JP6972029B2 - Equipment for co-reference analysis, information extraction and similar document retrieval, similar document retrieval system and information processing method - Google Patents

Description

The present invention relates to natural language processing (NLP) and, in particular, to, for example, devices for co-reference resolution, information extraction and similar document retrieval , similar document retrieval systems and information processing methods.

Today, the use and management of electronic medical documents is becoming more and more widespread. Based on the management of electronic medical documents, many applications that may benefit physicians can be developed, such as similar medical document retrieval, diagnostic support, and so on. The above application is realized by executing a text information extraction technique on a medical document. The expression of the result obtained through the text information extraction technique is called a medical element (language element). In general, a medical element in a medical document may indicate several different medical objects, which may be a specific physical object (such as an abnormality diagnosed from a test result) or an abstract medical object. It can be a concept (eg, a disease as judged by a doctor). For example, some medical elements may indicate an abnormality (eg, a tumor) diagnosed from a test result, and some medical elements may indicate a patient's disease or the like. That is, with respect to one medical object (eg, a tumor) in a medical document, the physician can use several different expressions (ie, medical elements) to record.

Therefore, it is necessary to have a technique capable of determining whether different expressions (ie, two different medical elements) indicate the same medical object. In addition, co-reference analysis is one of the important techniques. In Patent Document 1, a similarity measure of two candidate elements is calculated based on the similarity of word features between these two candidate elements in a document and the context of the two candidate elements. A co-reference solution is disclosed. And if the similarity measure of these two candidate elements is greater than or equal to a predetermined threshold, these two candidate elements will be determined as co-references.

However, in medical documents, some word features that are superficially identical or similar to each other may not actually indicate the same medical subject. For example, the word feature "tumor" in a medical document may indicate different anomalies because different anomalies may be described by using the same expression (ie, word features). In addition, some word features that are not superficially similar to each other may actually indicate the same medical object. For example, the word features "lumbar fracture" and "fragile fracture" in medical documents may indicate the same abnormality. This is because different aspects of one anomaly are described using different expressions (ie, word features) so as not to confuse other readers. Therefore, the accuracy of the co-reference analysis that only uses the similarity of word features is low.

U.S. Pat. No. 8,457,950

Accordingly, in view of what has been described in detail above in the description of the related art, this disclosure is intended to solve the above problems.

According to one aspect of the invention, an apparatus for co-reference analysis is provided, the apparatus comprising an acquisition unit configured to acquire a first medical element and a second medical element from an input medical document. To detect the diagnostic state of the first medical element, at least one characteristic of the first medical element, the diagnostic state of the second medical element, and at least one characteristic of the second medical element from the medical document. A conformability determination unit configured to determine the compatibility between the first medical element and the second medical element based on the detected diagnostic condition and characteristics. It has a co-reference analysis unit configured to determine whether a first medical element and a second medical element indicate the same medical object based on the compatibility made. Here, the diagnostic state represents the position of the medical element in the diagnostic process in the medical document, the characteristic represents the diagnostic item of the medical element in the medical document, and the suitability represents the same second medical element as the medical element. Represents the possibility of indicating a medical subject.

When the present invention is used, the accuracy of the co-reference analysis is improved.

Further features and advantages of the present invention will become apparent from the following description with reference to the drawings.

The accompanying drawings are incorporated herein by way of the present invention, are part of the present invention, show embodiments of the present invention, and serve to explain the principles of the present invention together with the present specification.
FIG. 1 is a block diagram schematically showing a hardware configuration capable of realizing a technique according to an embodiment of the present invention. FIG. 2 is a block diagram showing a configuration of an apparatus for co-reference analysis according to the first embodiment of the present invention. FIG. 3 is a diagram showing an exemplary input medical document of the present invention according to the first embodiment of FIG. FIG. 4 is a flowchart schematically showing the procedure of the whole processing according to the first embodiment in FIG. FIG. 5 is a block diagram showing a configuration of an apparatus for co-reference analysis according to the second embodiment of the present invention. FIG. 6 is a flowchart schematically showing the procedure of the whole process according to the second embodiment in FIG. FIG. 7 is another block diagram showing the configuration of an apparatus for co-reference analysis according to the second embodiment of the present invention. FIG. 8 is a flowchart schematically showing the procedure of the whole process according to the second embodiment in FIG. 7. FIG. 9 is a block diagram showing a configuration of an apparatus for extracting information according to a third embodiment of the present invention. FIG. 10 is a flowchart schematically showing a procedure of the whole processing according to the third embodiment in FIG. FIG. 11 is another flowchart schematically showing the procedure of the whole process according to the third embodiment in FIG. FIG. 12 is a block diagram showing a configuration of an apparatus for searching for similar documents according to a fourth embodiment of the present invention. FIG. 13 is a flowchart schematically showing the procedure of the whole processing according to the fourth embodiment in FIG. FIG. 14 shows the configuration of an exemplary similar document retrieval system according to the present invention.

Hereinafter, preferred embodiments of the present invention will be described in detail with reference to the accompanying drawings. It should be noted that the following description is merely exemplary and exemplary in nature and is by no means intended to limit the invention and its uses or uses. The relative arrangements, mathematical formulas, and numerical values of the components and processes described in the embodiments do not limit the scope of the invention unless otherwise specified. Moreover, techniques, methods and devices known to those of skill in the art may not be discussed in detail, but are intended to be part of the specification where appropriate.

Similar reference numbers and letters refer to similar items in the figure, so once an item is defined in one figure, there is no need to discuss the following figure.

In medical diagnosis, one complete diagnostic process can include several diagnostic states. The diagnostic state is a discovery state that identifies medical findings (eg, normal and abnormal findings) from test results (eg, computer tomography (CT), magnetic resonance imaging (MRI), blood tests, etc.). A descriptive state that describes detailed information on the results of the examination, a suspicious state that makes the first suspicion based on medical findings, a comparative state that compares the current situation with the results of the previous survey, a reason state that analyzes the current reason, and the final state. Judgment states that make decisions, delayed decision states that record information that a final decision could not be made, treatment states that suggest treatments such as surgery or medicine, request states that require further testing to find out more information, and more. It is the diagnostic status of. Here, the diagnostic state represents the position of medical findings in the diagnostic process. In the field of NLP technology, the medical findings described above can also be regarded as a medical element. Here, a medical element (ie, a medical discovery) is an expression or record corresponding to a medical subject. The medical object may be a specific physical object (such as an abnormality diagnosed from a test result) or an abstract medical concept (such as a disease judged by a doctor).

When creating a medical document, the order of the above diagnostic states is not fixed for one diagnostic process, and not all of the above diagnostic states are required. In addition, in one medical document, the medical findings (ie, medical elements) of the patient diagnosed from the test results can be recorded, and the physician's considerations and / or judgments for that patient can be recorded. Other necessary treatments and / or tests for the patient can also be recorded. That is, the contents corresponding to any diagnostic state can be recorded in one medical document.

Further, in one diagnostic state, a plurality of different expressions can be recorded to describe detailed information (such as different embodiments) of one medical subject. For example, in the discovery state, a plurality of different expressions can be recorded to describe the detailed information of one tumor. In addition, a single medical subject may be recorded several times between different diagnostic states. For example, one expression can be recorded to describe the detailed information of one tumor in the discovery state, and one different expression can be recorded to describe the level of the tumor in the determination state. Another other expression can then be recorded to describe another requirement to continue testing the tumor for further diagnostic conditions. That is, one medical subject can be described by using several different representations (ie, several different medical elements) in one medical document, whether in one diagnostic state or in different diagnostic states.

As mentioned above, the structure of medical documents is complex. On the other hand, the present inventors have the same medical elements when there is a certain compatibility (compatibility) between the medical elements based on the description criteria of the medical document and these medical elements are compatible with each other. It was found that a medical subject should be indicated. In one example, if one of the different diagnostic states has not completed one diagnostic process, one medical subject already described in one preceding diagnostic state is still described in the other subsequent diagnostic states. Therefore, these descriptions (ie, medical elements) recorded between different diagnostic conditions are compatible and indicate the same medical subject. In another example, if the description of one medical subject is not completed in one diagnostic state, the other description of this medical subject may continue to be recorded in this diagnostic state. Thus, these statements recorded in this diagnostic condition are compatible and indicate the same medical subject. Of course, if the description of one medical subject is recorded in several diagnostic states but not throughout the diagnostic process, and / or the description in one diagnostic state is not complete, the other description of one medical subject is this. It may not be possible to continue recording in the diagnostic state.

On the other hand, the present invention is not intended to process the entire diagnostic process in one medical document, but to identify compatibility between two medical elements in an input medical document. Here, the input medical document can be a part of the medical document or the whole medical document. That is, in the present invention, if the description of the current medical element is not complete, or if the diagnostic state of the current medical element is not the final diagnostic state, the other medical elements may be compatible with the current medical element. There is sex.
[Hardware configuration]

First, with reference to FIG. 1, a hardware configuration capable of realizing the technique described below will be described. FIG. 1 is a block diagram schematically showing a hardware configuration 100 capable of realizing a technique according to an embodiment of the present invention.

The hardware configuration 100 includes, for example, a central processing unit (CPU) 110, a random access memory (RAM) 120, a read-only memory (ROM) 130, a hard disk 140, an input device 150, an output device 160, a network interface 170, and a system bus. Can include 180. Further, the hardware configuration 100 may be implemented by a personal data assistant (PDA), mobile phone, laptop, desktop, tablet computer, or other suitable electronic device.

The CPU 110 can be any suitable programmable control device, and can execute various functions described later by executing various application programs stored in the ROM 130 or the hard disk 140. The RAM 120 is used to temporarily store programs or data loaded from the ROM 130 or the hard disk 140, as well as disclosure techniques and other disclosure techniques described in detail below by the CPU 110 with reference to FIGS. 2-14. It is also used as a space for executing various programs such as executing available functions. The hard disk 140 can store many types of information, such as an operating system (OS), various applications, control programs, and data pre-stored or set by the manufacturer or user. Here, the data can be a historical medical document, threshold (TH), rule or model, eg, described below.

In one embodiment, the input device 150 can be an input interface, for example, receiving an image of a medical document output from an image acquisition device. Here, the image acquisition device can be, for example, a camera, a digital camera or other suitable electronic device. Further, the output device 160 can be used as an output interface, and the processing result can be output to a subsequent operation described later.

In another embodiment, the input device 150 interacts with an electronic device that implements the hardware configuration 100, such as allowing a user to input a medical element, a medical document, or an image of a medical document through the input device 150. Allows the user to act. Further, the input device 150 can take various forms such as a button, a keypad, a dial, a click wheel, or a touch screen. The output device 160 can include a cathode ray tube (CRT) or a liquid crystal display and can display the processing result to the user. Further, when the electronic device that implements the hardware configuration 100 is a device such as a so-called intelligent mobile phone, PDA, tablet computer, or other suitable electronic device, the input device 150 and the output device 160 are integrally incorporated. be able to. Further, when the electronic device that implements the hardware configuration 100 is a device such as a so-called intelligent mobile phone, PDA, tablet computer, or other suitable electronic device, the input device 150 and the output device 160 are integrally incorporated. be able to.

The network interface 170 provides an interface for connecting an electronic device (such as the electronic device 1410 shown in FIG. 14) that realizes the hardware configuration 100 to a network (such as the network 1420 shown in FIG. 14). For example, the electronic device that implements the hardware configuration 100 can execute data communication with another electronic device (such as the server 1430 shown in FIG. 14) connected via the network via the network interface 170. .. Alternatively, a wireless interface can be provided to an electronic device that implements the hardware configuration 100 to perform wireless data communication. The system bus 180 can provide a data transfer path for transferring data to and from the CPU 110, RAM 120, ROM 130, hard disk 140, input device 150, output device 160, network interface 170, and the like. Although referred to as a bus, the system bus 180 is not limited to any particular data transfer technique.

Software that provides the same functionality as the hardware configuration above can be used instead.

In the example of one embodiment of the present invention such as co-reference analysis, the program of the present embodiment described later by referring to FIGS. 4, 6 and 8 is installed in advance on the hard disk 140, and the CPU 110 is used as a book. It can also be loaded into the RAM 120 when the program of the embodiment needs to be executed. In another example, the program of this embodiment can be recorded in the ROM 130 as part of the memory map and executed directly by the CPU 110. Further, programs of other embodiments described later with reference to FIGS. 10 to 11 and 13 such as extraction of information and retrieval of similar documents can also be stored and executed by the same method.

The hardware configuration 100 described above is merely exemplary and is by no means intended to limit the invention, its use, or its use. Also, for simplicity, FIG. 1 shows only one hardware configuration. However, it is possible to use multiple hardware configurations as needed.
[Device configuration for co-reference analysis using suitability]

The configuration for co-reference analysis will be described below with reference to FIG. FIG. 2 is a block diagram showing a configuration of an apparatus 200 for co-reference analysis according to the first embodiment of the present invention.

The block shown in FIG. 2 is implemented as the CPU 110 described above with reference to FIG. 1 and is used to execute a program loaded in the RAM 120 and cooperate with each hardware shown in FIG. Some or all of the blocks may be implemented by dedicated hardware.

As shown in FIG. 2, the apparatus 200 for co-reference analysis according to the first embodiment of the present invention includes an acquisition unit 210, a diagnostic feature detection unit 220, a compatibility determination unit 230, and a co-reference analysis unit 240. To prepare for.

As mentioned above, firstly, the input device 150 shown in FIG. 1 will receive a medical document input by a user (eg, a doctor). Here, the medical document can be a part of the medical document or the whole medical document. Further, the input device 150 will receive the first medical element and the second medical element selected from the medical document by the user. Second, the input device 150 transfers the received medical document, the first medical element, and the second medical element to the acquisition unit 210 via the system bus 180.

As mentioned above, the first medical element and the second medical element are selected by the user. As an alternative solution, existing text information extraction techniques can also be used to extract them from the medical documents received by the CPU 110. For example, a plurality of medical elements can be extracted from the received medical document received by the CPU 110, and then any two medical elements can be regarded as the first medical element and the second medical element.

The above medical document is in text format. However, medical documents can also be in image format. For example, the input device 150 can receive an image of a medical document input by the user or output from the image acquisition device. After the input device 150 receives the image of the medical document, the input device 150 transfers the received image of the medical document to the CPU 110 via the system bus 180. The CPU 110 then converts the medical document from an image format to a text format, for example by using existing optical character recognition (OCR) technology. Then, the first medical element and the second medical element are selected from the converted medical document by the user via the input device 150, or are extracted from the converted medical document by the CPU 110 by using the text information extraction technique. Can be

In order to make the present invention easier to understand, an exemplary input medical document that is part of the medical document is shown in FIG. Here, medical documents are recorded, for example, in Japanese. As shown in FIG. 3, the term "nodule" (ie, "nodule" 310, "nodule" 320, "nodule" 330, and "nodule" 340) indicated by the dashed ellipse is the medical element described above. , Any two of these "nodules" can be considered as the first and second medical elements described above. Here, "nodule" means "noodle".

Returning to FIG. 2, first, the acquisition unit 210 acquires the first medical element, the second element, and the medical document from the input device 150 via the system bus 180.

Second, the diagnostic feature detection unit 220, from the medical document, indicates the diagnostic state of the first medical element, at least one characteristic of the first medical element, the diagnostic state of the second medical element, and the second medical element. Detects at least one characteristic of. Here, the diagnostic state represents the position of the medical element in the diagnostic process in the medical document, the characteristic represents the diagnostic item of the medical element in the medical document, and the diagnostic item represents the item of interest to the doctor for making the diagnosis. show.

As mentioned above, the diagnostic state includes at least one of the above-mentioned diagnostic states such as a discovery state, a description state, a suspicious state, a comparison state, a reason state, a delay determination state, a treatment state, and a request state. Then, the characteristic includes at least the type of the characteristic corresponding to the diagnostic item and the value of the characteristic corresponding to the parameter of the diagnostic item. Furthermore, the type of characteristic is at least one of a test index (size, shape, position, level, numerical index, etc.), a medical terminology (test, disease, treatment, drug, etc.), and a diagnostic assertion (polarity, cause, etc.). Including one. Here, the polarities are, for example, negative or positive poles, and the above causes represent, for example, that an abnormality is caused by lifestyle or trauma.

For example, taking the first medical element as an example, in order to identify which diagnostic state the first medical element belongs to, first, the diagnostic feature detection unit 220 relates to the first medical element from the medical document. It is possible to extract the predefined contents (contents) to be performed. The predefined content may be predefined by the manufacturer or user, for example according to actual use or experience, or according to the descriptive criteria for each diagnostic condition. The predefined content is content that can support the medical diagnosis in the medical document. Second, the diagnostic feature detection unit 220 can identify the diagnostic state of the first medical element by analyzing the extracted contents. For example, the diagnostic state of the first medical element can be identified by analyzing the extracted content according to a pre-generated rule or by classifying the extracted content according to a pre-generated model. Here, the pre-generated rules and pre-generated models can be generated based on a plurality of expression samples for each diagnostic state.

For example, taking the medical element "nodule" 310 shown in FIG. 3 as an example, the contents extracted by the diagnostic feature detection unit 220 are "peripheral", "diameter", "cm", and "recognized", and the diagnosis is made. The diagnostic state of the medical element "nodule" 310 identified by the feature detector 220 is the discovery state. Here, "peripheral" means "terminal", "diameter" means "diameter", "cm" is a measure of length, and "recognized" means "be found". ..

Then, in order to identify what kind of information is recorded about the first medical element, as an example, the diagnostic feature detection unit 220 may use an existing NLP such as template-based information extraction or training language corpus-based information extraction. According to the technique, the characteristics of the first medical element (such as the type of characteristic and / or the value of the characteristic) can be further extracted. For example, when the medical element "nodule" 310 shown in FIG. 3 is continuously taken up, the characteristics of the medical element "nodule" 310 extracted by the diagnostic feature detection unit 220 are "position: right lung S4" and "size: 2". .5 cm ”,“ polarity: anode ”. Here, the extracted "position", "size" and "polarity" are the types of characteristics, and the extracted "right lung S4", "2.5 cm" and "positive" are the values of the characteristics. ..

As mentioned above, the diagnostic state and the characteristics of the medical element are all detected by one unit, namely the diagnostic feature detector 220. However, diagnostic status and characteristics of medical elements can also be detected by different units. As shown in FIG. 2, as an optional solution, the diagnostic state identification unit 221 can identify the diagnostic state of the medical element, and the characteristic extraction unit 222 can extract the characteristics of the medical element.

Returning to FIG. 2, after the diagnostic feature detection unit 220 detects the diagnostic state of the first medical element, the characteristic of the first medical element, the diagnostic state of the second medical element, and the characteristic of the second medical element, The compatibility determination unit 230 determines (determines) the compatibility between the first medical element and the second medical element based on the detected diagnostic condition and characteristics. Here, compatibility represents the possibility that one medical element and another medical element indicate the same medical object. Then, as a suitable solution, the compatibility determination unit 230 may further include a compatibility coefficient determination unit 231 and a compatibility determination unit 232.

First, the suitability coefficient determination unit 231 determines the suitability coefficient between the first medical element and the second medical element based on the detected diagnostic state and characteristics. Here, the suitability factor represents the semantic conflict between the first medical element and the second medical element. As an example, the fitness factor is a semantic value between the diagnostic state of the first medical element, the characteristic of the first medical element, the diagnostic state of the second medical element, and the characteristic of the second medical element. Semantic sequence between the diagnostic state of the first medical element, the characteristic of the first medical element, the diagnostic state of the second medical element, and the characteristic of the second medical element, including the conflict of Includes conflicts.

For example, when the diagnostic state of the first medical element and the diagnostic state of the second medical element are in the same diagnostic state, the type of the characteristic of the first medical element and the type of the characteristic of the second medical element are different, and the first type is different. If the distance between the medical element and the second medical element is small, the semantic values between the first medical element and the second medical element do not conflict. Alternatively, if the diagnostic state of the first medical element and the diagnostic state of the second medical element are in different diagnostic states, then a particular type of characteristic of the first medical element and a particular type of second medical element. If the elements are the same, the values of these properties are the same, and the distance between the first medical element and the second medical element is small, then the first medical element and the second medical element Semantic values with do not conflict.

Further, when the diagnostic state of the first medical element and the diagnostic state of the second medical element are continuously connected, or the characteristics of the first medical element and the characteristics of the second medical element are continuous. If connected to, if the resulting semantic meanings do not conflict (ie, if the contiguous connections meet the descriptive criteria of the medical document), then between the first and second medical elements. The semantic order of does not conflict.

Then, after the suitability coefficient determination unit 231 determines the suitability coefficient between the first medical element and the second medical element, the suitability determination unit 232 determines the first suitability coefficient based on the determined suitability coefficient. The compatibility between the medical element of the above and the second medical element is determined. Further, the conformity determination unit 232 determines that the first medical element and the second medical element are compatible, for example, when the semantic values and the semantic order do not conflict with each other.

In one example, the suitability coefficient determination unit 231 can determine the suitability coefficient according to the following steps.

First, the fitness factor determination unit 231 will calculate at least one of the following features:

1) The distance between the first medical element and the second medical element. Here, this distance can be determined, for example, based on the distance of the sentence between the first medical element and the second medical element. And the smaller the distance, the more likely the first and second elements are to indicate the same medical object. For example, if the sentence related to the first medical element and the sentence related to the second medical element are adjacent to each other (that is, the distance is zero), the first medical element and the second medical element are the same medical object. Is likely to indicate.

2) The order between the state of the first medical element and the state of the second medical element.

3) The distance between the diagnostic state of the first medical element and the diagnostic state of the second medical element. Here, this distance can be determined, for example, based on the sentence distance between the diagnostic state of the first medical element and the diagnostic state of the second medical element. And the smaller the distance, the more likely the first and second elements are to indicate the same medical object. For example, if the diagnostic state of the first medical element and the diagnostic state of the second medical element are adjacent to each other (ie, the distance is zero), the first medical element and the second medical element indicate the same medical object. Probability is high.

4) Types of characteristics of the first medical element and the second medical element.

5) The order between the type of characteristic of the first medical element and the type of characteristic of the second medical element.

6) Value of the property whose type belongs to both the first medical element and the second medical element.

Second, the fitness factor determination unit 231 will determine semantic value conflicts and semantic order conflicts based on calculated features and predefined rules.

Through the descriptive criteria of medical documents, it can be seen that the medical elements in medical documents meet certain disciplines. In one example, if the two medical elements are in the same diagnostic state, if the two medical elements are compatible (ie, if the two medical elements indicate the same medical object). Except for the polar types mentioned above, the types of properties of these two medical elements should be different (ie, the content of these two medical elements is likely to be different aspects of the same medical subject). In another example, if the two medical elements are in the same diagnostic state, if the two medical elements are compatible (ie, if the two medical elements indicate the same medical subject). The characteristics of certain types of these two medical elements should be the same (eg, the positions of these two medical elements should be the same). Therefore, the predetermined rules differ from the conflicting conditions of characteristics when the first and second medical elements are in the same diagnostic state, and the first and second medical elements. It can at least include competing conditions of features when in diagnosis. In addition, predefined rules may be preset by the manufacturer or user, eg, according to experience or statistical training.

As an example, the medical element "nodule" 310 and the medical element "nodule" 320 shown in FIG. 3 are taken up, the medical element "nodule" 310 is regarded as the first medical element, and the medical element "nodule" 320 is regarded as the second medical element. Consider it as. As described above, the diagnostic feature detection unit 220 detects the diagnostic state and the diagnostic state and characteristics of the medical element “nodule” 310 and the medical element “nodule” 320, respectively, so that the compatibility coefficient determination unit 231 is for medical use. It can be easily determined that the element "nodule" 310 and the medical element "nodule" 320 are in the same diagnostic state. The characteristics of the medical element "nodule" 310 extracted by the diagnostic feature detection unit 220 are "position: right lung S4", "size: diameter 2.5 cm", and "polarity: positive". The characteristics of the medical element "nodule" 320 extracted by 220 are "shape: irregular" and "polarity: positive". Here, "irregular" means "irregular".

According to the characteristics of the medical element "nodule" 310 and the medical element "nodule" 320, the medical element "nodule" 310 and the medical element "nodule" 320 have one and the same kind of characteristic (ie, "polarity"). It is easy to see that the values of the same type of property are the same (ie, positive). Further, as shown in FIG. 3, since the sentence related to the medical element "nodule" 310 and the sentence related to the medical element "nodule" 320 are adjacent to each other, the medical element "nodule" 310 and the medical element The distance to the "nodule" 320 can be calculated as, for example, zero. Therefore, according to predefined rules, other types of properties of the medical element "nodule" 310 and the medical element "nodule" 320 are the type of polarity and the medical element "nodule" 310 and the medical element "nodule" 320. Since the distance between them is different except that the distance between them is very small, the fitness factor 231 determines that the semantic values between the medical element "nodule" 310 and the medical element "nodule" 320 do not conflict. Can be. Further, if the characteristic types of the medical element "nodule" 310 and the medical element "nodule" 320 are continuously connected (that is, "position" → "size" → "polarity" → "shape" → "polarity" ), The acquired semantic meanings do not conflict (ie, continuous connections meet the descriptive criteria of medical documents). Therefore, according to the predefined rules, the fitness factor determination unit 231 determines that the semantic order between the medical element "nodule" 310 and the medical element "nodule" 320 does not conflict.

Then, the conformity determination unit 232 determines that the medical element "nodule" 310 and the medical element "nodule" 320 are compatible because the semantic order and the semantic value between them do not conflict with each other.

Further, as another example, the medical element "nodule" 310 and the medical element "nodule" 330 shown in FIG. 3 are taken up, the medical element "nodule" 310 is regarded as the first medical element, and the medical element "nodule" 330 is regarded as the first medical element. Considered as a medical element of 2. According to the detected diagnostic states of the medical element "nodule" 310 and the medical element "nodule" 330, the suitability coefficient determination unit 231 has the medical element "nodule" 310 and the medical element "nodule" 330 in the same diagnostic state. That can be easily determined. The characteristics of the medical element "nodule" 330 extracted by the diagnostic feature detection unit 220 are "location: mediastinum", "size: 1 cm", and "polarity: positive". Here, "mediastinum" means "mediastinum".

According to the characteristics of the medical element "nodule" 310 and the medical element "nodule" 330, the medical element "nodule" 310 and the medical element "nodule" 330 have three same types of characteristics (ie, "position", "size"). And it is easy to see that it has "polarity"). Further, as shown in FIG. 3, since there are three sentences between the sentence related to the medical element "nodule" 310 and the sentence related to the medical element "nodule" 330, the medical element "nodule" 310 The distance between and the medical element "nodule" 330 can be calculated as, for example, 3. Therefore, according to the predefined rules, some types of characteristics of the medical element "nodule" 310 and the medical element nodule 330 are the same (eg "position" and "size") and the medical element "nodule". Since the distance between 310 and the medical element "nodule" 330 is not small, the suitability factor 231 competes for semantic values between the medical element "nodule" 310 and the medical element "nodule" 330. It can be judged that there is. Further, if the types of characteristics of the medical element "nodule" 310 and the medical element "nodule" 330 are continuously connected (that is, "position" → "size" → "polarity" → "shape" → "polarity" ), The acquired semantic meanings conflict (ie, continuous connections do not meet the descriptive criteria of medical documents). Therefore, according to the predefined rules, the fitness factor determination unit 231 determines that the semantic order between the medical element "nodule" 310 and the medical element "nodule" 330 conflicts.

Then, the conformity determination unit 232 determines that the medical element "nodule" 310 and the medical element "nodule" 330 are incompatible because the semantic order and the semantic value between them conflict with each other. ..

Further, as another example, the medical element "nodule" 310 and the medical element "nodule" 34 shown in FIG. 3 are taken up, the medical element "nodule" 310 is regarded as the first medical element, and the medical element "nodule" 340 is regarded as the first medical element. Considered as a second medical element. According to the detected diagnostic states of the medical element "nodule" 310 and the medical element "nodule" 340, the suitability coefficient determination unit 231 has the medical element "nodule" 310 and the medical element "nodule" 340 in different diagnostic states. That can be easily determined. Here, the medical element "nodule" 310 is in the discovered state and the medical element "nodule" 340 is in the comparative state. The characteristics of the medical "nodule" 340 extracted by the diagnostic feature detection unit 220 are "location: mediastinum", "target: previous time", and "tendency: increase", where "last time" is "last". "time" and "increase" mean "enlarge".

According to the characteristics of the medical element "nodule" 310 and the medical element "nodule" 340, the medical element "nodule" 310 and the medical element "nodule" 340 have one and the same kind of characteristic (ie, "polarity"). It is easy to see that the values of the same type of property are different. Further, as shown in FIG. 3, since there are five sentences between the sentence related to the medical element "nodule" 310 and the sentence related to the medical element "nodule" 340, the medical element "nodule" 310 The distance between and the medical element "nodule" 340 can be calculated, for example, 5. Therefore, according to the predefined rules, the medical element "nodule" 310 and the medical element "nodule" 340 have different values of the same type, and the medical element "nodule" 310 and the medical element "nodule" 340 have different values. Due to the large distance between them, the suitability coefficient determination unit 231 determines that the semantic values between the medical element "nodule" 310 and the medical element "nodule" 340 conflict. Furthermore, if the diagnostic states of the medical element "nodule" 310 and the medical element "nodule" 340 are continuously connected (ie, "discovery state" → "comparison state"), the acquired semantic meanings are in conflict. No (ie, the continuous connection meets the descriptive criteria of the medical document, or the distance between the diagnostic states of the medical element "nodule" 310 and the medical element "nodule" 340 is zero). Therefore, according to the predefined rules, the fitness factor determination unit 231 determines that the semantic order between the medical element "nodule" 310 and the medical element "nodule" 340 does not conflict.

Then, the conformity determination unit 232 determines that the medical element "nodule" 310 and the medical element "nodule" 340, which have conflicting semantic values between them, are not compatible.

Further, as another example, the medical element "nodule" 330 and the medical element "nodule" 340 shown in FIG. 3 are taken up, the medical element "nodule" 330 is regarded as the first medical element, and the medical element "nodule" 340 is regarded as the first medical element. Considered as a second medical element. According to the detected diagnostic states of the medical element "nodule" 330 and the medical element "nodule" 340, the suitability coefficient determination unit 231 has the medical element "nodule" 330 and the medical element "nodule" 340 in different diagnostic states. It can be easily determined. Here, the medical element "nodule" 330 is in the discovered state, and the medical element "nodule" 340 is in the comparative state.

According to the characteristics of the medical element "nodule" 330 and the medical element "nodule" 340, the medical element "nodule" 330 and the medical element "nodule" 340 have one and the same kind of characteristic (ie, "polarity"). It is easy to see that the values of the same type of property are the same (ie, positive). Further, as shown in FIG. 3, since there is one sentence between the sentence related to the medical element "nodule" 330 and the sentence related to the medical element "nodule" 340, the medical element "nodule" 330 The distance between and the medical element "nodule" 340 can be calculated as, for example, 1. Therefore, according to the predefined rules, the medical element "nodule" 330 and the medical element "nodule" 340 have the same characteristic value of the same type, with the medical element "nodule" 330 and the medical element "nodule" 340. Since the distance between them is small, the suitability coefficient determination unit 231 determines that the semantic values between the medical element "nodule" 330 and the medical element "nodule" 340 do not conflict. Furthermore, if the diagnostic states of the medical element "nodule" 330 and the medical element "nodule" 340 are continuously connected (ie, "discovery state" → "comparison state"), the acquired semantic meanings are in conflict. No (ie, the continuous connection meets the descriptive criteria of the medical document, or the distance between the diagnostic states of the medical element "nodule" 330 and the medical element "nodule" 340 is zero). Therefore, according to the predefined rules, the fitness factor determination unit 231 determines that the semantic order between the medical element "nodule" 330 and the medical element "nodule" 340 does not conflict.

Then, the conformity determination unit 232 determines that the medical element "nodule" 330 and the medical element "nodule" 340 are compatible because the semantic order and the semantic value between them do not conflict with each other. ..

Further, the above example determines the compatibility between the first medical element and the second medical element by determining whether the calculated features are competing according to a predefined rule. On the other hand, the compatibility score between the first medical element and the second medical element is calculated according to a predefined rule, based on the degree of competition of the calculated features. It can also be judged by. In one example, if the first and second medical elements are in the same diagnostic object, the fitness factor determination unit 231 may calculate the compatibility score between them according to the following equation.

は、第１および第２の医用要素の特性の種類の総数に対する同じ種類の特性の数の比率を表し、

は、値が第１および第２の医用要素の特性の種類の総数と異なる、同じ種類の特性の数の比率を表し、

は、第１および第２の医用要素における特性の種類の総数に対する、意味論的順序が異常である特性の種類の数の比率を表し、

は、診断状態にある文の総数に対する、第１の医用要素と第２の医用要素との間の文の距離の比率を表し、W_type、 W_value、 W_sequence、 W_{sentence distance}は、例えば経験に従って製造者またはユーザによって予め設定された所定の重みを表す。 here,

Represents the ratio of the number of characteristics of the same type to the total number of types of characteristics of the first and second medical elements.

Represents the ratio of the number of traits of the same type whose values differ from the total number of trait types of the first and second medical elements.

Represents the ratio of the number of trait types that are out of semantic order to the total number of trait types in the first and second medical elements.

Represents the ratio of the sentence distance between the first and second medical elements to the total number of sentences in the diagnostic state, where W _type , W _value , W _sequence , W _{sentence distance} are, for example, experience. Represents a predetermined weight preset by the manufacturer or user according to.

Then, the conformity determination unit 232 can determine that the first medical element and the second medical element are compatible when the calculated conformity score is equal to or higher than the predetermined threshold value. Here, the predefined thresholds may be preset by the manufacturer or user, for example, according to experience.

In another example, if the first and second medical elements are in different diagnostic targets, the fitness factor determination unit 231 may calculate the compatibility score between them according to the following equation.

は、第１および第２の医用要素の特性の種類の総数に対する、値が異なる同じ種類の特性の数の比率を表し、

は、診断プロセス全体における状態の総数に対する、第１および第２の医用要素第１および第２の医用要素において意味論的順序が異常である状態の数の比率を表し、

は、診断プロセス全体における文の総数に対する、第１の医用要素と第２の医用要素の間の文の距離の比率を表し、

は、診断プロセス全体における状態の総数に対する、第１の医用要素と第２の医用要素との間の状態距離の比率を表し、W_value、 W_sequence、 W_{sentence distance}、 W_{state distance}は、例えば経験に従って製造者またはユーザによって予め設定された所定の重みを表す。 here,

Represents the ratio of the number of traits of the same type with different values to the total number of trait types of the first and second medical elements.

Represents the ratio of the number of states in which the semantic order is out of order in the first and second medical elements to the total number of states in the entire diagnostic process.

Represents the ratio of the sentence distance between the first and second medical elements to the total number of sentences in the entire diagnostic process.

Represents the ratio of the state distance between the first and second medical elements to the total number of states in the entire diagnostic process, where W _value , W _sequence , W _{sentence distance} , W _{state distance} are, for example, experience. Represents a predetermined weight preset by the manufacturer or user according to.

Then, the conformity determination unit 232 may determine that the first medical element and the second medical element are compatible when the calculated conformity score is equal to or higher than the predetermined threshold value. Here, the predefined thresholds may be preset by the manufacturer or user, for example, according to experience.

As described above, the suitability coefficient determination unit 231 determines the suitability coefficient according to a predefined rule. As an alternative solution, the suitability factor determination unit 231 may also determine the suitability factor according to the following steps.

First, the fitness factor determination unit 231 will calculate at least one of the above features.

Second, the fitness factor determination unit 231 will determine semantic value competition and semantic order competition based on a pre-generated model. Here, the pre-generated model is a model for the characteristics when the first medical element and the second medical element are in the same diagnostic state, and the diagnosis that the first medical element and the second medical element are different from each other. Includes at least a model for features when in state. In addition, the pre-generated model can be pre-generated by the manufacturer, eg, according to statistical training. The detailed description is repeated here because the operation of the fitness factor determined based on the pre-generated model is similar to the above operation of the fitness factor determined based on the predefined rules. No.

Returning to FIG. 2, after the compatibility determination unit 230 determines the compatibility between the first medical element and the second medical element, the co-reference analysis unit 240 determines the compatibility between the first medical element and the second medical element. Whether or not the elements indicate the same medical object is determined based on the determined suitability. Further, the co-reference analysis unit 240 may determine that the first medical element and the second medical element indicate the same medical object when the determined compatibility is compatible.

Referring to the above example, as shown in FIG. 3, the medical element "nodule" 310 and the medical element nodule are compatible with each other, and the medical element "nodule" 310 and the medical element "nodule" 330 are not compatible with each other. , The medical element "nodule" 310 and the medical element "nodule" 340 are not compatible with each other, and the medical element "nodule" 330 and the medical element "nodule" 340 are compatible with each other. Therefore, in the co-reference analysis unit 240, the medical element "nodule" 310 and the medical element "nodule" 320 indicate one and the same nodule, and the medical element "nodule" 330 and the medical element "nodule" 340 are different and the same. Judged to indicate a nodule.

As mentioned above, the present invention determines whether two medical elements are co-referenced to each other based on the compatibility between them. Conformity can be explained as a set of features that meet certain constraints (such as the description criteria of a medical document) should not conflict with each other, otherwise the set of features are incompatible with each other. That is, if the features associated with the two medical elements do not conflict, the two medical elements are compatible (ie, these two medical elements indicate the same medical object). In addition, the invention semantically determines the conflict between features associated with two medical elements based on the descriptive criteria of the medical document, so that medical elements whose descriptions are superficially dissimilar to each other are also treated accurately. can do. Therefore, the present invention improves the accuracy of the co-reference analysis.
[Overall processing 1]

With reference to FIG. 4, the whole process executed by the configuration of the first embodiment in FIG. 2 will be described. FIG. 4 is a flowchart 400 schematically showing a procedure of the whole processing according to the first embodiment in FIG.

As mentioned above, first, the input device 150 shown in FIG. 1 receives a medical document input by the user or output from the image acquisition device. Here, the medical document can be a part of the medical document or the whole medical document. Further, when the medical document is in an image format, the CPU 110 receives the medical document from the input device 150 via the system bus 180, and first converts the medical document from the image format to the text format.

Further, the input device 150 receives the first medical element and the second medical element selected from the medical document by the user or extracted by the CPU 110 by using the existing text information extraction technique. Second, the input device 150 transfers the received medical document, the first medical element, and the second medical element to the acquisition unit 210 shown in FIG. 2 via the system bus 180.

Then, as shown in FIG. 4, in the acquisition step S410, the acquisition unit 210 acquires the first medical element, the second medical element, and the medical document from the input device 150 via the system bus 180.

In the diagnostic feature detection step S420, the diagnostic feature detection unit 220, from the medical document, describes the diagnostic state of the first medical element, at least one characteristic of the first medical element, the diagnostic state of the second medical element, and the second. Detects at least one characteristic of the medical element of.

In the suitability coefficient determination step S430, the suitability determination unit 230 determines the suitability coefficient between the first medical element and the second medical element based on the detected diagnostic condition and characteristics.

In the suitability determination step S440, the suitability determination unit 230 determines the suitability between the first medical element and the second medical element based on the determined suitability coefficient.

Then, when the first medical element and the second medical element are compatible, in the co-reference analysis step S450, the co-reference analysis unit 240 has the same medical element as the first medical element and the second medical element. Judge to indicate the target. Otherwise, in the co-reference analysis step S460, the co-reference analysis unit 240 determines that the first medical element and the second medical element do not point to the same medical object.

And finally, the co-reference resolution unit 240 resolves the co-reference via the system bus 180 for the subsequent operations described below, such as to display the processing result to the user or to extract information, etc. Is transferred to the output device 160 shown in FIG.
[Device configuration for co-reference analysis using similarity and suitability]

As described in the first embodiment, the apparatus 200 for semantic role analysis shown in FIG. 2 only uses the compatibility between the first medical element and the second medical element. Determine if the first medical element and the second medical element indicate the same medical object. However, the invention can also perform co-reference analysis by using a measure of similarity and suitability between a first medical element and a second medical element. In this embodiment, with reference to FIGS. 5 and 7, the configuration of a device for co-reference that performs co-reference analysis using a similarity measure and suitability will be described below. The apparatus for co-reference analysis of this embodiment has the same hardware configuration as that described in FIG.

FIG. 5 is a block diagram showing a configuration of an apparatus 500 for co-reference analysis according to the second embodiment of the present invention.

The block shown in FIG. 5 is realized as the CPU 110 described above with reference to FIG. 1 and is used to execute a program loaded in the RAM 120 and cooperate with each hardware shown in FIG. Some or all of the blocks may be implemented by dedicated hardware.

Comparing FIG. 5 with FIG. 2, there are two major differences in the apparatus 500 for semantic role analysis shown in FIG.

First, the device 500 for co-reference analysis further measures the similarity of word features between the first and second medical elements in the medical document and the content of the first and second medical elements in the medical document. It has a first similarity measure determination unit 510 used to determine a first similarity measure between a first medical element and a second medical element based on the similarity measure of. Here, the similarity measure between the first medical element and the second medical element represents the similarity between the first medical element and the second medical element. For example, if the similarity is very high, it means that the first medical element and the second medical element use the characteristics of almost the same word. And if the degree of similarity is high, either the first medical element and the second medical element use word features that are very similar, or the word features used in the first medical element and the second medical element. It means different or replaced by a synonym such as the word feature "failure". Further, the degree of similarity can be determined by using, for example, the above [Patent Document 1].

Secondly, in the diagnostic feature detection unit 220, the first similarity measure determination unit 510 has a threshold value (that is, the first similarity measure between the first medical element and the second medical element is shown in FIG. 6). Only when it is determined that the value is TH) or higher, the diagnostic state of the first medical element, the characteristics of the first medical element, the diagnostic state of the second medical element, and the characteristics of the second medical element are detected. .. Here, the threshold may be predefined by the manufacturer or user, for example according to actual use or experience. That is, in the present embodiment, the present invention uses the suitability between two medical elements to correct the inaccurate judgment caused by performing a collateral analysis using a similarity measure. Can be done. Here, an inaccurate judgment is determined by determining two medical elements as co-references, but these two medical elements are superficially the same or similar, but are not actually co-referenced to each other. be. Therefore, the accuracy of the co-reference analysis is improved.

Other detailed explanations regarding the acquisition unit 210, the diagnostic feature detection unit 220, the suitability determination unit 230, and the co-reference analysis unit 240 shown in FIG. 5 are the same as the corresponding parts shown in FIG. 2, and thus the details thereof. The explanation is not repeated here.

With reference to FIG. 6, the whole process executed by the configuration of the second embodiment in FIG. 5 will be described. FIG. 6 is a flowchart 600 schematically showing the procedure of the whole process according to the second embodiment in FIG.

Comparing FIG. 6 with FIG. 4, there are the following major differences in the flowchart 600 shown in FIG.

After the acquisition unit 210 acquires the first medical element, the second medical element, and the medical document from the input device 150 via the system bus 180 in the acquisition step S410, the first similarity measure determination step S610 is performed. The similarity measure determination unit 510 of 1 is based on the similarity measure of the word feature between the first and second medical elements in the medical document and the similarity measure of the contents of the first and second medical elements in the medical document. , Determine a first similarity measure between the first medical element and the second medical element. Further, in S610, in the first similarity measure determination unit 510, the first similarity measure between the first medical element and the second medical element is equal to or higher than the threshold value (that is, TH shown in FIG. 6). If it is determined, the procedure proceeds to step S420, and if not, the procedure ends.

Since the other detailed description for steps S410 to S460 shown in FIG. 6 is the same as the corresponding step shown in FIG. 4, the detailed description is not repeated here.

As mentioned above, the embodiment shown in FIG. 5 uses suitability to correct inaccurate judgments caused by using a similarity measure to perform a co-reference analysis. In addition, there is another solution for performing a collateral analysis by using a similarity measure and suitability. FIG. 7 is another block diagram showing the configuration of the apparatus 700 for co-reference analysis according to the second embodiment of the present invention.

The block shown in FIG. 7 is realized as the CPU 110 described above with reference to FIG. 1 and is used to execute a program loaded in the RAM 120 and cooperate with each hardware shown in FIG. Some or all of the blocks may be implemented by dedicated hardware.

Comparing FIG. 7 with FIG. 2, there are two major differences in the apparatus 700 for semantic role analysis shown in FIG.

The apparatus 700 for co-reference analysis further measures the similarity measure of the word feature between the first and second medical elements in the medical document and the similarity measure of the contents of the first and second medical elements in the medical document. Has a second similarity measure determinant 710, which is used to determine a second similarity measure between the first medical element and the second medical element. Here, the similarity measure between the first medical element and the second medical element can be determined by using, for example, the above-mentioned [Patent Document 1].

Second, the co-reference analysis unit 240 determines whether the first medical element and the second medical element indicate the same medical object based on the determined second similarity measure and the determined goodness of fit. to decide. For example, the co-reference analysis unit 240 determines when the determined second similarity measure and the determined goodness of fit are equal to or greater than the first threshold value (TH1), or when the determined second similarity measure is determined. If the goodness of fit is greater than or equal to the second threshold (TH2), it can be determined that the first medical element and the second medical element represent the same medical object. Here, TH1 and TH2 may be predefined by the manufacturer or user, for example according to actual use or experience.

As described in the first embodiment, the determined compatibility between the two medical elements can be recorded as a compatibility score, or is compatible (compatible) or non-compatible (non-conforming). Can be recorded as. Therefore, if it is recorded as conforming, the determined conformance may be considered as 1, and if it is recorded as non-conforming, the determined conformance may be considered as 0. Further, the weights of both the determined first similarity measure and the determined suitability can be set to 1, in which case the weighted sum described above, i.e., the determined first similarity. It is the sum of the sex measure and the determined suitability.

Other detailed explanations regarding the acquisition unit 210, the diagnostic feature detection unit 220, the suitability determination unit 230, and the co-reference analysis unit 240 shown in FIG. 7 are the same as the corresponding parts shown in FIG. 2, and thus the details thereof. The explanation is not repeated here. In this embodiment, for some medical elements whose description does not meet the criteria of a medical document, the present invention uses a similarity measure between the medical elements to perform a co-reference analysis using suitability. Inaccurate judgments caused can be corrected. Therefore, the accuracy of the co-reference analysis is improved.

With reference to FIG. 8, the whole process executed by the configuration of the second embodiment in FIG. 7 will be described. FIG. 8 is a flowchart 800 schematically showing the procedure of the whole process according to the second embodiment in FIG. 7.

Comparing FIG. 8 with FIG. 4, there are two main differences in the flowchart 800 shown in FIG.

First, in addition to determining the compatibility between the first medical element and the second medical element, the flowchart 800 further includes a second similarity measure determination step S810. In this step, the second similarity measure determination unit 710 measures the similarity of word features between the first and second medical elements in the medical document, and between the first and second medical elements in the medical document. Based on the content similarity measure, a second similarity measure between the first medical element and the second medical element is determined.

Second, the suitability determination unit 230 determines the suitability in steps S430 to S440, the second similarity measure determination unit 710 determines the second similarity measure in step S810, and then the co-reference analysis step S820. Then, the co-reference analysis unit 240 determines whether or not the first medical element and the second medical element indicate the same medical object based on the determined second similarity measure and the determined suitability. decide.

Since the other detailed description for steps S410 to S440 shown in FIG. 8 is the same as the corresponding step shown in FIG. 4, the detailed description is not repeated here.
[Device configuration for information extraction]

As described in the first and second embodiments, in order to extract information, the device 200 for co-reference analysis shown in FIG. 2, the device 500 for co-reference analysis shown in FIG. 5, and FIG. 7 The device 700 for the shown co-reference analysis can be used. In the present embodiment, next, with reference to FIG. 9, the configuration of the device for information extraction to which the devices 200, 500, and 700 for the above-mentioned co-reference analysis are applied will be described. The device for extracting information of this embodiment has the same hardware configuration as that described in FIG.

FIG. 9 is a block diagram showing the configuration of the device 900 for information extraction according to the third embodiment of the present invention.

The block shown in FIG. 9 is realized as the CPU 110 described above with reference to FIG. 1 and is used to execute a program loaded in the RAM 120 and cooperate with each hardware shown in FIG. Some or all of the blocks may be implemented by dedicated hardware.

As shown in FIG. 9, the apparatus 900 for information extraction according to the third embodiment of the present invention includes the above-mentioned apparatus 200, 500, or 700 for co-reference analysis, an acquisition unit 910, and a medical element extraction unit 920. , And a medical element integration unit 930.

First, as described in the first embodiment, the input device 150 shown in FIG. 1 receives a medical document input by the user or output from the image acquisition device. Here, the medical document can be a part of the medical document or the whole medical document. Further, when the medical document is in an image format, the CPU 110 receives the medical document from the input device 150 via the system bus 180, and first converts the medical document from the image format to the text format.

Second, as shown in FIG. 9, the acquisition unit 910 acquires a medical document from the input device 150 or the CPU 110 via the system bus 180.

Third, the medical element extraction unit 920 extracts at least two medical elements from the medical document acquired by the acquisition unit 910. Here, the medical element extraction unit 920 can use an existing text information extraction technique for extracting a medical element, for example.

And the device for co-reference analysis, which may be the device 200, 500, or 700 described above, indicates whether any two of the medical elements indicate the same medical object according to the above description with reference to FIGS. 2-8. To judge.

And finally, the medical element integration unit 930 integrates with the diagnostic state and characteristics of the mutually co-referenced medical elements determined by the device for co-reference. The medical element integration unit 930 then displays the system bus 180 through the integration results for subsequent operations as described below, such as displaying the processing results to the user or searching for similar documents. Transfer to the output device 160 shown in 1.

Thereby, according to the embodiment of the present information extraction, each of the medical objects in the acquired medical document can be acquired as one descriptive data.

In order to obtain more accurate descriptive data for the medical object, as a preferred solution, the medical element integration unit 930 is based on the appearance order (sequence) of the medical elements in the acquired medical document, and the medical elements are co-referenced with each other. Diagnostic status and characteristics can be integrated. Further, the medical element integration unit 930 can carry out the integration by various methods described below with reference to FIGS. 10 to 11.
[Overall processing 2]

With reference to FIGS. 10 to 11, the entire process executed by the configuration of the third embodiment in FIG. 9 will be described. FIG. 10 is a flowchart 1000 schematically showing a procedure of the whole processing according to the third embodiment of FIG.

As mentioned above, first, the input device 150 shown in FIG. 1 receives a medical document input by the user or output from the image acquisition device. Here, the medical document can be a part of the medical document or the whole medical document. Further, when the medical document is in an image format, the CPU 110 receives the medical document from the input device 150 via the system bus 180, and first converts the medical document from the image format to the text format.

Then, as shown in FIG. 10, in the acquisition step S1010, the acquisition unit 910 acquires a medical document from the input device 150 or the CPU 110 via the system bus 180.

In the medical element extraction step S1020, the medical element extraction unit 920 extracts at least two medical elements from the medical document acquired by the acquisition unit 910.

In co-reference analysis steps S103 to S1040, first in step S1030, the device for co-reference analysis, which may be the device 200, 500 or 700 described above, is the extracted medical element that first appears in the obtained medical document. , Select as the first medical element to appear. Then, in step S1040, the device for co-reference analysis, which may be the device 200, 500, or 700 described above, is co-referenced with the first appearing medical element according to the above description with reference to FIGS. 2-8. Determine the medical factors that are in use.

In the medical element integration step S1050, the medical element integration unit 930 integrates the diagnostic state and characteristics of the first appearing medical element and the medical element co-referenced with the first appearing medical element.

Then, in step S1060, the medical element integration unit 930 determines whether or not the extracted medical element remains. Then, with respect to the remaining medical elements, if the medical elements still remain, the operation executed in steps S103 to S1050 is repeated. Otherwise, the procedure ends and descriptive data for each medical subject is displayed to the user or for subsequent operations.

In the following, any solution will be described with reference to FIG. FIG. 11 is another flowchart 1100 schematically showing the procedure of the whole process according to the third embodiment in FIG.

As shown in FIG. 11, since the detailed description of steps S101 to S1030 and S1060 shown in FIG. 11 is the same as the corresponding step shown in FIG. 10, the detailed description is not repeated here.

After the first appearing medical element in step S1030 is selected, in step S1110 the device for semantic role analysis, which may be the device 200, 500 or 700 described above, is described above with reference to FIGS. 2-8. Follow the instructions to determine the medical element that is co-referenced with the closest and first appearing medical element.

In step S1120, the medical element integration unit 930 integrates the diagnostic state and characteristics of the first appearing medical element and the medical element determined in step S1110.

In step S1130, with respect to the integrated medical element, the device for co-reference analysis, which may be the device 200, 500 or 700 described above, first appeared in accordance with the description described above with reference to FIGS. 2-8. Determine if a medical element that is closest to and co-referenced with the element is included. And if the integrated medical element still contains the closest and co-referenced medical element, the procedure proceeds to step S1140. If not, the process proceeds to step S1060.

In step S1140, the medical element integration unit 930 integrates the diagnostic state and characteristics of the medical element determined in step S1130 into the integrated medical element. Then, the procedure repeats the operation performed in steps S1130 to S1140 until there is no medical element closest to and co-referenced with the integrated medical element.

As will be appreciated by those skilled in the art, the above-mentioned integrated operations described with reference to FIGS. 10 to 11 are merely exemplary and not limiting.

As described above, taking the above-mentioned medical element "nodule" 310-340 shown in FIG. 3 as an example, the medical element "nodule" 310 and the medical element "nodule" 320 indicate one identical nodule and are medical. The element "nodule" 330 and the medical element nodule indicate another identical nodule. Thus, the diagnostic states and characteristics of the medical element "nodule" 310 and the medical element "nodule" 320 can be integrated into one descriptive data relating to the medical subject, the corresponding descriptive data being shown, for example, in Table 1 below. .. Further, the diagnostic status and characteristics of the medical element "nodule" 330 and the medical element "nodule" 340 can be integrated into different descriptive data for another medical subject, the corresponding descriptive data can be found in, for example, Table 2 below. Shown.

Table 1:

［類似文書検索のための装置の構成］ Table 2:

[Device configuration for similar document retrieval]

As described in the third embodiment, the information extraction device 900 shown in FIG. 9 can be used for similar document retrieval. Here, if the two documents are the same document, it means that the contents of these two documents are almost the same. And if the two documents are similar, it means that there is considerable overlap of conceptual content between the two documents. In the present embodiment, next, with reference to FIG. 12, the configuration of the device for searching for similar documents to which the above-mentioned information extraction device 900 is applied will be described. The apparatus for searching for similar documents according to this embodiment has the same hardware configuration as that described in FIG.

FIG. 12 is a block diagram showing the configuration of the apparatus 1200 for searching for similar documents according to the fourth embodiment of the present invention.

The block shown in FIG. 12 is implemented as the CPU 110 described above with reference to FIG. 1 and is used to execute a program loaded in the RAM 120 and cooperate with each hardware shown in FIG. Some or all of the blocks may be implemented by dedicated hardware.

As shown in FIG. 12, the apparatus 1200 for searching for similar documents according to the fourth embodiment of the present invention includes the above-mentioned device 900 for extracting information, a similarity measure calculation unit 1210, and a similar document search unit 1220. To prepare for.

When a user (eg, a doctor) writes or reads a medical document, he / she usually wants to search for and refer to a historical medical document that is similar to the medical document he or she wrote or read.

Thus, first, the input device 150 shown in FIG. 1 will receive a medical document written or read by the user. Here, the medical document can be directly input by the user or output from the image acquisition device, and the document can be a part of the medical document or the entire medical document.

With respect to the medical document received by the input device 150, the device 900 for information extraction is an integrated diagnostic state of the medical element from the obtained medical document according to the above description with reference to FIGS. 9-11. And the characteristics (ie, the above descriptive data for each medical subject) are extracted.

On the other hand, in one example, the CPU 110 shown in FIG. 1 acquires a history medical document stored in the ROM 130 or the hard disk 140. In another example, the CPU 110 acquires a history medical document stored in a server connected to a device 1200 for searching for similar documents via a network. A detailed description will be described later with reference to FIG. Then, for each of the historical medical documents, the device 900 for information extraction provides the integrated diagnostic state and characteristics of the medical elements from the corresponding historical medical document according to the above description with reference to FIGS. 9-11. Extract.

Then, as shown in FIG. 12, the similarity measure calculation unit 1210 integrates the integrated diagnostic state and characteristics of the medical elements extracted from the acquired medical document and the medical elements extracted from the history medical document. Calculate a measure of similarity between the diagnosed state and the characteristic. Here, the similarity measure can be, for example, an edit distance.

Finally, the similarity document search unit 1220 searches for at least one medical document similar to the medical document obtained from the history medical document based on the calculated similarity measure. In one example, the similarity document search unit 1220 can search for one medical document corresponding to the largest similarity measure as the final processing result. In another example, the similarity document search unit 1220 can search for a particular medical document ranked based on the calculated similarity measure value.

Then, the similar document search unit 1220, an output device showing similar medical documents searched via the system bus 180 for subsequent operations such as displaying the processing result to the user or diagnostic support, is shown in FIG. Transfer to 160.

As a result, according to this embodiment of the similar document search, historical medical documents that are superficially similar but not actually similar to the medical document entered by the user are not searched and are entered by the user. History medical documents that are actually similar to but not superficially similar to the medical documents are searched. That is, a historical medical document whose concept content has a significant overlap with a medical document input by the user and / or a historical medical document describing a subject similar to the medical document input by the user is searched according to the present invention. Become. Therefore, the accuracy of similar document retrieval is improved.
[Overall processing 3]

With reference to FIG. 13, the whole process executed by the configuration of the fourth embodiment in FIG. 12 will be described. FIG. 13 is a flowchart 1300 schematically showing the procedure of the whole processing according to the fourth embodiment of FIG.

As described above, for the medical document input by the user and each history medical document stored in the ROM 130 or the hard disk 140 or the server, in the information extraction step S1310, according to the above description with reference to FIGS. 9 to 11. The device 900 for information extraction extracts the integrated diagnostic object and characteristics of the medical element from the acquired medical document, and extracts the integrated diagnostic state and characteristic of the medical element from the historical medical document.

In the similarity measure calculation step S1320, the similarity measure calculation unit 1210 integrates the integrated diagnostic state and characteristics of the medical elements extracted from the acquired medical document and the medical elements extracted from the history medical document. Calculate a measure of similarity between diagnostic status and characteristics.

In the similar document search step S1330, the similar document search unit 1220 searches for at least one medical document similar to the medical document obtained from the historical medical document based on the calculated similarity measure.

Then, in order to display the processing result to the user or to perform a subsequent operation, the similar document search unit 1220 displays the similar medical document searched via the system bus 180 on the output device 160 shown in FIG. Forward.
[Similar document search system]

As described in the fourth embodiment, the historical medical document may be stored in a server connected to the similar document retrieval device 1200 via a network. In the present embodiment, next, an exemplary similar document retrieval system 1400 to which the above-mentioned device 1200 for similar document retrieval is applied will be described with reference to FIG. 14. FIG. 14 shows the configuration of an exemplary similar document retrieval system 1400 according to the present invention.

As shown in FIG. 14, the similar document retrieval system 1400 includes an electronic device 1410 as the above-mentioned device for searching similar documents and a server 1430 used for storing historical medical documents and other historical data. Be prepared. The electronic device 1410 of this embodiment can have the same hardware configuration as that shown in FIG. 1 and the same configuration as that shown in FIG.

The electronic device 1410 is configured to acquire historical medical documents from the server 1430 via the network 1420. Then, the electronic device 1410 is configured to search for a medical document similar to the medical document input by the user from the acquired historical medical documents according to the above description with reference to FIGS. 12 to 13. Further, the electronic device 1410 may be implemented by a personal data assistant (PDA), mobile phone, laptop, desktop, tablet computer, or other suitable electronic device. As shown in FIG. 14, the electronic device 1410 is, for example, a laptop (ie, a personal computer).

All of the above units are exemplary and / or preferred modules for performing the processes described in this disclosure. These units can be hardware units (such as field programmable gate arrays (FPGAs)), digital signal processors, application-specific integrated circuits, and / or software modules (such as computer-readable programs). The units for performing the various steps are not exhaustively listed above. However, if there is a step to perform a particular process, there may be a corresponding functional module or unit (performed by hardware and / or software) to perform the same process. Technical solutions with the steps described and all combinations of units corresponding to these steps are included in the disclosure of this application as long as the technical solutions they constitute are complete and applicable.

Further, the software is 200 for co-reference analysis shown in FIGS. 2, 5 and 7, 900 for information extraction shown in FIG. 9, or 1200 for similar document retrieval shown in FIG. If configured partially or entirely by, it may be stored on the hard disk 140 shown in FIG. In another embodiment, the device for co-reference analysis shown in FIGS. 2, 5, and 7, 900, the device 900 for information extraction shown in FIG. 9, or the device for searching similar documents shown in FIG. If the 1200 is partially or wholly configured by hardware or firmware, it is a functional module as long as there is a need for co-reference analysis, a need for information extraction, or a need for similar document retrieval in an electronic device. It can also be incorporated into an electronic device such as a computer.

It is possible to implement the methods and devices of the present invention in many ways. For example, the methods and devices of the invention can be implemented through hardware, firmware, or any combination thereof. The order of the steps in the above method is merely exemplary, and the steps in the method of the present invention are not limited to the specific order described above, unless otherwise specified. Further, in some embodiments, the invention can also be implemented as a program recorded on a recording medium, including machine-readable instructions for carrying out the method according to the invention. As described above, the present invention also includes a recording medium on which a program for carrying out the method of the present invention is recorded.

Although some particular embodiments of the invention have been described in detail by way of examples, those skilled in the art will appreciate that the above examples are merely exemplary and do not limit the scope of the invention. sea bream. It will be appreciated by those skilled in the art that the above embodiments can be modified without departing from the scope and spirit of the invention. The scope of the present invention is defined by the appended claims.

This application claims the benefit of Chinese Patent Application No. 20160428860.4 filed June 16, 2016, which is incorporated herein by reference in its entirety.

Claims (21)

A device for co-reference analysis
An acquisition means configured to acquire the first and second medical elements from the input medical document, and
From the medical document, the diagnostic state of the first medical element, the diagnostic state of the second medical element, at least one characteristic of the first medical element, and at least one characteristic of the second medical element. Diagnostic feature detection means configured to detect,
A compatibility determining means configured to determine compatibility between the first medical element and the second medical element based on the detected diagnostic condition and characteristics.
It has a co-reference analysis means configured to determine whether the first medical element and the second medical element indicate the same medical object based on the determined suitability.
The diagnostic status represents the position of the medical element in the diagnostic process in the medical document.
A characteristic represents a diagnostic item of a medical element in a medical document.
Compatibility is a device that represents the possibility that one medical element and another medical element indicate the same medical object. The device according to claim 1, wherein the diagnostic feature detecting means is for one medical element.
Extracting predefined content related to the medical element from the medical document
A device that identifies the diagnostic state of the medical element by analyzing the extracted content. The apparatus according to claim 1, wherein the conformity determining means is used.
A suitability coefficient determining means configured to determine a suitability coefficient between the first medical element and the second medical element based on the detected diagnostic condition and characteristics. The fitness factor represents a means and means that represent the semantic conflict between the first and second medical elements.
An apparatus having a compatibility determining means configured to determine compatibility between the first medical element and the second medical element based on the determined compatibility coefficient. The device according to claim 3, wherein the compatibility coefficient is
Semantic values between the diagnostic state of the first medical element, the characteristic of the first medical element, the diagnostic state of the second medical element, and the characteristic of the second medical element. Conflict and
The semantic order between the diagnostic state of the first medical element, the characteristic of the first medical element, the diagnostic state of the second medical element, and the characteristic of the second medical element. Equipment, including conflicts. The apparatus according to claim 4, wherein the suitability coefficient determining means is used.
The distance between the first medical element and the second medical element,
The sequence between the diagnostic state of the first medical element and the diagnostic state of the second medical element,
The distance between the diagnostic state of the first medical element and the diagnostic state of the second medical element,
The type of property of the first medical element and the second medical element,
The order between the type of the property of the first medical element and the type of the property of the second medical element,
At least one of the features of the property, of which the type belongs to both the first medical element and the second medical element, is calculated.
A device for determining said conflicts of said semantic values and said conflicts of said semantic order based on the calculated features and predefined rules. The apparatus according to claim 4, wherein the suitability coefficient determining means is used.
The distance between the first medical element and the second medical element,
The sequence between the diagnostic state of the first medical element and the diagnostic state of the second medical element,
The distance between the diagnostic state of the first medical element and the diagnostic state of the second medical element,
The type of property of the first medical element and the second medical element,
The order between the type of the property of the first medical element and the type of the property of the second medical element,
At least one of the features of the property, of which the type belongs to both the first medical element and the second medical element, is calculated.
A device for determining said competition of the semantic values and said competition of the semantic conflicts based on the calculated features and a pre-generated model. In the apparatus according to claim 4, when the semantic value and the semantic order do not conflict with each other, the first medical element and the second medical element are compatible with each other. A device that determines that there is. The apparatus according to claim 1, wherein the first medical element and the second medical element are the same medical element when the determined compatibility is compatible. A device that determines to indicate an object. The device according to claim 1, further
The first is based on a measure of similarity of word features between the first and second medical elements in the medical document and a measure of similarity of the contents of the first and second medical elements in the medical document. Has a first similarity measure determining means configured to determine a first similarity measure between the medical element of the device and the second medical element.
When the first similarity measure determining means determines that the first similarity measure between the first medical element and the second medical element is equal to or greater than a threshold value, the first medical element is used. An apparatus for detecting the diagnostic state of an element, the characteristic of the first medical element, the diagnostic state of the second medical element, and the characteristic of the second medical element. The device according to claim 1, further
The first is based on a measure of similarity of word features between the first and second medical elements in the medical document and a measure of similarity of the context of the first and second medical elements in the medical document. Has a second similarity measure determining means configured to determine a second similarity measure between the medical element and said second medical element.
Does the co-reference analysis means indicate that the first medical element and the second medical element indicate the same medical object based on the determined second similarity measure and the determined compatibility? The device to judge. A device for extracting information
An acquisition method configured to acquire medical documents,
A medical element extraction means configured to extract at least two medical elements from the obtained medical document, and a medical element extraction means.
The apparatus for co-reference analysis according to any one of claims 1 to 10, configured to determine if any two of the medical elements represent the same medical object.
An apparatus having a medical element integration means configured to integrate the diagnostic state and the characteristic of the medical element that are co-referenced to each other. The device according to claim 11, wherein the medical element integrating means has the diagnostic state of the medical element and said, which are co-referenced to each other, based on the order of appearance of the medical element in the acquired medical document. A device that integrates characteristics. A device for searching similar documents
A claim configured to extract the integrated diagnostic state and characteristics of the medical element from the acquired medical document and the integrated diagnostic state and characteristics of the medical element from the history medical document. The apparatus for extracting information according to item 11 or 12, and the apparatus.
Similarities between the integrated diagnostic state and characteristics of the medical element extracted from the acquired medical document and the integrated diagnostic state and characteristics of the medical element extracted from the history medical document. A similarity measure calculation means configured to calculate a measure,
A device comprising a similar document search means configured to search for at least one medical document similar to the acquired medical document from the history medical document based on the calculated similarity measure. A similar document retrieval system composed of at least one server and a device for searching similar documents connected to the server via a network.
The server is configured to store historical medical documents.
The device for searching for similar documents is
A claim configured to extract the integrated diagnostic state and characteristics of the medical element from the acquired medical document and the integrated diagnostic state and characteristics of the medical element from the history medical document. The apparatus for extracting information according to item 11 or 12, and the apparatus.
Similarities between the integrated diagnostic state and characteristics of the medical element extracted from the acquired medical document and the integrated diagnostic state and characteristics of the medical element extracted from the history medical document. A similarity measure calculation means configured to calculate a measure,
A similar document search system comprising a similar document search means configured to search at least one medical document similar to the acquired medical document from the historical medical document based on the calculated similarity measure. .. It is an information processing method for co-reference analysis performed by a computer under the control of software.
The acquisition process of acquiring the first medical element and the second medical element from the input medical document, and the acquisition process.
From the medical document, the diagnostic feature detecting means is a diagnostic state of the first medical element, at least one characteristic of the first medical element, a diagnostic state of the second medical element, and at least one of the second medical elements. Diagnostic feature detection process to detect characteristics and
A compatibility determination step in which the compatibility determination means determines the compatibility between the first medical element and the second medical element based on the detected diagnostic condition and characteristics.
The co-reference analysis means includes a co-reference analysis step of determining whether the first medical element and the second medical element indicate the same medical object based on the determined compatibility.
The diagnostic status represents the position of the medical element in the diagnostic process in the medical document.
A characteristic represents a diagnostic item of a medical element in a medical document.
Compatibility is an information processing method that represents the possibility that one medical element and another medical element indicate the same medical object. The information processing method according to claim 15, wherein the conformity determination step is performed.
A compatibility coefficient determination step for determining a compatibility coefficient between the first medical element and the second medical element based on the detected diagnostic condition and characteristics, wherein the compatibility coefficient is a method. A process that represents a semantic conflict between a first medical element and a second medical element,
An information processing method comprising a suitability determination step of determining the suitability between the first medical element and the second medical element based on the determined suitability coefficient. The information processing method according to claim 15, wherein the first medical element and the second medical element are the same when the determined compatibility is applicable in the co-reference analysis step. An information processing method that is determined to indicate a medical subject. The information processing method according to claim 15, further.
The first similarity measure determining means is based on the similarity measure of the word feature between the first and second medical elements in the medical document, and the first and second in the medical document. It comprises a first similarity rate determination step of determining a first similarity measure between the first medical element and the second medical element based on the content similarity measure of the medical element. When it is determined in the diagnostic feature detection step that the first similarity measure between the first medical element and the second medical element is equal to or greater than the threshold in the first similarity determination step. , said diagnostic state of the first medical element, wherein the characteristic of the first medical element, said diagnostic state of the second medical element, and the characteristics of the second medical element is detected, the information processing Method. The information processing method according to claim 15, further.
The second similarity measure determining means is based on the similarity measure of the word feature between the first and second medical elements in the medical document, and the first and second medical use in the medical document. A second similarity measure determining step of determining a second similarity measure between the first medical element and the second medical element based on the content similarity measure of the element is included. In the reference analysis step, whether the first medical element and the second medical element indicate the same medical object based on the determined second similarity measure and the determined compatibility. Information processing method for which is determined. It is an information processing method for information extraction performed by a computer under the control of software.
The acquisition method is the acquisition process for acquiring medical documents,
A medical element extraction step in which the medical element extraction means extracts at least two medical elements from the acquired medical document, and a medical element extraction step.
A co-reference analysis step of determining whether any two of the medical elements indicate the same medical object by using the information processing method according to any one of claims 15 to 19 as the co-reference analysis means.
An information processing method comprising a medical element integration step configured such that the medical element integration means integrates the diagnostic state and the characteristic of the medical element that are co-referenced to each other. It is an information processing method for searching similar documents performed by a computer under the control of software.
By using the information processing method according to claim 20 , the information extraction means extracts the integrated diagnostic state and characteristics of the medical element from the acquired medical document, and extracts the integrated diagnostic state and characteristics of the medical element from the history medical document. Information extraction process to extract integrated diagnostic status and characteristics,
The similarity rate calculation means has the integrated diagnostic state and characteristics of the medical element extracted from the acquired medical document and the integrated diagnostic state of the medical element extracted from the historical medical document. The similarity rate calculation process, which calculates the similarity measure between the characteristics, and
An information processing method comprising a similar document search step of searching for at least one medical document similar to the acquired medical document from the history medical document based on the calculated similarity measure. ..

Images