemented.

[0008] The present invention has been made in view of the above circumstances, and aims to provide a medical information processing device, an operating method for a medical information processing device, and a program that can provide new support for creating medical documents. [Means for solving the problem]

[0009] The medical information processing device according to the present disclosure includes one or more processors and one or more memories storing programs to be executed by the one or more processors, and the one or more processors execute the instructions of the program to obtain sentences entered by a user in accordance with the input, determine a change in the attributes of the information contained in the sentence as a sentence boundary, perform structuring processing for each unit separated by the boundary, terminate the determination upon completion of creation of a medical document containing one or more sentences, and output the results of the structuring processing.

[0010] According to the medical information processing device of the present disclosure, the transition of attributes of terms described in sentences constituting medical text is used as a division of the medical text, and structuring processing is performed for each unit divided using the division, and the structuring processing results are output. As a result, the structuring processing results for each unit divided can be output before the creation of the medical text is completed.

[0011] A sentence entered by a user includes multiple words and periods.

[0012] The information contained in the sentence entered by the user may include the words that make up the sentence.

[0013] The attributes of information may be the attributes of words that make up a sentence.

[0014] Examples of medical documents include radiology reports, findings, and various summaries.

[0015] In a medical image processing device according to another aspect, one or more processors may acquire a sentence representing a finding of a medical image.

[0016] According to this aspect, a structuring process of the finding sentence can be performed, and the result of the structuring process of the finding sentence can be output.

[0017] A medical image is an image generated by imaging a subject using a modality such as a CT scanner. Medical images may be three-dimensional data or two-dimensional reconstructed images.

[0018] The term image may include the meaning of image data, which is a signal representing an image.

[0019] In a medical information processing device according to another aspect, the one or more processors may determine a change in the attribute of information contained in an acquired sentence as a change in a term expressing a finding of a medical image as a break in the sentence.

[0020] According to this aspect, it is possible to perform structuring processing of the sentences that make up the medical image each time a term expressing a finding in the medical image is changed.

[0021] In a medical image processing apparatus according to another aspect, the one or more processors may output the results of the structuring process to a display device.

[0022] According to this aspect, a user such as a doctor can visually recognize the results of the structuring process.

[0023] In a medical image processing apparatus according to another aspect, the one or more processors may output the results of the structuring process for each attribute.

[0024] According to this aspect, a user such as a doctor can recognize the structuring process results for each attribute.

[0025] In another aspect of a medical information processing device, when a term representing an attribute is detected, one or more processors may start a structuring process for a unit including the term representing the attribute, and output a provisional result of the structuring process.

[0026] According to this aspect, a user such as a doctor can recognize the results of the structuring process for the created sentence before completing the creation of the medical document.

[0027] In another aspect of a medical information processing device, one or more processors may acquire past medical text created in the past, and output the results of the structuring processing on the acquired sentence in association with the results of the structuring processing on the past medical text.

[0028] According to this aspect, the structuring process results from the past inspections can be reflected in the structuring process results from the current inspection.

[0029] In another aspect of a medical information processing device, one or more processors may add and output information that is described in past medical documents but is not described in the acquired medical documents to the results of structuring processing on the acquired sentences.

[0030] According to this aspect, the amount of information in the structuring process result can be relatively increased.

[0031] In a medical image processing apparatus according to another aspect, the one or more processors may output a human body schema in association with the result of the structuring process.

[0032] According to this aspect, a user such as a doctor can visually recognize the correspondence between the structuring processing result and the organ.

[0033] In a medical information processing device according to another aspect, the one or more processors may determine a change in a term representing a part of the human body as a sentence break, as a change in an attribute of information described in an acquired sentence.

[0034] According to this aspect, structuring processing of sentences that make up medical text can be performed for each part of the human body.

[0035] In a medical information processing device according to another aspect, the one or more processors may determine a change in a term representing an organ of the human body as a sentence break, as a change in an attribute of information described in an acquired sentence.

[0036] According to this aspect, structuring processing of sentences that make up medical text can be performed for each organ.

[0037] In a medical image processing device according to another aspect, one or more processors may perform a structuring process including a term extraction process that acquires terms in a unit and determines the type of the acquired terms.

[0038] The types of terms in such an embodiment may include date and time, modality used, lesion, location in an organ, degree, and disease name.

[0039] In a medical image processing device according to another aspect, one or more processors may perform a structuring process including a relationship determination process for determining the relationship between a plurality of terms extracted in the term extraction process.

[0040] According to this aspect, a structuring process using the relationship between terms can be performed.

[0041] In a medical image processing apparatus according to another aspect, the one or more processors may group the acquired sentences based on the result of the structuring process.

[0042] According to this aspect, a user such as a doctor can recognize the relationship between the structuring process result and the grouping of sentences included in the medical text.

[0043] In a medical image processing apparatus according to another aspect, one or more processors may perform grouping based on the relationships between multiple terms extracted in the structuring process.

[0044] According to this aspect, it is possible to group sentences that make up medical text based on terms that have a defined relationship.

[0045] In a medical image processing device according to another aspect, the one or more processors may output identification information that identifies the grouping results in the acquired sentences.

[0046] In this case, the identification information may be changed by changing the display mode for each term, such as by color coding for each term.

[0047] In a medical image processing apparatus according to another aspect, the one or more processors may acquire information indicating replacement of each grouped unit in the acquired sentence.

[0048] According to this aspect, when correcting sentences that make up medical text, a group of multiple sentences can be used as a unit of correction.

[0049] In a medical image processing device according to another aspect, the one or more processors may store the acquired sentence breaks.

[0050] According to this aspect, it is possible to reuse segments in medical text.

[0051] In another aspect of the medical information processing device, one or more processors may acquire a correction input representing a correction to a sentence entered by a user, and when the correction input is acquired, update the delimiter in accordance with the correction.

[0052] According to this aspect, the structuring process result is corrected in response to correction of the sentences that make up the medical text.

[0053] In a medical image processing device according to another aspect, the one or more processors may identify the range of correction based on the stored delimiters.

[0054] According to this aspect, correction can be performed for each section of the medical text.

[0055] In a medical image processing device according to another aspect, the one or more processors may output a result of structuring the corrected sentence.

[0056] According to this aspect, a user such as a doctor can recognize the correction of the structuring processing result in response to the correction of the medical text.

[0057] The operating method of a medical information processing device according to the present disclosure is a method of operating a medical information processing device that includes one or more processors and one or more memories that store programs to be executed by the one or more processors, in which the one or more processors acquire sentences input by a user in response to the input, determine a change in the attributes of the information contained in the sentence as a sentence boundary, perform structuring processing for each unit separated by the boundary, terminate the determination upon completion of creation of a medical document containing one or more sentences, and output the results of the structuring processing.

[0058] According to the operation method of the medical image processing device of the present disclosure, it is possible to obtain the same effects as the medical image processing device of the present disclosure. The constituent elements of the medical image processing device of other aspects can be applied to the constituent elements of the operation method of the medical image processing device of other aspects.

[0059] The program of the present disclosure is a program that enables a computer to perform the functions of acquiring a sentence entered by a user in response to the input, determining a change in the attributes of information contained in the sentence as a sentence boundary, performing structuring processing for each unit separated by the boundary, terminating the determination upon completion of the creation of a medical document containing one or more sentences, and outputting the results of the structuring processing.

[0060] According to the program of the present disclosure, it is possible to obtain the same effects as the medical information processing device of the present disclosure. The components of the medical information processing device of other aspects may be applied to the components of the program of other aspects. [Effects of the Invention]

[0061] According to the present invention, the transition of attributes of terms contained in sentences that constitute medical text is used as a division of the medical text, and structuring processing is performed for each unit divided using the division, and the structuring processing results are output. This makes it possible to output the structuring processing results for each unit divided before the creation of the medical text is completed. [Brief explanation of the drawings]

[0062] [Figure 1] FIG. 1 is a functional block diagram showing an overview of the processing functions of a medical image processing apparatus according to an embodiment. [Figure 2] FIG. 2 is a block diagram schematically illustrating an example of the hardware configuration of the medical image processing apparatus shown in FIG. [Figure 3] FIG. 3 is a block diagram showing an outline of a configuration example of a medical information management system to which the medical information processing apparatus shown in FIG. 1 is applied. [Figure 4] FIG. 4 is a flowchart showing the procedure of the structuring processing method according to the embodiment. [Figure 5] FIG. 5 is a diagram showing the configuration of a display screen applied to the medical image processing apparatus shown in FIG. [Figure 6] FIG. 6 is an explanatory diagram of the structuring process applied to the medical image processing apparatus shown in FIG. [Figure 7] FIG. 7 is a schematic diagram showing an example of a display during input of a finding statement. [Figure 8] FIG. 8 is a schematic diagram showing an example of a display screen when a break is detected. [Figure 9] FIG. 9 is a schematic diagram showing a display example when a break is detected during input of a sentence. [Figure 10] FIG. 10 is a schematic diagram showing a display example applied to the structuring process according to the modified example. [Figure 11] FIG. 11 is a schematic diagram showing an example of a display in which the results of the provisional structuring process are displayed. [Figure 12] FIG. 12 is a schematic diagram showing a display example in a state where the input of the observation sentence shown in FIG. 11 has progressed. [Figure 13] FIG. 13 is a schematic diagram showing a display example when the provisional result of the structuring process is updated. [Figure 14] FIG. 14 is a schematic diagram showing a display example in a state where the input of the observation sentence shown in FIG. 13 has progressed. [Figure 15] FIG. 15 is a schematic diagram showing a display example when a break is detected. [Figure 16] FIG. 16 is a schematic diagram showing an example of a display of a finding statement according to a modified example. [Figure 17] FIG. 17 is a schematic diagram showing a display example in a state where the input of the observation sentence shown in FIG. 16 has progressed. [Figure 18] FIG. 18 is a schematic diagram showing a display example in a state where the input of the observation sentence shown in FIG. 17 has progressed. [Figure 19] FIG. 19 is a schematic diagram showing a display example in a state where the input of the observation sentence shown in FIG. 18 has progressed. [Figure 20] FIG. 20 is a schematic diagram showing a display example when a delimiter is defined. [Figure 21] FIG. 21 is a schematic diagram showing another example of the display of findings when the structuring process is finalized. [Figure 22] FIG. 22 is a schematic diagram showing a display example in which confirmation of structuring processing is indicated by changing the display mode of the findings. [Figure 23] FIG. 23 is a schematic diagram showing a display example of the structural processing result associated with the human body schema. [Figure 24] FIG. 24 is a schematic diagram showing another example of displaying the structural processing result associated with the human body schema. [Figure 25] FIG. 25 is a schematic diagram showing an example of a finding sentence and a structuring process result in a previous examination. [Figure 26] FIG. 26 is a schematic diagram showing an example of a display screen during input of a finding statement. [Figure 27] FIG. 27 is a schematic diagram showing a display example of the structured processing result associated with the past examination. [Figure 28] FIG. 28 is a schematic diagram showing a display example before the findings are corrected. [Figure 29] FIG. 29 is a schematic diagram showing a display example during correction of a finding statement. [Figure 30] FIG. 30 is a schematic diagram showing a display example when the findings are corrected. [Figure 31] FIG. 31 is a schematic diagram showing a display example before a finding statement is added. [Figure 32] FIG. 32 is a schematic diagram showing a display example when a finding statement is added. [Figure 33] FIG. 33 is a schematic diagram showing a display example before a finding statement according to another example is added. [Figure 34] FIG. 34 is a schematic diagram showing an example of a screen display when addition of a finding statement according to another example is completed. [Figure 35] FIG. 35 is a schematic diagram showing a display example before the sentences in the findings sentences are rearranged. [Figure 36] FIG. 36 is a schematic diagram showing a display example when the replacement of the findings has been completed. DETAILED DESCRIPTION OF THE INVENTION

[0063] DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Preferred embodiments of the present invention will now be described in detail with reference to the accompanying drawings. In this specification, the same components are designated by the same reference numerals, and redundant explanations will be omitted where appropriate.

[0064] [Configuration example of medical information processing device] 1 is a functional block diagram showing an overview of the processing functions of a medical image processing apparatus according to an embodiment. The medical image processing apparatus 10 shown in the figure can be realized using computer hardware and software.

[0065] The medical information processing device 10 includes a medical text acquisition unit 12, a medical text processing unit 14, and a medical text storage unit 16. The medical text acquisition unit 12 acquires medical text entered by a user such as a doctor using an input device such as a keyboard. Examples of medical text include radiology reports and admission / discharge summaries, which are created by medical professionals and managed as medical information.

[0066] A medical text includes one or more sentences. A sentence includes one or more words and one period. The medical text can be acquired and stored as text data.

[0067] The medical text acquisition unit 12 acquires at least a part of the input sentence in response to an input operation of at least a part of the sentence constituting the input medical text until the input of the sentence being input is completed. "During input" refers to any timing between the start timing of input and the end timing of input.

[0068] The medical text acquisition unit 12 transmits the acquired medical text to the medical text processing unit 14. For example, the medical text acquisition unit 12 may acquire a portion of a sentence that has already been input while the sentence that constitutes the medical text is being input, and transmit the portion of the acquired sentence to the medical text processing unit 14 while the sentence that constitutes the medical text is being input.

[0069] The medical text processing unit 14 acquires the medical text sent from the medical text acquisition unit 12. The medical text processing unit 14 performs specified processing on the acquired medical text. An example of the specified processing is setting display conditions such as font type, size, and color. The medical text processing unit 14 stores the acquired medical text and the processed medical text in the medical text storage unit 16.

[0070] The medical text processing unit 14 transmits the medical text that has been subjected to the specified processing to the medical text output unit 18. The medical text output unit 18 converts the medical text into a format applicable to the display unit 30 and transmits it to the display unit 30.

[0071] The medical information processing device 10 includes a delimiter monitoring unit 20, a delimiter information storage unit 22, a structuring processing unit 24, a structuring result storage unit 26, and a structuring result output unit .

[0072] The delimiter monitoring unit 20 applies the same conditions as the medical text processing unit 14 to acquire medical text via the medical text acquisition unit 12. The delimiter monitoring unit 20 monitors the delimiters of the acquired medical text. The delimiter monitoring unit 20 continues to monitor the delimiters of the acquired medical text until the creation of the medical text is completed.

[0073] The medical text may be divided by changing the attributes of the sentences that make up the medical text. The attribute of the sentence may be changed by changing the attributes of the terms that appear in the sentence. The division monitoring unit 20 stores information about divisions of the medical text in the division information storage unit 22.

[0074] That is, the break monitoring unit 20 determines breaks in the medical text and defines the breaks in the medical text. In other words, the break monitoring unit 20 defines boundaries between groups when grouping the medical text. Details of determining breaks in the medical text will be described later.

[0075] The structuring processing unit 24 acquires medical text with defined delimiters from the delimiter monitoring unit 20, and performs structuring processing for each unit delimited using the delimiters. For example, when a medical text is acquired as a finding sentence containing a doctor's findings, the structuring processing of the finding sentence may include processing such as finding sentence determination, organ determination, finding term extraction, factuality determination, and relationship extraction.

[0076] Finding sentence determination classifies the content of each sentence that makes up the finding sentence into four types: findings, diagnosis, comparison, and others. Organ determination determines the type of anatomical structure, such as an organ, for each sentence. Anatomical structures include organs such as the brain, esophagus, and lungs, as well as blood vessels, nerves, and bones.

[0077] Finding term extraction extracts words related to date and time, modality, lesion, location, severity, disease name, etc. for each sentence. Factuality determination determines whether each sentence is positive or negative. For example, in factuality determination, each sentence is determined to be positive, such as admitted, possible, or suspected, or negative, such as not admitted.

[0078] In relation extraction, relationships between sentences are extracted. For example, in relation extraction, the extracted finding terms for each organ and each lesion are grouped together for multiple sentences that make up a finding sentence.

[0079] The structuring processing unit 24 may apply a trained learning model. For example, a trained learning model that has been trained using pairs of sentences to be structurized and structuring processing results as training data may be applied to the structuring processing unit 24.

[0080] Structuring processing unit 24 stores the results of the structuring processing in structuring result storage unit 26. Structuring processing unit 24 also transmits the results of the structuring processing to structuring result output unit 28. Structuring result output unit 28 converts the results of the structuring processing into a format applicable to display unit 30 and transmits them to display unit 30.

[0081] The display unit 30 may be one or more monitor devices. The display unit 30 may display text representing medical text and the results of the structuring process. The display unit 30 may also display medical images to be interpreted. The display unit 30 may include a first monitor device that displays the medical images, and a second monitor device that displays the medical text and the results of the structuring process for the medical text.

[0082] The medical information processing device 10 may be incorporated into a medical information processing system for processing various types of medical information acquired in a medical institution such as a hospital. In addition, the processing function of the medical information processing device 10 may be provided as a cloud service.

[0083] [Hardware configuration of medical information processing device] Figure 2 is a block diagram showing an example of the hardware configuration of the medical information processing device shown in Figure 1. The medical information processing device 10 can be realized by a computer system configured using one or more computers. Here, an example is shown in which one computer executes a program to realize various functions of the medical information processing device 10.

[0084] The type of computer that functions as the medical information processing device 10 is not particularly limited, and may be a server computer, a workstation, a personal computer, a tablet terminal, etc. The computer may also be a virtual machine.

[0085] The medical information processing apparatus 10 includes a processor 50, a computer-readable medium 52 which is a non-transitory tangible entity, a communication interface 54, an input / output interface 56, and a bus 58. Note that IF shown in Fig. 2 represents an interface.

[0086] The processor 50 includes a CPU (Central Processing Unit). The processor 50 may also include a GPU (Graphics Processing Unit). The processor 50 is connected to a computer-readable medium 52, a communication interface 54, and an input / output interface 56 via a bus 58. The processor 50 reads various programs, data, etc. stored in the computer-readable medium 52 and executes various processes.

[0087] The computer-readable medium 52 includes a memory 60, which is a main storage device, and a storage 62, which is an auxiliary storage device. The storage 62 may be configured using a hard disk drive, a solid-state drive, an optical disk, a magneto-optical disk, or a semiconductor memory. The storage 62 may be configured using an appropriate combination of hard disk drives and the like. The storage 62 stores various programs, data, and the like.

[0088] A hard disk drive may be referred to as an HDD, which is an abbreviation of the English term Hard Disk Drive, and a solid state drive may be referred to as an SSD, which is an abbreviation of the English term Solid State Drive.

[0089] The memory 60 is used as a working area for the processor 50, and as a storage unit that temporarily stores programs and various data read from the storage 62. A program stored in the storage 62 is loaded into the memory 60, and instructions of the program are executed by the processor 50, which functions as a processing unit that performs various processes specified by the program. The memory 60 stores a segmentation monitoring program 70, a structuring program 72, a medical text processing program 74, various data, and the like, which are executed by the processor 50.

[0090] The delimiter monitoring program 70 causes the processor 50 to execute the delimiter monitoring process performed using the delimiter monitoring unit 20 shown in Figure 1. The structuring program 72 causes the processor 50 to execute the structuring process performed using the structuring processing unit 24.

[0091] The medical text processing program 74 causes the processor 50 to execute medical text processing performed using the medical text processing unit 14. The segmentation monitoring program 70, the structuring program 72, and the medical text processing program 74 may include trained learning models. Each program shown in Figure 2 includes one or more instructions. The processor 50 executes the instructions included in each program to realize the function corresponding to each program.

[0092] The communication interface 54 performs communication processing with an external device via a wired or wireless connection, and exchanges information with the external device. The medical information processing device 10 is connected to a communication line via the communication interface 54. The communication line may be a local area network or a wide area network. The communication interface 54 can serve as a data acquisition unit that accepts input of data such as images. The communication line is not shown in the figure.

[0093] The medical information processing device 10 may include a display device 80 and an input device 82. The display device 80 and the input device 82 are connected to the bus 58 via the input / output interface 56. Examples of the input device 82 include a keyboard, a mouse, a multi-touch panel, other pointing devices, and a voice input device. The input device 82 may be an appropriate combination of a keyboard and the like.

[0094] The display device 80 is an output interface that displays various types of information. Examples of the display device 80 include a liquid crystal display, an organic electroluminescence (EL) display, and a projector. The display device 80 may be an appropriate combination of liquid crystal displays and the like. Organic EL is abbreviated as OEL, which stands for organic electroluminescence in English. The display device 80 shown in FIG. 2 can be used as the display unit 30 shown in FIG. 1.

[0095] [Example of medical information management system configuration] Fig. 3 is a block diagram showing an example of the configuration of a medical information management system to which the medical information processing device shown in Fig. 1 is applied. The medical information processing device 10 shown in Fig. 1 can be incorporated into a medical information management system 100 shown in Fig. 3.

[0096] The medical information management system 100 is a computer network established in a medical institution such as a hospital. The medical information management system 100 includes an examination order system 102, an electronic medical record system 104, an image management system 106, an image interpretation report system 108, an image processing system 109, a modality 110, and a terminal device 132.

[0097] A server-type computer may be applied to each of the test order system 102, the electronic medical record system 104, the image management system 106, and the radiology report system 108. The server-type computer may be configured in such a way that multiple computers work together.

[0098] The examination order system 102, the electronic medical record system 104, the image management system 106, the radiology report system 108, the modality 110, and the terminal device 132 are electrically connected to each other via a communication line 130. This allows the components of the medical information management system 100 to communicate with each other.

[0099] The test order system 102 manages test orders issued based on test order requests issued by doctors. Test orders include various information related to tests, such as patient identification information such as a patient ID, identification information of the doctor in charge of the examination such as the doctor ID, and the type of test. The test order system 102 may be equipped with a test order storage device that stores test orders. Note that ID is an abbreviation for identification.

[0100] The electronic medical record system 104 manages electronic medical records for each patient. The electronic medical record system 104 stores patient identification information and electronic medical records in association with each other, and can search for the electronic medical record for each patient using the patient identification information as a parameter.

[0101] The electronic medical record system 104 performs a search in response to a request to read out the medical record for each patient sent from the terminal device 132, and transmits to the terminal device 132 various pieces of information contained in the electronic medical record corresponding to the read out request.

[0102] The electronic medical record system 104 acquires additional information, change information, deletion information, etc. of the electronic medical record transmitted from the terminal device 132, and updates the electronic medical record based on the additional information, etc. The electronic medical record system may include an electronic medical record storage device that stores the electronic medical record.

[0103] The image management system 106 manages medical images taken and generated using the modalities 110. The image management system 106 may include a DICOM server. The image management system 106 may include an image storage device that stores the medical images.

[0104] The DICOM server is a server that operates in accordance with the DICOM specifications. The DICOM server is a computer that stores and manages various data, including images captured using the modality 110. The DICOM server is equipped with a large-capacity external storage device and a database management program.

[0105] The DICOM server communicates with other devices via the communication line 130, sending and receiving various data including image data. The DICOM server receives various data including image data generated using the modality 110 via the communication line 130, and stores and manages the data on a recording medium such as a large-capacity external storage device. The storage format of the image data and communication between devices via the communication line 130 are based on the DICOM protocol.

[0106] The image management system 106 may store three-dimensional images configured from imaging data of the modality 110, or may store two-dimensional reconstructed images reconstructed from the three-dimensional images. The image management system 106 may transmit three-dimensional images to the image processing system 109, the terminal device 132, etc., or may transmit two-dimensional reconstructed images reconstructed from the three-dimensional images.

[0107] The image interpretation report system 108 manages image interpretation reports created by image interpretation doctors. The image interpretation report system 108 stores image interpretation reports associated with electronic medical records and medical images. The image interpretation report system 108 includes an image interpretation report creation support device that supports the creation of image interpretation reports.

[0108] The image processing system 109 can acquire data from the image management system 106 or the like via the communication line 130. The image processing system 109 performs various processes, such as image analysis, on medical images captured using the modality 110. The image processing system 109 may be configured to perform various analytical processes, such as computer-aided diagnosis, for example, a process for recognizing a lesion area from an image, a process for identifying a disease name or other classification, and a segmentation process for recognizing an organ or other area. Note that computer-aided diagnosis may be referred to as CAD, which is an abbreviation for Computer Aided Diagnosis or Computer Aided Detection.

[0109] Furthermore, the image processing system 109 can send the processing results to the image management system 106 and the terminal device 132. The processing functions of the image processing system 109 may be installed in the image management system 106 or the terminal device 132.

[0110] Various pieces of information, including various data stored in the database of the image management system 106 and processing results generated using the image processing system 109, can be displayed on a monitor device connected to the terminal device 132.

[0111] The terminal device 132 can function as a terminal for viewing images called a PACS viewer or a DICOM viewer. Multiple terminal devices 132 can be connected to the communication line 130. PACS is an abbreviation for Picture Archiving and Communication System. The form of the terminal device 132 is not particularly limited, and it may be a personal computer, a workstation, a tablet terminal, or the like.

[0112] 1 and 2 may be configured using an interpretation report system 108 and a terminal device 132 shown in Fig. 3. For example, the medical information processing device 10 may be configured by linking an interpretation report creation support device provided in the interpretation report system 108 with the terminal device 132.

[0113] Specific examples of the modality 110 include a CT device 112, an MRI device 114, an ultrasound diagnostic device 116, a PET device 118, an X-ray diagnostic device 120, an X-ray fluoroscopy diagnostic device 122, and an endoscopy device 124. There may be various combinations of types of modality 110 connected to the communication line 130 depending on the medical institution. Note that MRI is an abbreviation for Magnetic Resonance Imaging. PET is an abbreviation for Positron Emission Tomography.

[0114] The communication line 130 may be a local communication line within a medical institution, or a part of the communication line 130 may be a wide area communication line.

[0115] [Structuring process for creating medical documents] Fig. 4 is a flowchart showing the steps of a structuring processing method according to an embodiment. The structuring processing method shown in Fig. 4 is applied when, in the medical image acquisition step S10, a doctor interprets medical images acquired by the medical information processing device 10 shown in Fig. 1, and when the doctor discovers a lesion, the doctor enters a written observation. The structuring processing method shown in Fig. 4 is an example of an operating method of the medical information processing device that executes various processes using each unit of the medical information processing device 10, which is a computer.

[0116] In the medical text acquisition step S11, the medical text acquisition unit 12 shown in Fig. 1 acquires the findings entered by the doctor as medical text. In the medical text acquisition step S11, a part of the findings that have already been entered may be acquired while the findings are being entered. After the medical text acquisition step S11, the process proceeds to the medical text storage step S12.

[0117] In the medical text storage step S12, the medical text processing unit 14 stores the finding sentences acquired in the medical text acquisition step S11 in the medical text storage unit 16. The finding sentences are stored in the medical text storage unit 16 for each acquired finding sentence. After the medical text storage step S12, the process proceeds to the delimiter monitoring step S14.

[0118] In the break monitoring step S14, the break monitoring unit 20 monitors breaks in the finding sentences acquired in the medical text acquisition step S11. For example, in the break monitoring step S14, organ recognition may be performed for each finding sentence, and a change in organ may be determined as a break in the finding sentence.

[0119] In the break monitoring step S14, if the break monitoring unit 20 does not find a break in the observation sentence, the result is No. If the result is No, the break monitoring step S14 continues until the result is Yes.

[0120] On the other hand, if the delimiter monitoring unit 20 detects a break in the finding sentence in the break monitoring step S14, the result is determined to be Yes. If the determination is Yes, the process proceeds to the structuring processing step S16. In the break monitoring step S14, the break in the finding sentence determined by the break monitoring unit 20 is stored in the break information storage unit 22 as break information.

[0121] In the structuring process step S16, the structuring processor 24 performs structuring process for each unit of observation sentence separated by the separator identified in the separator monitoring step S14. After the structuring process step S16, the process proceeds to the structuring result storage step S18.

[0122] In the structuring result storage step S18, the structuring processing unit 24 stores the structuring process results of the observation sentences in the structuring process step S16 in the structuring result storage unit 26. After the structuring result storage step S18, the process proceeds to the structuring result output step S20.

[0123] In the structuring result output step S20, the structuring result output unit 28 outputs the results of the structuring process of the finding sentences performed in the structuring process step S16. In the structuring result output step S20, a mode can be implemented in which the results of the structuring process of the finding sentences are displayed using the display unit 30. After the structuring result output step S20, the process proceeds to the medical text output step S22.

[0124] In the medical text output step S22, the medical text output unit 18 outputs the findings acquired in the medical text acquisition step S11 and subjected to specified processing using the medical text processing unit 14. In the medical text output step S22, a mode in which the findings are displayed using the display unit 30 can be implemented.

[0125] The order of the structuring result output step S20 and the medical text output step S22 may be reversed or may be executed in parallel. After the structuring result output step S20 and the medical text output step S22 are executed, the process proceeds to the medical text acquisition completion determination step S24.

[0126] In the medical text acquisition completion determination step S24, the medical text acquisition unit 12 determines whether or not the acquisition of medical text has been completed. If the medical text acquisition unit 12 determines that the acquisition of medical text should continue in the medical text acquisition completion determination step S24, the determination is No. If the determination is No, the process proceeds to the medical text acquisition step S11, and each step from the medical text acquisition completion determination step S11 to the medical text acquisition completion determination step S24 is repeatedly executed until the determination is Yes in the medical text acquisition completion determination step S24.

[0127] On the other hand, in the medical text acquisition completion determination step S24, if the medical text acquisition unit 12 determines that the acquisition of the medical text has been completed, the determination is Yes. If the determination is Yes, the structuring processing method is terminated.

[0128] [Example of display screen configuration] Fig. 5 is a configuration diagram of a display screen applied to the medical information processing device shown in Fig. 1. The display screen 200 shown in the figure is displayed on the display unit 30 shown in Fig. 1 when a medical image 202 is acquired, a doctor interprets the medical image 202, and creates a statement of findings 204.

[0129] The display screen 200 includes a medical image display section 210 that displays a medical image 202, and a text display section 212 that displays a finding statement 204 and a structuration processing result 206. Note that the configuration of the display screen 200 shown in Fig. 5 is an example, and can be changed, added, or deleted as appropriate.

[0130] [Examples of structured processing] Fig. 6 is an explanatory diagram of the structuring process applied to the medical information processing device shown in Fig. 1. Fig. 6 shows a schematic diagram of a display example of a finding statement 204 and a structuring process result 206. Note that the frame surrounding the finding statement 204 shown in Fig. 6 is shown for convenience of explanation and is not actually displayed on the display screen 200. In addition, the arrow line and organ name shown on the right side of the finding statement 204 in Fig. 6 are also shown for convenience of explanation and are not actually displayed on the display screen 200.

[0131] First, a doctor interprets a medical image and, if a lesion is found, inputs a finding statement. The medical information processing device 10 shown in Figure 1 acquires the finding statements entered by the doctor, performs organ recognition for each statement, and determines whether the organ has changed. Figure 6 illustrates a case in which a first statement 220, a second statement 222, a third statement 224, a fourth statement 226, and a fifth statement 228 are acquired, and the organ recognition results for the first to fourth statements are the lungs, and the organ recognition result for the fifth statement is the liver.

[0132] Based on the result of organ recognition in the fifth sentence 228, the medical information processing device 10 determines that there is an organ change between the fourth sentence 226 and the fifth sentence 228, and defines the space between the fourth sentence 226 and the fifth sentence 228 as a division 240.

[0133] That is, the medical information processing device 10 extracts terms representing organs used in sentences constituting the finding sentences 204, and groups the finding sentences 204 by organ. In other words, the medical information processing device 10 performs a relationship determination process to determine the relationship between terms used in sentences constituting the finding sentences 204, and groups the sentences constituting the finding sentences 204 based on the relationship between the terms. Note that the change of organs described in the embodiment is an example of a change of body part.

[0134] Once the division 240 is defined, structuring processing is performed on a section of text including the first sentence 220 to the fourth sentence 226. Figure 6 shows, as an example of structuring processing, a structuring process for classifying terms used in the findings sentence 204 into organ name, location in the organ, lesion name, factuality of the lesion, disease name, factuality of the disease name, size and characteristics of the lesion.

[0135] When the structuring process result for a block of sentence, which is the processing unit of the structuring process, is derived, the structuring process result 206 is displayed. Figure 6 shows the structuring process result for the lungs, which is written in the first sentence 220 to the fourth sentence 226. 6 are shown for convenience of explanation, and are not actually displayed on the display screen 200. The same applies to the partitions 240 in the figures from FIG.

[0136] Fig. 7 is a schematic diagram showing an example of a display during input of a finding sentence. Fig. 7 shows a finding sentence 204 in which a first sentence 220, a second sentence 222, and a third sentence 224 have been input. When the finding sentence 204 shown in Fig. 7 is displayed, it has not been determined that an organ change has occurred, a delimiter has not been defined, structuring processing has not been performed, and the structuring processing result 206 is not displayed. Note that Fig. 7 shows a table with blank fields as an example of a non-display mode of the structuring processing result 206.

[0137] Fig. 8 is a schematic diagram showing an example of a display screen when a break is detected. Fig. 8 shows a finding sentence 204 when a fifth sentence 228 is input, it is determined that there is a change in organ between the fourth sentence 226 and the fifth sentence 228, and a break 240 is defined between the fourth sentence 226 and the fifth sentence 228. When the finding sentence 204 shown in Fig. 8 is displayed, a structuring process result 206 of a sentence related to the lungs is also displayed.

[0138] Fig. 9 is a schematic diagram showing a display example when a break is detected during sentence input. Fig. 9 shows a finding sentence 204 in which a change in organ is determined during input of the fifth sentence 228, and a break 240 is defined between the fourth sentence 226 and the fifth sentence 228 during input of the fifth sentence 228. When the break 240 is defined during input of the fifth sentence 228, structuring processing is performed during input of the fifth sentence 228, and the structuring processing result 206 is displayed.

[0139] [Variations of structuring processing] In the structuring process according to the modified example, a provisional structuring process result 206A, which is a provisional structuring process result, is displayed until the organ change is determined, and when a delimiter 240 is defined and the medical text to be structured is confirmed, the structuring process result is confirmed.

[0140] Fig. 10 is a schematic diagram showing a display example applied to structuring processing according to a modified example. Fig. 10 illustrates an observation sentence 204 in which a first sentence 220 has been partially input. While the first sentence 220 is being input, from the time the period of the first sentence 220 is input until the first sentence 220 is finalized, structuring processing is not performed on the first sentence 220, and the structuring processing result 206 and the tentative structuring processing result 206A are not displayed.

[0141] Fig. 11 is a schematic diagram showing an example of a display in which the tentative structuring process result is displayed. Fig. 11 illustrates a case in which the first sentence 220 is finalized, structuring process is performed on the first sentence 220, and the tentative structuring process result 206A of the first sentence 220 is displayed.

[0142] Fig. 12 is a schematic diagram showing a display example in a state where the input of the observation sentence shown in Fig. 11 has progressed. Fig. 12 illustrates an observation sentence 204 in which a second sentence 222 has been input partway. During the input of the second sentence 222, from when the period of the second sentence 222 is input until the second sentence 222 is finalized, structuring processing is not performed on the second sentence 222, and the display of the provisional structuring processing result 206A of the first sentence 220 continues.

[0143] Fig. 13 is a schematic diagram showing a display example when the provisional results of the structuring process have been updated. Fig. 13 illustrates a case where the second sentence 222 has been finalized, structuring process has been performed on the second sentence 222, and the provisional structuring process result 206A has been updated. In the provisional structuring process result 206A shown in Fig. 13, the structuring process result of the second sentence 222 has been added to the attribute column.

[0144] Fig. 14 is a schematic diagram showing a display example in a state where the input of the observation sentences shown in Fig. 13 has progressed. Fig. 14 shows a provisional structuring processing result 206A in which the third sentence 224 and the fourth sentence 226 have been finalized, structuring processing has been performed on the third sentence 224 and the fourth sentence 226, and the structuring processing results of the third sentence 224 and the fourth sentence 226 have been added. In the provisional structuring processing result 206A shown in Fig. 14, the structuring processing results of the third sentence 224 and the fourth sentence 226 have been added to the characteristics column.

[0145] Fig. 15 is a schematic diagram showing a display example when a break is detected. Fig. 15 illustrates a finding sentence 204 when a change in organ is detected during input of the fifth sentence 228 and a break 240 is defined.

[0146] When the delimiter 240 is defined, the structuring process result 206 is finalized from the first sentence 220 to the fourth sentence 226. Fig. 15 shows a display mode of the structuring process result 206 indicating that the structuring process result 206 has been finalized.

[0147] For example, when the structuring process result 206 is finalized, the color applied to the provisional structuring process result 206A may be changed in the structuring process result 206. Instead of or in addition to changing the color, changes in font, size, etc. may also be implemented.

[0148] [Modification of the findings display] Fig. 16 is a schematic diagram showing an example of the display of a finding sentence according to a modified example. Fig. 16 illustrates a case in which right lung S4 is extracted as a result of the term extraction process in the finding sentence 204 being input in the first sentence 220, and the characters "right lung S4" are colored red to indicate the organ and location. The solid underline added to right lung S4 indicates that the characters "right lung S4" have been colored red.

[0149] Fig. 17 is a schematic diagram showing a display example when the input of the findings shown in Fig. 16 has progressed. Fig. 17 shows the state when the input of the first sentence 220 has been completed. As a result of the term extraction process for the first sentence 220, the following terms are extracted: long diameter 3 cm, solid mass, and observed.

[0150] The illustration shows the case where the term "longest diameter 3 cm" is colored blue to indicate size, the term "solid mass" is colored green to indicate lesion, and the term "observed" is colored orange to indicate factuality.

[0151] The dotted underline around the 3cm long diameter indicates that the words "3cm long diameter" have been colored blue, the dashed line around the solid mass indicates that the words "solid mass" have been colored green, and the two-dot chain line around the word "recognized" indicates that the words "recognized" have been colored orange.

[0152] Fig. 18 is a schematic diagram showing a display example in a state where the input of the finding sentence shown in Fig. 17 has progressed. Fig. 18 illustrates a case where, in the finding sentence 204 during input of the second sentence 222, the terms boundary, unclear, marginal, and lobulated are extracted as a result of the term extraction process for the second sentence 222.

[0153] The finding sentence 204 shown in Figure 18 illustrates a case where the terms "borderline," "indistinct," "marginal," and "lobulated" are colored purple to indicate the characteristics. The dashed dotted underlines added to the borders, etc. indicate that the borders, etc. are colored purple.

[0154] FIG. 19 is a schematic diagram showing a display example in a state where the input of the findings shown in FIG. 18 has progressed. FIG. 19 shows the state where the input of the first sentence 220 to the fourth sentence 226 has been completed. FIG. 19 illustrates a case where the terms "boundary invagination" and "observed" have been extracted as a result of the term extraction process for the third sentence 224. FIG. 19 also illustrates a case where the terms "interior," "calcification," "cavity," "bronchial radiograph," and "does not include" have been extracted as a result of the term extraction process for the fourth sentence 226.

[0155] The finding sentence 204 shown in FIG. 19 illustrates a case in which the terms "internal," "calcification," "cavity," and "bronchial radiograph" are colored purple to indicate characteristics, and the term "does not include" is colored orange to indicate factuality.

[0156] Fig. 20 is a schematic diagram showing a display example when a separator is defined. Fig. 20 shows an observation sentence 204 when a separator 240 is defined during input of the fifth sentence 228. Note that the line segment indicating the separator 240 is omitted from Fig. 20. The same applies to Fig. 21.

[0157] In the fifth sentence 228 shown in Fig. 20, the term "liver" is colored red to represent an organ, and the term "microcyst" is colored green to represent a lesion. Fig. 20 also illustrates a case where the structuring process results 206 from the first sentence 220 to the fourth sentence 226 are finalized and the structuring process results 206 from the first sentence 220 to the fourth sentence 226 related to the lungs are displayed.

[0158] Fig. 21 is a schematic diagram showing another example of the display of the observation sentences when the structuring process has been finalized. Fig. 21 shows an example in which the color of the first sentence 220 to the fourth sentence 226, for which the structuring process result 206 shown in Fig. 20 has been finalized, is returned to black.

[0159] In the observation sentence 204 shown in FIG. 21, it is easy to distinguish between the first sentence 220 to the fourth sentence 226 for which the structuring process result 206 has been determined, and the fifth sentence 228 for which the structuring process result 206 has not been determined.

[0160] [Display of structured processing confirmation using change in the display mode of findings] Fig. 22 is a schematic diagram showing a display example in which confirmation of structuring processing is indicated by changing the display mode of the findings. When the structuring processing results 206 of the first sentence 220 to the fourth sentence 226 are confirmed, the findings 204 shown in Fig. 22 changes from black when input to blue.

[0161] 22 illustrates a change in character color as an example of a change in the display mode of the finding statement 204, but the change in the display mode of the finding statement 204 may also be achieved by applying a change in character size, a change in character font, emphasis processing, etc. Furthermore, the change in the display mode of the finding statement 204 may also be achieved by appropriately combining a change in character color, a change in character size, a change in character font, and emphasis processing.

[0162] The display mode of the observation sentences shown in FIG. 22 makes it easy to distinguish between the first sentence 220 to the fourth sentence 226 for which structuring processing has been confirmed and the fifth sentence 228 for which structuring processing has not been confirmed.

[0163] [Matching between human body schema and structural processing results] Fig. 23 is a schematic diagram showing a display example of the structuring processing result associated with a human body schema. Fig. 23 shows an example in which the finding statement 204 and the structuring processing result 206 are displayed, and also a human body schema 250 is displayed, using the text display unit 212 shown in Fig. 5. In the human body schema 250 shown in Fig. 23, the lung 252, which is the organ to be structuring processed, is displayed in color.

[0164] Fig. 24 is a schematic diagram showing another example of the display of the structuring processing results associated with a human body schema. Fig. 24 shows a case where a fifth sentence 228 for the liver is determined, structuring processing is performed on the fifth sentence 228, and the structuring processing result 206 of the fifth sentence 228 is displayed. In the human body schema 250 shown in Fig. 24, the liver 254, which is the organ to be structuring processed, is displayed in color.

[0165] The structuring process results 206 of the first sentence 220 to the fourth sentence 226 for the lungs shown in Fig. 23 and the structuring process result 206 of the fifth sentence 228 for the liver shown in Fig. 24 may be displayed on the same screen. Furthermore, the structuring process results 206 of the finding sentences 204 may be displayed for each organ specified in the human body schema 250.

[0166] The correspondence between the human body schema 250 and the structuring processing result 206 can be easily understood by using the human body schema 250, which organ displays the structuring processing result.

[0167] [Matching with past examinations] The medical information processing device 10 shown in Figure 1 may acquire past examination results from the electronic medical record system 104 shown in Figure 3, and when creating an interpretation report for the current examination, may reflect the results of structuring the interpretation report and findings contained in the past examination results.

[0168] Fig. 25 is a schematic diagram showing an example of a finding sentence and a structuring process result in a past examination. Fig. 25 shows a finding sentence 204B and a structuring process result 206B included in the past examination result corresponding to the current examination.

[0169] Fig. 26 is a schematic diagram showing an example of a display screen during input of a finding sentence. Fig. 26 shows a case where a separator 240 is defined between a first sentence 260 and a second sentence 262, structuring processing is performed on the first sentence 260, and the structuring processing result 206 is displayed.

[0170] Fig. 27 is a schematic diagram showing a display example of the structuring process result associated with the past examination. Fig. 27 shows the structuring process result 206C in which the structuring process result 206B of the finding sentence 204B in the past examination shown in Fig. 25 is associated with the structuring process result 206 shown in Fig. 26.

[0171] The structuring processing result 206C shown in Figure 27 is the structuring processing result 206 shown in Figure 26, with the right lung and middle lobe added as information in the location column, and information indicating that the size of the tumor has increased from 40 millimeters to 50 millimeters added as information in the size column.

[0172] The structuring process result 206C associated with the structuring process result 206 included in the past examination result has an increased amount of information compared to the structuring process result 206 of the finding sentence 204 in the current examination.

[0173] [Example of display corresponding to correction of findings] Fig. 28 is a schematic diagram showing an example of a screen display before the finding sentence is corrected. Fig. 28 illustrates a finding sentence 204 in which a first sentence 280 about the lungs, a second sentence 282 about the liver, and a third sentence 284 about the gallbladder have been input, and a fourth sentence 286 about the kidneys is being input, as well as a structuring process result 206 of the finding sentence 204.

[0174] Figure 29 is a schematic diagram showing a display example of a finding sentence being corrected. Figure 29 shows the finding sentence 204 when the fourth sentence 286 related to the kidney shown in Figure 28 has been deleted. Since the structuring process for the lungs, liver, and gallbladder has been finalized, the structuring process result 206 remains unchanged.

[0175] Fig. 30 is a schematic diagram showing a display example when a finding sentence is corrected. In finding sentence 204 shown in Fig. 30, fourth sentence 286D related to the pancreas is input instead of fourth sentence 286 related to the kidney shown in Fig. 28. Structuring process result 206D shown in Fig. 30 is obtained by adding the structuring process result of fourth sentence 286D related to the pancreas to structuring process result 206 shown in Figs. 28 and 29.

[0176] 28 to 30, the structuring process result 206 is corrected in accordance with the correction of the finding sentence 204. As a result, the structuring process result 206 is generated while maintaining the correspondence with the corrected finding sentence 204.

[0177] [Example of display for adding findings] Fig. 31 is a schematic diagram showing a display example before the finding sentence is added. Fig. 31 shows a finding sentence 204 and a structuring process result 206 of the finding sentence 204, which includes a first sentence 280 about the lungs, a second sentence 282 about the liver, and a third sentence 284 about the kidneys.

[0178] Fig. 32 is a schematic diagram showing a display example when a finding sentence is added. In finding sentence 204E shown in Fig. 32, a third sentence 284E about the gallbladder is added to finding sentence 204 shown in Fig. 31, and third sentence 284E about the kidney shown in Fig. 31 is changed to a fourth sentence 286E.

[0179] In the structuring process result 206E shown in FIG. 32, a structuring process result column 294 for the gallbladder is inserted between a structuring process result column 290 for the liver and a structuring process result column 292 for the liver related to the pancreas, in accordance with the order in which each sentence constituting the finding sentence 204E is written.

[0180] Furthermore, the positions of the divisions are also updated in response to the addition of the third sentence 284E related to the gallbladder in the finding sentence 204E. That is, the positions of the divisions in the finding sentence 204 shown in Fig. 31 are between the first sentence 280 related to the lungs and the second sentence 282 related to the liver, and between the second sentence 282 related to the liver and the third sentence 284 related to the pancreas.

[0181] On the other hand, the division positions of the finding sentence 204E shown in FIG. 32 are between the first sentence 280 regarding the lungs and the second sentence 282 regarding the liver, between the second sentence 282 regarding the liver and the third sentence 284E regarding the gallbladder, and between the third sentence 284E regarding the gallbladder and the fourth sentence 286E regarding the pancreas.

[0182] Fig. 33 is a schematic diagram showing a display example before a finding sentence according to another example is added. The finding sentence 204 shown in Fig. 33 includes a first sentence 280 about the lungs, a second sentence 282 about the liver, a third sentence 284 about the gallbladder, a fourth sentence 286 about the kidneys, and a fifth sentence 288 about the pancreas.

[0183] The structuring process result 206 shown in FIG. 33 includes the structuring process results for each of the first sentence 280 to the fifth sentence 288.

[0184] In finding sentence 204F shown in Figure 34, a fifth sentence 286F regarding a kidney cyst is added to finding sentence 204 shown in Figure 33. In addition, fifth sentence 288 shown in Figure 33 is changed to sixth sentence 288F. In finding sentence 204F shown in Figure 34, the position of the delimiter is changed compared to finding sentence 204 shown in Figure 33.

[0185] In structuration processing result 206F shown in Figure 34, a structuration processing result column 296 related to kidney cysts is added in response to the addition of fifth sentence 286F related to kidneys. In structuration processing result 206F shown in Figure 34, the order of organs is not changed compared to structuration processing result 206 shown in Figure 33. That is, the order of organs in structuration processing result 206 shown in Figure 33 is lungs, liver, gallbladder, kidneys, and pancreas, and the order of organs in structuration processing result 206F shown in Figure 34 is the same.

[0186] 31 to 34, the structuration processing result 206 is added in response to the addition of the finding sentence 204. As a result, the structuration processing result 206 corresponding to the addition of the finding sentence 204 is generated.

[0187] [Example of grouping display] Figure 35 is a schematic diagram showing a display example before the sentences of the findings are rearranged. Figure 35 includes a first sentence 280 about the lungs, a second sentence 282 about the liver, a third sentence 284 about the gallbladder, a fourth sentence 286 about the kidneys, a fifth sentence 287 about renal cysts, and a sixth sentence 289 about the pancreas. Figure 35 also shows a structural processing result 206 corresponding to the finding sentence 204.

[0188] Figure 35 shows a state in which, when the order of the fourth sentence 286 regarding the kidney and the fifth sentence 287 regarding the renal cyst is swapped with the sixth sentence 289 in the finding sentence 204, the fourth sentence 286 regarding the kidney and the fifth sentence 287 regarding the renal cyst are selected and highlighted.

[0189] Figure 35 shows an example in which, when selecting the fourth sentence 286 relating to the kidney and the fifth sentence 287 relating to the renal cyst, the operator uses the mouse 82A to select the fourth sentence 286 relating to the kidney and the fifth sentence 287 relating to the renal cyst, and the medical information processing device 10 acquires this as a correction input.

[0190] 36 is a schematic diagram showing a display example when the replacement of the observation sentences is completed. In the observation sentence 204G shown in FIG. 36, the positions of the fourth sentence 286 and the fifth sentence 287 are changed from the positions following the third sentence 284 to the line following the sixth sentence 289.

[0191] In addition, in structuration processing result 206G shown in FIG. 36, the positions of structuration processing result column 298 and structuration processing result column 299 relating to the kidney are changed compared to structuration processing result 206 shown in FIG. 35 in accordance with the change in finding sentence 204G.

[0192] 35 and 36, segments are monitored according to the attributes of the sentences, and sentences are grouped based on the attributes of the sentences. Then, a sentence to be replaced can be moved by dragging it with the mouse 82A. Even if there are multiple sentences to be replaced, multiple sentences can be selected. Furthermore, the structuring process result 206 can be changed in response to changes in the observation sentence 204.

[0193] [About the programs that run the computer] A program that causes a computer to realize the processing functions of the medical information processing device 10 can be recorded on a computer-readable medium, which is a tangible, non-transitory information storage medium such as an optical disk, a magnetic disk, or a semiconductor memory, and the program can be provided through this information storage medium.

[0194] In addition, instead of providing the program by storing it on such a tangible, non-transitory computer-readable medium, it is also possible to provide the program signal as a download service using a telecommunications line such as the Internet.

[0195] Furthermore, some or all of the processing functions of the medical information processing device 10 may be realized by cloud computing, and may also be provided as a Software as a Service (SasS) service. SasS is an abbreviation for Software as a Service.

[0196] [Hardware configuration of each processing unit] The hardware structure of the processing units in the medical information processing device 10 that perform various processes, such as the medical text acquisition unit 12, medical text processing unit 14, medical text output unit 18, segmentation monitoring unit 20, structuring processing unit 24, and structuring result output unit 28, is, for example, various processors as shown below.

[0197] Various types of processors include CPUs, which are general-purpose processors that execute programs and function as various processing units, GPUs, which are processors specialized for image processing, programmable logic devices such as FPGAs (Field Programmable Gate Arrays), which are processors whose circuit configuration can be changed after manufacturing, and dedicated electrical circuits such as ASICs, which are processors with circuit configurations designed specifically to perform specific processes.

[0198] A programmable logic device (PLD) is an abbreviation of Programmable Logic Device (ASIC), and Application Specific Integrated Circuit (ASIC).

[0199] A single processing unit may be configured with one of these various processors, or may be configured with two or more processors of the same or different types. For example, a single processing unit may be configured using multiple FPGAs, a combination of a CPU and an FPGA, or a combination of a CPU and a GPU.

[0200] Alternatively, multiple processing units may be configured as a single processor. Examples of multiple processing units configured as a single processor include, first, a form in which a single processor is configured using a combination of one or more CPUs and software, as typified by computers such as client and server computers, and this processor functions as multiple processing units. Second, a form in which a processor is used to realize the functions of an entire system including multiple processing units on a single IC chip, as typified by system-on-chip. A system-on-chip may be referred to as SoC, an abbreviation for System On a Chip. IC is an abbreviation for Integrated Circuit.

[0201] In this way, the various processing units are configured as hardware structures using one or more of the various processors described above.More specifically, the hardware structures of these various processors are electric circuits (circuitry) that combine circuit elements such as semiconductor elements.

[0202] [Effects of the embodiment] The medical image processing apparatus according to the embodiment can achieve the following advantageous effects.

[0203] [1] While the user is entering a finding sentence, the sentences that make up the finding sentence are acquired, and a change in the attribute of a term described in the acquired sentence is determined as a division in the finding sentence. A structuring process is performed for each unit separated by the division, and the results of the structuring process are displayed. This makes it possible to output the results of the structuring process for each unit separated without waiting for the completion of the creation of the finding sentence.

[0204] In other words, when a break is determined, structuring processing is performed for each unit separated by the break. This allows the structuring processing, which has a relatively heavy processing load compared to when structuring processing is performed after the observation statement is created, to be performed while the observation statement is being created, thereby reducing the stress that users such as doctors experience while waiting for processing.

[0205] [2] When a break in a finding sentence is detected, structuring processing is performed for each break, even while the next sentence is being input. This makes it possible to output the results of structuring processing for each break without waiting for the finding sentence to be finalized.

[0206] [3] Until a break in the observation sentence is detected, a provisional structuring process result for each sentence is output. This makes it possible to output a structuring process result for a sentence whose input has been completed while the observation sentence is being input.

[0207] [4] When the structuring process result is confirmed, the display mode of the structuring process result is changed, so that the user can visually confirm that the structuring process result has been confirmed.

[0208] [5] Before determining the boundaries of the findings, term extraction is performed for each sentence, and a different display mode is applied to each attribute of the term, allowing the user to visually recognize the attributes of the terms in the sentences that make up the findings.

[0209] [6] When the structuring process result is confirmed, the display mode of the sentence for which the structuring process result has been confirmed is changed, so that the user can visually recognize the sentence for which the structuring process result has been confirmed.

[0210] [7] A human body schema associated with the structuring process result is displayed, allowing the user to visually recognize the correspondence between the structuring process result and the organ.

[0211] [8] Observation sentences created in past examinations are acquired, and the structuring process results of the observation sentences created in the past examinations are reflected in the structuring process results of the observation sentences created for the current examination, thereby making it possible to relatively increase the amount of information in the structuring process results.

[0212] [9] The structural processing result is corrected in accordance with the correction of the finding sentence, so that the finding sentence and the structural processing result are consistent with each other.

[0213] The technical scope of the present invention is not limited to the scope described in the above embodiments. The configurations and the like in each embodiment can be appropriately combined with each other within the scope that does not deviate from the spirit of the present invention. [Explanation of symbols]

[0214] 10 Medical information processing device 12 Medical Text Acquisition Department 14 Medical Text Processing Unit 16 Medical text storage section 18 Medical text output unit 20 Separation monitoring unit 22 Delimiter information storage unit 24 Structured Processing Section 26 Structured result storage 28 Structured result output section 30 Display section 50 processors 52 Computer-readable medium 54 Communication Interface 56 Input / Output Interface 58 Bus 60 memory 62 Storage 70 Delimiter Monitoring Program 72 Structured Program 74 Medical text processing program 80 Display device 82 Input Device 82A Mouse 100 Medical Information Management System 102 Inspection Order System 104 Electronic Medical Record System 106 Image Management System 108 Image Reading Report System 109 Image Processing System 110 Modalities 112 CT device 114 MRI machine 116 Ultrasound diagnostic equipment 118 PET equipment 120 X-ray diagnostic equipment 122 X-ray fluoroscopy equipment 124 Endoscopic Devices 130 Communication Lines 132 Terminal Equipment 200 display screen 202 Medical Imaging 204 Observations 204B Observations 204E Observations 204F Observations 204G Observations 210 Medical image display unit 212 Text display section 220 1st sentence 222 Second sentence 224 Third sentence 226 4th sentence 228 Sentence 5 240 divisions 250 Human Body Schema 252 Lungs 254 Liver 260 1st sentence 262 Second sentence 280 1st sentence 282 Second sentence 284 3rd sentence 284E 3rd sentence 286 4th sentence 286D 4th sentence 286E 4th sentence 286F 5th sentence 287 Sentence 5 288 Sentence 5 288F 6th sentence 289 Sentence 6 290 Structured Processing Result Column 292 Structured Processing Result Column 294 Structured Processing Result Column 296 Structured Processing Result Column 298 Structured Processing Result Column 299 Structured Processing Result Column Each step from S10 to S24 Each step in the structuring process

Claims

1. one or more processors; one or more memories in which programs to be executed by the one or more processors are stored; Equipped with The one or more processors execute instructions of the program, Obtaining a sentence input by a user in response to said input; determining a change in the attribute of the information described in the sentence as a break in the sentence; A structuring process is performed for each unit separated by the separator, The determination is terminated in response to completion of creation of a medical document including one or more of the sentences; A medical image processing device that outputs the results of the structuring process.

2. The medical image processing apparatus according to claim 1 , wherein the one or more processors acquire the sentences representing findings of medical images.

3. The medical information processing apparatus according to claim 2 , wherein the one or more processors determine a change in a term representing a finding of a medical image as a break in the sentence, as a change in an attribute of information described in the acquired sentence.

4. The medical image processing apparatus according to claim 1 , wherein the one or more processors output the results of the structuring process to a display device.

5. The medical image processing apparatus according to claim 1 , wherein the one or more processors output a result of the structuring process for each of the attributes.

6. The medical information processing device according to any one of claims 1 to 5, wherein the one or more processors, when a term representing the attribute is detected, start the structuring process of the unit containing the term representing the attribute, and output a provisional result of the structuring process.

7. The one or more processors: Obtaining past medical documents created in the past, The medical information processing apparatus according to claim 1 , wherein the result of the structuring process on the acquired sentence is output in association with the result of the structuring process on the past medical text.

8. The medical information processing device described in claim 7, wherein the one or more processors add information that is described in the past medical text but is not described in the acquired medical text to the results of the structuring processing on the acquired sentence and output the result.

9. The medical information processing apparatus according to claim 1 , wherein the one or more processors associate a human body schema with the result of the structuring process and output the result.

10. The medical information processing device according to any one of claims 1 to 9, wherein the one or more processors determine a change in the attributes of the information contained in the acquired sentence as a change in a term representing a part of the human body as a break in the sentence.

11. The medical information processing device according to any one of claims 1 to 10, wherein the one or more processors determine a change in the attributes of the information contained in the acquired sentence as a break in the sentence when a term representing an organ of the human body is changed.

12. The medical information processing device according to claim 1 , wherein the one or more processors perform the structuring process including a term extraction process for acquiring terms in the units and determining the types of the acquired terms.

13. The medical image processing apparatus according to claim 12 , wherein the one or more processors perform the structuring process including a relationship determination process for determining a relationship between a plurality of terms extracted in the term extraction process.

14. The medical image processing apparatus according to claim 1 , wherein the one or more processors group the acquired sentences based on a result of the structuring process.

15. The medical image processing apparatus according to claim 14 , wherein the one or more processors perform the grouping based on relationships between a plurality of terms extracted in the structuring process.

16. The medical information processing apparatus according to claim 14 or 15, wherein the one or more processors output identification information that identifies the grouping result in the acquired sentence.

17. The medical information processing apparatus according to claim 14 , wherein the one or more processors acquire information indicating replacement of each grouped unit in the acquired sentence.

18. The medical image processing device according to claim 1 , wherein the one or more processors store the acquired sentence breaks.

19. The one or more processors: obtaining a correction input representing a correction of said sentence entered by a user; The medical image processing apparatus according to claim 18 , wherein, when the correction input is acquired, the division is updated in accordance with the correction.

20. The medical image processing apparatus according to claim 19 , wherein the one or more processors identify the range of the correction based on the stored delimiters.

21. The medical image processing device according to claim 19 or 20, wherein the one or more processors output a result of the structuring process of the corrected sentence.

22. A method for operating a medical information processing device including one or more processors and one or more memories that store programs to be executed by the one or more processors, comprising: The one or more processors Obtaining a sentence input by a user in response to said input; determining a change in the attribute of the information described in the sentence as a break in the sentence; Structuring process is carried out for each unit separated by a separator, The determination is terminated in response to completion of creation of a medical document including one or more of the sentences; A method for operating a medical image processing device that outputs the results of the structuring process.

23. On the computer, a function of acquiring a sentence input by a user in accordance with said input; a function of determining a change in the attribute of information described in the sentence as a break in the sentence; a function of performing structuring processing for each unit separated by the separator; A function of terminating the determination in response to completion of creation of a medical document containing one or more of the sentences; and A program that realizes the function of outputting the results of the structuring process.

Images