Electronic medical record processing method, device, storage medium and electronic equipment

Abstract

The present disclosure relates to a method, device, storage medium and electronic device for processing electronic medical records to solve the problem of inaccurate electronic medical record recognition. The processing method includes: based on the separator in the electronic medical record, the electronic medical record is divided into multiple medical record short sentences, the symptom entities in the multiple medical record short sentences and the entity attributes corresponding to the symptom entities are obtained, for each of the multiple medical record short sentences, the entity attributes corresponding to the target symptom entity in the medical record short sentence are shared with at least one other symptom entity in the same medical record short sentence, and multiple target medical record short sentences are obtained, and based on the multiple target medical record short sentences, a target electronic medical record is obtained. Thus, each symptom entity corresponds to an entity attribute, and a structured target electronic medical record is obtained by output, thereby improving the accuracy of electronic medical record recognition.

Description

Technical Field

[0001] The present disclosure relates to the field of data processing, and in particular, to a method, device, storage medium and electronic device for processing electronic medical records. Background Art

[0002] In the prior art, electronic medical record structuring is usually achieved by named entity recognition and attribute recognition. Sequence labeling is used to label entities in electronic medical records. First, the entities in the text are labeled, and then the attributes of the entities are labeled. However, in the process of medical record sequence labeling, due to the non-standard writing of medical records, it is often encountered that only some entities have attributes, which seriously affects the effect of the named entity recognition model, and thus the electronic medical record information obtained by recognition cannot accurately reflect the entities and corresponding attributes. Summary of the invention

[0003] The purpose of the present disclosure is to provide an electronic medical record processing method, device, storage medium and electronic device to solve the problem of inaccurate electronic medical record information recognition in related technologies, resulting in some entities having no corresponding attributes or incorrect entity attributes.

[0004] In order to achieve the above object, the present disclosure provides a method for processing electronic medical records, the method comprising:

[0005] Based on the separator in the electronic medical record, the electronic medical record is divided into a plurality of medical record short sentences;

[0006] Acquire symptom entities in the plurality of medical record short sentences, and entity attributes corresponding to the symptom entities;

[0007] For each of the multiple medical record sentences, sharing the entity attribute corresponding to the target symptom entity in the medical record sentence with at least one other symptom entity in the same medical record sentence, to obtain multiple target medical record sentences;

[0008] Based on the multiple target medical record sentences, the target electronic medical record is obtained.

[0009] Optionally, the entity attribute includes a yin-yang sub-attribute and / or a degree sub-attribute, and for each of the multiple medical record short sentences, the entity attribute corresponding to the target symptom entity in the medical record short sentence is shared with at least one other symptom entity in the same medical record short sentence to obtain multiple target medical record short sentences, and the multiple target medical record short sentences are obtained, including:

[0010] Based on the position of the first symptom entity in the first medical record short sentence, the yin and yang sub-attributes of the first symptom entity are shared with at least one second symptom entity in the first medical record short sentence in a first direction, and the first medical record short sentence is any one of the multiple medical record short sentences; and / or,

[0011] Sharing the degree sub-attribute of the first symptom entity with at least one third symptom entity in the first medical history sentence in a second direction, wherein the second direction is opposite to the first direction;

[0012] Based on multiple symptom entities obtained after sharing the yin-yang sub-attributes and / or degree sub-attributes, a target medical record short sentence corresponding to the first case short sentence is obtained.

[0013] Optionally, before sharing the yin and yang sub-attributes of the first symptom entity with at least one second symptom entity in the first medical history sentence in a first direction based on the position of the first symptom entity in the first medical history sentence, the method further includes:

[0014] identifying the yin and yang sub-attributes in the plurality of medical record short sentences;

[0015] If multiple yin and yang sub-attributes are identified in the second medical record sentence, the second medical record sentence is divided into multiple medical record sentences based on the positions of the symptom entities corresponding to the multiple yin and yang sub-attributes in the second medical record sentence, and the second medical record is any one of the multiple medical record sentences divided from the electronic medical record.

[0016] Optionally, after sharing the degree sub-attribute of the first symptom entity with at least one third symptom entity in the first medical history sentence in the second direction, the method further includes:

[0017] If during the sharing process the third entity attribute corresponding to the third symptom entity and the degree sub-attribute of the first symptom entity are mutually exclusive, the sharing of the degree sub-attribute of the first symptom entity in the second direction is stopped starting from the third symptom entity.

[0018] Optionally, for each of the multiple medical record sentences, sharing the entity attribute corresponding to the target symptom entity in the medical record sentence with at least one other symptom entity in the same medical record sentence to obtain multiple target medical record sentences includes:

[0019] Obtaining a symptom category corresponding to each symptom entity in a first medical record short sentence, where the first medical record short sentence is any short sentence among the multiple medical record short sentences;

[0020] The first entity attribute corresponding to the first symptom entity in the first medical record sentence is shared with the second symptom entity to obtain a target medical record sentence corresponding to the first medical record sentence, wherein the first symptom entity and the second symptom entity are adjacent in the first medical record sentence and belong to the same symptom category.

[0021] Optionally, obtaining the target electronic medical record based on the multiple target medical record sentences includes:

[0022] Based on the multiple target medical record short sentences, multiple target medical record entities are obtained, and the symptom entities in the target medical record entities all correspond to entity attributes;

[0023] Extract symptom entities and corresponding entity attributes from the multiple target medical record entities to obtain the target electronic medical record.

[0024] Optionally, before dividing the electronic medical record into a plurality of medical record short sentences based on the separator in the electronic medical record, the method further comprises:

[0025] identifying connectives in the initial electronic medical record;

[0026] The connecting word is replaced with the separator to obtain the electronic medical record.

[0027] The second part of the embodiment of the present disclosure provides an electronic medical record processing device, the device comprising:

[0028] A segmentation module, used for segmenting the electronic medical record into a plurality of medical record short sentences based on the separators in the electronic medical record;

[0029] A first acquisition module is used to acquire symptom entities in the plurality of medical record short sentences and entity attributes corresponding to the symptom entities;

[0030] A sharing module, for sharing, for each of the plurality of medical record sentences, an entity attribute corresponding to a target symptom entity in the medical record sentence with at least one other symptom entity in the same medical record sentence, to obtain a plurality of target medical record sentences;

[0031] The second acquisition module is used to obtain the target electronic medical record based on the multiple target medical record sentences.

[0032] The third part of the embodiment of the present disclosure provides a non-temporary computer-readable storage medium, on which a computer program is stored, and when the program is executed by a processor, the steps of any one of the methods described in the first part are implemented.

[0033] A fourth aspect of the embodiments of the present disclosure provides an electronic device, including:

[0034] a memory having a computer program stored thereon;

[0035] A processor, configured to execute the computer program in the memory to implement the steps of any one of the methods described in the first part.

[0036] Through the above technical solution, at least the following beneficial technical effects can be achieved:

[0037] Based on the separator in the electronic medical record, the electronic medical record is divided into multiple medical record short sentences, the symptom entities in the multiple medical record short sentences and the entity attributes corresponding to the symptom entities are obtained, and for each of the multiple medical record short sentences, the entity attributes corresponding to the target symptom entity in the medical record short sentence are shared with at least one other symptom entity in the same medical record short sentence, and multiple target medical record short sentences are obtained. Based on the multiple target medical record short sentences, the target electronic medical record is obtained. The above technical solution can accurately identify the symptom entities in the electronic medical record, and share the entity attributes corresponding to the target symptom entity in the electronic medical record with other symptom entities in the same medical record short sentence, thereby determining the entity attributes corresponding to each symptom entity, and outputting a structured target electronic medical record, so that the expression of the electronic medical record can be reconstructed and the accuracy of electronic medical record recognition can be improved.

[0038] Other features and advantages of the present disclosure will be described in detail in the following detailed description. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] The accompanying drawings are used to provide a further understanding of the present disclosure and constitute a part of the specification. Together with the following specific embodiments, they are used to explain the present disclosure but do not constitute a limitation of the present disclosure. In the accompanying drawings:

[0040] Figure 1 It is a flowchart of a method for processing electronic medical records proposed according to an exemplary embodiment.

[0041] Figure 2 It is a flowchart of a method for sharing entity attributes proposed according to an exemplary embodiment.

[0042] Figure 3 The present invention is a block diagram of a device for processing electronic medical records according to an exemplary embodiment.

[0043] Figure 4 It is a block diagram of an electronic device according to an exemplary embodiment.

[0044] Figure 5 is a block diagram of another electronic device according to an exemplary embodiment. DETAILED DESCRIPTION

[0045] The specific implementation of the present disclosure is described in detail below in conjunction with the accompanying drawings. It should be understood that the specific implementation described herein is only used to illustrate and explain the present disclosure, and is not used to limit the present disclosure.

[0046] In order to solve the technical problem that some entities identified by electronic medical records in the prior art do not have corresponding entity attributes, resulting in the inability of the reconstructed electronic medical records to accurately represent the entities and the corresponding attributes, and causing the electronic medical records to be inaccurate, the present disclosure provides a method for processing electronic medical records. Figure 1 FIG. 1 is a flowchart of a method for processing an electronic medical record according to an exemplary embodiment. Figure 1 As shown, the method for processing the electronic medical record includes the following steps:

[0047] Step S11, based on the separators in the electronic medical record, the electronic medical record is divided into a plurality of medical record short sentences.

[0048] It is understandable that the electronic medical record is a series of medical conditions recorded by doctors based on the description of the patient's condition when the doctor inquires with the patient. The medical conditions recorded by the doctor are only simply described, and the symptom entities and entity attributes corresponding to the patient are not clear enough. The electronic medical record cannot intuitively reflect the patient's condition. At the same time, the patient's medical record data is a relatively large amount of data to be checked. In order to facilitate the intuitive presentation and effective storage of electronic medical records, the electronic medical records made by doctors need to be processed. Usually, in the same sentence, a sentence separated by a delimiter has the same sentence meaning. Therefore, in this embodiment, the electronic medical record is first divided into multiple medical record short sentences based on the separators in the electronic medical record. Usually, when recording medical records, doctors will use commas as separators to separate two short sentences with different semantics according to conventional writing habits. Therefore, the commas in the electronic medical record can be used as separators; for example, to adapt to the doctor's writing description habits, space characters, semicolons, periods, etc. can be used as separators, and the electronic medical record can be divided into multiple medical record short sentences based on one or more separators, but this embodiment does not limit this.

[0049] It should be noted that in order to ensure the accuracy of the electronic medical records subsequently identified, the symbols in the electronic medical records that represent parallel and equal relationships in this embodiment cannot be used as separators. For example, the semicolon in the electronic medical record cannot be used as a separator. The semicolon needs to be retained as a parallel and equal relationship in a single medical record sentence.

[0050] Step S12, obtaining symptom entities in multiple medical record short sentences and entity attributes corresponding to the symptom entities.

[0051] Identify the multiple medical record sentences obtained, and obtain the symptom entities and corresponding entity attributes included in each medical record sentence. It should be noted that the symptom entity can include multiple categories. For example, it can be a symptom entity representing the patient's specific symptoms, such as "fever", "cough", "wheezing", "weakness" and other specific symptoms, and it can also be a symptom entity representing the name of the disease, such as "cold", "anemia", "allergy", "bronchitis", "asthma", etc. The entity attribute is the property that represents the strength of the symptom entity. For example, it can be a yin-yang sub-attribute indicating whether the patient has specific symptoms. When the patient has related diseases or related symptoms, the entity attribute corresponding to the symptom entity is "positive". Due to the particularity of the medical record situation, when the patient does not have related symptoms and diseases, it is also necessary to make relevant records. At this time, the entity attribute corresponding to the symptom entity is "negative". Optionally, the entity attribute can also be the degree sub-attribute of the symptom entity, for example, used to record the duration of the symptom entity, or the start time, or the strength.

[0052] It is understandable that when identifying medical record sentences, the keyword recognition method can be used to identify the symptom entities and entity attributes in the medical record sentences, and then the symptom entities and entity attributes in the same medical record sentence are matched according to the conventional writing and description habits, so as to obtain multiple symptom entities and corresponding entity attributes. For example, under normal circumstances, the entity attribute representing the yin and yang sub-attribute is located at the starting position of the same medical record sentence, and the entity attribute representing the degree sub-attribute is located at the end of the same medical record sentence. The symptom entity and the corresponding entity attribute are matched together based on the positional relationship between the entity attribute and the symptom entity and the distance relationship in the same medical record sentence. It should be noted that in this embodiment, under normal circumstances, the conventional writing method and the conventional description habit are written or described in the order of a sentence from left to right, where the starting position is the leftmost position of a sentence and the end position is the last position of a sentence.

[0053] Step S13, for each of the multiple medical record sentences, share the entity attribute corresponding to the target symptom entity in the medical record sentence with at least one other symptom entity in the same medical record sentence to obtain multiple target medical record sentences.

[0054] It should be noted that among the multiple symptom entities obtained through the above steps, because there is a parallel description relationship in the medical record short sentence, it will be recognized during the recognition process that some symptom entities do not have corresponding entity attributes. Therefore, in this embodiment, the symptom entity with entity attributes obtained through the above steps is used as the target symptom entity, and the entity attributes corresponding to the target symptom entity are shared with at least one other symptom entity in the same medical record short sentence, thereby obtaining multiple target medical record short sentences.

[0055] It is understandable that before sharing entity attributes, it is necessary to first count the number of symptom entities of the same type in the multiple medical record sentences obtained. For medical record sentences that only contain one symptom entity of the same type in the same medical record sentence, entity attributes will not be shared, and medical record sentences containing two or more symptom entities of the same type will be screened out for entity attribute sharing. It should be noted that when sharing the entity attributes of the target symptom entity, the entity attributes can be shared with other symptom entities that do not have entity attributes in the same medical record sentence, and comprehensive sharing can also be implemented to synchronously share the entity attributes of the target symptom entity with other symptom entities (including symptom entities with entity attributes and symptom entities without entity attributes). For symptom entities with entity attributes, in order to distinguish between shared entity attributes and original entity attributes, the shared entity attributes need to be marked accordingly for subsequent screening and judgment, so as to accurately identify the entity attributes corresponding to the symptom entity.

[0056] Figure 2 FIG. 1 is a flow chart of a method for sharing entity attributes according to an exemplary embodiment. Figure 2 As shown, the above step S13 may include:

[0057] Step S131, based on the position of the first symptom entity in the first medical record short sentence, the yin and yang sub-attributes of the first symptom entity are shared with at least one second symptom entity in the first medical record short sentence in a first direction. The first medical record short sentence is any one of the multiple medical record short sentences segmented by the electronic medical record.

[0058] and / or,

[0059] Step S132, sharing the degree sub-attribute of the first symptom entity with at least one third symptom entity in the first medical history sentence in a second direction, wherein the second direction is opposite to the first direction.

[0060] In this embodiment, the symptom entity includes a yin-yang sub-attribute for indicating whether there are relevant symptoms, and / or a degree sub-attribute for indicating the strength of the symptom entity. It can be understood that when the yin-yang sub-attribute is identified, in order to improve the accuracy of identification, only the yin-yang sub-attribute indicating "negative" is identified. When it is determined that the symptom entity has no corresponding "negative" yin-yang sub-attribute, the corresponding first symptom entity is marked as a "positive" yin-yang sub-attribute. It can be understood that according to writing habits, under normal circumstances, the yin-yang sub-attribute of the symptom entity is located at the starting position of the same medical record short sentence, and the corresponding degree sub-attribute is located at the end position of the same medical record short sentence. The semantics expressed in the same medical record short sentence are basically the same. Therefore, it is necessary to synchronize the yin-yang sub-attribute and / or degree sub-attribute corresponding to the first symptom entity to other symptom entities in the same medical record short sentence, and determine the position of the first symptom entity corresponding to the entity attribute in the first medical record short sentence by identification, and the yin-yang sub-attribute located at the starting position is synchronously shared with other symptom entities in the same medical record short sentence according to the first direction of positive writing, and the degree sub-attribute located at the end position is synchronously shared with other symptom entities in the same medical record short sentence according to the second direction opposite to the first direction.For example, the medical record short sentence is "the patient has no fever, cough, abdominal pain and other symptoms". According to the above method, the symptom entity determined from the starting position of "the patient has no fever, cough, abdominal pain and other symptoms" is "fever", and the corresponding yin and yang sub-attributes of the symptom entity "fever" are "none", and the symptom entity "fever" has no degree sub-attribute, so the yin and yang sub-attribute "none" is shared with the symptom entities "cough" and "abdominal pain", and then the degree sub-attribute is identified. The corresponding degree sub-attribute is not identified for the above medical record short sentence, so the degree sub-attribute is not correspondingly identified. The target medical record sentence after sharing is "the patient has no fever, no cough, no abdominal pain and other symptoms"; the medical record sentence is "the patient has runny nose, chest tightness, cough, and sputum for 6 days", and the symptom entity determined from the end of "the patient has runny nose, chest tightness, cough, and sputum for 6 days" according to the above method is "sputum", and the corresponding degree sub-attribute of the symptom entity "sputum" is "6 days", and the symptom entity "sputum" has no yin and yang sub-attributes, so the degree sub-attribute "6 days" is shared with the symptom entities "cough", "chest tightness" and "snot", and then the yin and yang sub-attributes are performed. Yang sub-attribute identification: For the above medical record short sentence, no corresponding yin and yang sub-attributes are identified, so the corresponding yin and yang sub-attributes are not shared, so the target medical record short sentence after sharing is "the patient has had runny nose for 6 days, chest tightness for 6 days, cough for 6 days, and expectoration for 6 days"; the medical record short sentence is "the patient has abdominal pain, diarrhea, indigestion, bloating, and weakness for 3 days". According to the above method, the symptom entity and entity attribute determined from the starting position and the end position of the medical record short sentence "the patient has abdominal pain, diarrhea, indigestion, bloating, and weakness for 3 days" are "abdominal pain" and " Weakness for 3 days", the yin and yang sub-attributes "yes" corresponding to the symptom entity "abdominal pain" at the initial position are shared with other symptom entities in order from left to right, and the medical record short sentence "the patient has abdominal pain, diarrhea, indigestion, bloating, and weakness for 3 days" is obtained. Then the degree sub-attribute "3 days" corresponding to the symptom entity "weakness" at the last position is shared with other symptom entities in order from right to left, so as to obtain the target medical record short sentence "the patient has abdominal pain for 3 days, diarrhea for 3 days, indigestion for 3 days, bloating, and weakness for 3 days". It should be noted that in this embodiment, when sharing entity attributes through the above steps, according to the conventional writing method, there is no restriction on the order of sharing entity attributes from left to right and from right to left.

[0061] Optionally, in order to improve the processing efficiency of electronic medical records, the symptom entities in the same medical record short sentence can be classified according to the above-mentioned sharing rules, and can be divided into the first category of symptom entities: the symptom entity and the immediately following symptom entity are of the same category, the second category of symptom entities: the symptom entity and the previous symptom entity and the next symptom entity are of the same category, the third category of symptom entities: the symptom entity is only of the same category as the previous symptom entity, and the fourth category of symptom entities: the symptom entity and the previous and next symptom entities are not of the same category. Based on the above-mentioned sharing method, the yin and yang sub-attributes of the first category of symptom entities are shared with the next symptom entity, the yin and yang sub-attributes of the second category of symptom entities are shared with the next symptom entity and / or the degree sub-attributes are shared with the previous symptom entity, the degree sub-attributes of the third category of symptom entities are shared with the previous symptom entity, and the fourth category of symptom entities are not shared. The sharing steps are continued until all symptom entities in the same medical record short sentence contain corresponding entity attributes.

[0062] Optionally, before the above step S132, the method may further include:

[0063] Step S134, identifying the yin and yang sub-attributes in a plurality of medical record short sentences.

[0064] Step S135, if multiple yin and yang sub-attributes are identified in the second medical record sentence, the second medical record sentence is divided into multiple medical record sentences based on the positions of the symptom entities corresponding to the multiple yin and yang sub-attributes in the second medical record sentence, and the second medical record sentence is any one of the multiple medical record sentences divided by the electronic medical record.

[0065] It is understandable that in the actual process of recording the patient's symptoms, different yin and yang sub-attributes may appear in the same medical record sentence. For example, the patient is accompanied by fever, fever, dizziness and other symptoms, but has no cough, runny nose, sputum and other conditions. For such a medical record sentence, the "positive" yin and yang sub-attribute corresponding to "accompanying" covers the symptom entity of "fever, fever, dizziness", and the "negative" yin and yang sub-attribute corresponding to "none" covers the symptom entity of "cough, runny nose, sputum". Therefore, when the yin and yang sub-attributes corresponding to the symptom entity at the starting position are shared with other symptom entities in the first direction, for the symptom entities corresponding to other yin and yang sub-attributes that appear later, there are multiple identical or mutually exclusive yin and yang sub-attributes after sharing. Therefore, before sharing the yin and yang sub-attributes, it is necessary to process the second medical record sentence first, identify the multiple yin and yang sub-attributes in the second medical record sentence, and divide the medical record sentence into multiple medical record sentences from the position of the yin and yang sub-attributes to avoid mutual exclusion when sharing entity attributes and improve the accuracy of electronic medical record recognition.

[0066] Optionally, after the above step S133, the method may further include:

[0067] Step S136, if the third entity attribute corresponding to the third symptom entity and the degree sub-attribute of the first symptom entity are mutually exclusive during the sharing process, the sharing of the degree sub-attribute of the first symptom entity in the second direction is stopped starting from the third symptom entity.

[0068] It can be understood that in this embodiment, according to the position of the first symptom entity in the first medical record sentence, the degree sub-attribute corresponding to the first symptom entity is shared with the third symptom entity in sequence in the second direction (it is set that under normal circumstances, the forward reading direction is the first direction, and the corresponding second direction is the direction opposite to the first direction). When sharing to a third symptom entity with an initial degree sub-attribute, the sharing of the degree sub-attribute corresponding to the first symptom entity in the second direction is stopped from the position of the third symptom entity, and the initial sub-attribute corresponding to the third symptom entity is shared in the second direction synchronously. For example, the first medical record sentence is "the patient coughed, had a fever, and had asthma for 3 days, had a lot of sputum, coughed up phlegm, and had a stuffy nose for 6 days". When sharing in the above manner, first identify the symptom entity and the corresponding entity attribute in the medical record sentence, and determine the degree sub-attribute " 3 days" and "6 days", and then according to the above sharing method, the "degree sub-attribute" is shared in the second direction to obtain "coughing for 3 days", "fever for 3 days", "asthma for 3 days" and "excessive sputum for 6 days", "coughing and sputum for 6 days", and "nasal congestion for 6 days". When the degree sub-attribute "6 days" corresponding to "excessive sputum for 6 days" is shared with "asthma for 3 days", since the degree sub-attribute "6 days" to be shared is mutually exclusive with the degree sub-attribute "3 days" of the symptom entity "asthma", the sharing of the degree sub-attribute "6 days" in the second direction is stopped, and the target medical record short sentence obtained is "the patient has coughed for 3 days, had a fever for 3 days, and asthma for 3 days, excessive sputum for 6 days, coughed and sputum for 6 days, and nasal congestion for 6 days". For example, when entity attributes are mutually exclusive during sharing, delimiters can be added between the mutually exclusive entity attributes to separate the medical record short sentences, so that the corresponding symptom conditions expressed in the electronic medical record are more accurate.

[0069] Step S133, based on obtaining multiple symptom entities after sharing the yin-yang sub-attributes and / or the degree sub-attributes, obtain the target medical record short sentence corresponding to the first case short sentence.

[0070] The yin and yang sub-attributes and / or degree sub-attributes are retained after sharing, so that the symptom entities in the same medical record sentence all include the corresponding yin and yang sub-attributes and / or degree sub-attributes, and then the target medical record sentence corresponding to the first medical record sentence is obtained, and other medical record sentences are shared according to the above-mentioned sharing method to obtain multiple target medical record sentences.

[0071] Optionally, the above step S13 may further include:

[0072] Step S137, obtaining the symptom category corresponding to each symptom entity in the first medical record short sentence, where the first medical record short sentence is any short sentence among the multiple medical record short sentences.

[0073] Step S138, sharing the first entity attribute corresponding to the first symptom entity in the first medical record sentence with the second symptom entity, obtaining the target medical record sentence corresponding to the first medical record sentence, the first symptom entity and the second symptom entity are adjacent in the first medical record sentence and belong to the same symptom category.

[0074] It can be understood that in the present embodiment, there are symptom entities of multiple symptom categories, which may be symptom entities representing specific symptoms of patients, for example, specific symptoms such as "fever", "cough", "wheezing", "weakness", etc., or symptom entities representing disease names, for example, "cold", "anemia", "allergy", "bronchitis", "asthma", etc., or symptom entities indicating that patients have received corresponding treatments, for example, "6 days of chemotherapy", "6 days of saline infusion", "2 vaccinations", etc. When doctors describe the condition of the disease, they usually describe the symptom entities of the same category together. Therefore, when sharing attributes, it is necessary to first identify the symptom entities of the same type and share the first entity attribute corresponding to the first symptom entity of the same type in the first medical record sentence with the second symptom entity to obtain the target medical record sentence corresponding to the first medical record sentence.

[0075] Step S14, obtaining a target electronic medical record based on multiple target medical record short sentences.

[0076] The multiple target medical record short sentences obtained through the above steps are sorted and sorted again according to the order in the electronic medical record to obtain a target electronic medical record with more accurate entity attributes corresponding to the symptom entity.

[0077] Optionally, the above step S14 may include:

[0078] Step S141, based on multiple target medical record phrases, multiple target medical record entities are obtained, and the symptom entities in the target medical record entities all correspond to entity attributes.

[0079] Step S142, extracting symptom entities and corresponding entity attributes from multiple target medical record entities to obtain target electronic medical records.

[0080] It is understandable that in the hospital's storage system, the patient's electronic medical record is a large and cumbersome data. The electronic medical record can be used by doctors for medical record analysis and medical record teaching, but when doctors identify the electronic medical record, they only need to identify and understand the symptom entities and corresponding entity attributes in the electronic medical record. Other related vernacular descriptions in the electronic medical record are not important. Therefore, it is necessary to output a structured electronic medical record. In this embodiment, based on the obtained multiple target medical record short sentences, the symptom entities and corresponding entity attributes in the target medical record short sentences are extracted, and the structured target electronic medical record is output based on the proposed multiple symptom entities and corresponding entity attributes.

[0081] Optionally, before the above step S11, the method for processing electronic medical records may further include:

[0082] Step S21, identifying connecting words in the initial electronic medical record.

[0083] Step S22, replacing the conjunction with a separator to obtain the electronic medical record.

[0084] In order to ensure the accuracy of electronic medical record recognition and improve the efficiency of processing, it is necessary to pre-process the electronic medical records according to writing and description habits. It is understandable that when doctors record the condition of the patient, in addition to the necessary separators such as commas to separate the sentences, they usually use conjunctions to explain the medical record. When making semantic explanations, conjunctions can represent "parallel", "equivalent", "opposite", etc. By identifying the conjunctions in the initial electronic medical record and replacing the conjunctions with separators, the corresponding electronic medical record is obtained, thereby realizing the pre-processing of the initial electronic medical record. For example, the conjunction can be "and", "and", "but", "also", "however", etc., which represent parallel relationships.

[0085] Through the above technical solution, based on the separator in the electronic medical record, the electronic medical record is divided into multiple medical record short sentences, the symptom entities in the multiple medical record short sentences and the entity attributes corresponding to the symptom entities are obtained, and for each of the multiple medical record short sentences, the entity attributes corresponding to the target symptom entity in the medical record short sentence are shared with at least one other symptom entity in the same medical record short sentence, and multiple target medical record short sentences are obtained. Based on the multiple target medical record short sentences, the target electronic medical record is obtained. By adopting this method of the present disclosure, the symptom entities in the electronic medical record are accurately identified, and according to the sharing method provided by the present disclosure, the entity attributes corresponding to the target symptom entity in the electronic medical record are shared with other symptom entities in the same medical record short sentence, so that each symptom entity has corresponding entity attributes, and the structured target electronic medical record is obtained by output. The expression method of the electronic medical record can be reconstructed through the method of the present disclosure, which improves the accuracy of electronic medical record recognition.

[0086] Figure 3is a block diagram of an electronic medical record processing device according to an exemplary embodiment. Figure 3 The electronic medical record processing device 100 includes a segmentation module 110 , a first acquisition module 120 , a sharing module 130 and a second acquisition module 140 .

[0087] The segmentation module 110 is used to segment the electronic medical record into a plurality of medical record short sentences based on the separators in the electronic medical record.

[0088] The first acquisition module 120 is used to acquire the symptom entities in the multiple medical record short sentences and the entity attributes corresponding to the symptom entities.

[0089] The sharing module 130 is used to share the entity attribute corresponding to the target symptom entity in the medical record short sentence with at least one other symptom entity in the same medical record short sentence for each of the multiple medical record short sentences, so as to obtain multiple target medical record short sentences.

[0090] The second acquisition module 140 is used to obtain a target electronic medical record based on the multiple target medical record short sentences.

[0091] Optionally, the entity attribute includes a yin-yang sub-attribute and / or a degree sub-attribute, and the sharing module 130 may include:

[0092] A first sharing submodule is used to share the yin and yang sub-attributes of the first symptom entity with at least one second symptom entity in the first medical history sentence in a first direction based on the position of the first symptom entity in the first medical history sentence, where the first medical history sentence is any one of the multiple medical history sentences. And / or,

[0093] The second sharing submodule is used to share the degree sub-attribute of the first symptom entity with at least one third symptom entity in the first medical history sentence in a second direction, and the second direction is opposite to the first direction.

[0094] The obtaining submodule is used to obtain a plurality of symptom entities based on sharing of yin-yang sub-attributes and / or degree sub-attributes, and obtain a target medical record short sentence corresponding to the first case short sentence.

[0095] Optionally, the sharing module 130 may further include:

[0096] The identification submodule is used to identify the yin and yang sub-attributes in multiple medical record short sentences.

[0097] A segmentation submodule is used to segment the second medical record short sentence into multiple medical record short sentences based on the positions of the symptom entities corresponding to the multiple yin and yang sub-attributes in the second medical record short sentence if multiple yin and yang sub-attributes are identified in the second medical record short sentence. The second medical record is any one of the multiple medical record short sentences segmented from the electronic medical record.

[0098] Optionally, the sharing module 130 may further include:

[0099] The execution submodule is used to stop sharing the degree subattribute of the first symptom entity in the second direction starting from the third symptom entity if the third entity attribute corresponding to the third symptom entity and the degree subattribute of the first symptom entity are mutually exclusive during the sharing process.

[0100] Optionally, the sharing module 130 may be used to:

[0101] The symptom category corresponding to each symptom entity in the first medical record short sentence is obtained, where the first medical record short sentence is any short sentence among multiple medical record short sentences.

[0102] The first entity attribute corresponding to the first symptom entity in the first medical record sentence is shared with the second symptom entity to obtain a target medical record sentence corresponding to the first medical record sentence. The first symptom entity and the second symptom entity are adjacent in the first medical record sentence and belong to the same symptom category.

[0103] Optionally, the second acquisition module 140 may also be used to:

[0104] Based on multiple target medical record sentences, multiple target medical record entities are obtained, and the symptom entities in the target medical record entities all correspond to entity attributes.

[0105] Extract symptom entities and corresponding entity attributes from multiple target medical record entities to obtain target electronic medical records.

[0106] Optionally, the segmentation module 110 is further configured to:

[0107] Before dividing the electronic medical record into a plurality of medical record short sentences based on the separators in the electronic medical record, connecting words in the initial electronic medical record are identified.

[0108] Replace the conjunctions with separators to obtain the electronic medical records.

[0109] Regarding the device in the above embodiment, the specific manner in which each module performs operations has been described in detail in the embodiment of the method, and will not be elaborated here.

[0110] Figure 4 FIG. 4 is a block diagram of an electronic device 400 according to an exemplary embodiment. Figure 4 As shown, the electronic device 400 may include: a processor 401 and a memory 402. The electronic device 400 may also include one or more of a multimedia component 403, an input / output (I / O) interface 404, and a communication component 405.

[0111] The processor 401 is used to control the overall operation of the electronic device 400 to complete all or part of the steps in the above-mentioned electronic medical record processing method. The memory 402 is used to store various types of data to support the operation of the electronic device 400, and these data may include, for example, instructions for any application or method used to operate on the electronic device 400, and application-related data, such as contact data, sent and received messages, pictures, audio, video, etc. The memory 402 can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (Static Random Access Memory, referred to as SRAM), electrically erasable programmable read-only memory (Electrically Erasable Programmable Read-Only Memory, referred to as EEPROM), erasable programmable read-only memory (Erasable Programmable Read-Only Memory, referred to as EPROM), programmable read-only memory (Programmable Read-Only Memory, referred to as PROM), read-only memory (Read-Only Memory, referred to as ROM), magnetic memory, flash memory, magnetic disk or optical disk. The multimedia component 403 may include a screen and an audio component. The screen may be, for example, a touch screen, and the audio component is used to output and / or input audio signals. For example, the audio component may include a microphone for receiving external audio signals. The received audio signal may be further stored in the memory 402 or sent through the communication component 405. The audio component also includes at least one speaker for outputting audio signals. The I / O interface 404 provides an interface between the processor 401 and other interface modules, and the other interface modules may be keyboards, mice, buttons, etc. These buttons may be virtual buttons or physical buttons. The communication component 405 is used for wired or wireless communication between the electronic device 400 and other devices. Wireless communication, such as Wi-Fi, Bluetooth, Near Field Communication (NFC), 2G, 3G, 4G, NB-IOT, eMTC, or other 5G, etc., or a combination of one or more of them, is not limited here. Therefore, the corresponding communication component 405 may include: Wi-Fi module, Bluetooth module, NFC module, etc.

[0112] In an exemplary embodiment, the electronic device 400 can be implemented by one or more application specific integrated circuits (ASIC), digital signal processors (DSP), digital signal processing devices (DSPD), programmable logic devices (PLD), field programmable gate arrays (FPGA), controllers, microcontrollers, microprocessors or other electronic components to execute the above-mentioned electronic medical record processing method.

[0113] In another exemplary embodiment, a computer-readable storage medium including program instructions is also provided, and when the program instructions are executed by a processor, the steps of the above-mentioned electronic medical record processing method are implemented. For example, the computer-readable storage medium can be the above-mentioned memory 402 including program instructions, and the above-mentioned program instructions can be executed by the processor 401 of the electronic device 400 to complete the above-mentioned electronic medical record processing method.

[0114] Figure 5 is a block diagram of an electronic device 500 according to an exemplary embodiment. For example, the electronic device 500 may be provided as a server. Figure 5 The electronic device 500 includes a processor 522, which may be one or more, and a memory 532 for storing a computer program executable by the processor 522. The computer program stored in the memory 532 may include one or more modules, each corresponding to a set of instructions. In addition, the processor 522 may be configured to execute the computer program to perform the above-mentioned method for processing electronic medical records.

[0115] In addition, the electronic device 500 may further include a power supply component 526 and a communication component 550. The power supply component 526 may be configured to perform power management of the electronic device 500, and the communication component 550 may be configured to implement communication of the electronic device 500, for example, wired or wireless communication. In addition, the electronic device 500 may further include an input / output (I / O) interface 558. The electronic device 500 may operate based on an operating system stored in the memory 532, such as Windows Server 2008. TM , Mac OSX TM , Unix TM , Linux TM etc.

[0116] In another exemplary embodiment, a computer-readable storage medium including program instructions is also provided, and when the program instructions are executed by a processor, the steps of the above-mentioned electronic medical record processing method are implemented. For example, the non-transitory computer-readable storage medium can be the above-mentioned memory 532 including program instructions, and the above-mentioned program instructions can be executed by the processor 522 of the electronic device 500 to complete the above-mentioned electronic medical record processing method.

[0117] In another exemplary embodiment, a computer program product is also provided. The computer program product includes a computer program that can be executed by a programmable device. The computer program has a code portion for executing the above-mentioned electronic medical record processing method when executed by the programmable device.

[0118] The preferred embodiments of the present disclosure are described in detail above in conjunction with the accompanying drawings. However, the present disclosure is not limited to the specific details in the above embodiments. Within the technical concept of the present disclosure, the technical solution of the present disclosure can be subjected to a variety of simple modifications, and these simple modifications all belong to the protection scope of the present disclosure. It should also be noted that the various specific technical features described in the above specific embodiments can be combined in any suitable manner without contradiction.

[0119] In addition, various embodiments of the present disclosure may be arbitrarily combined, and as long as they do not violate the concept of the present disclosure, they should also be regarded as the contents disclosed by the present disclosure.

Claims

1. A method for processing electronic medical records, characterized in that: include: Based on the separator in the electronic medical record, the electronic medical record is divided into a plurality of medical record short sentences; Acquire symptom entities in the plurality of medical record short sentences, and entity attributes corresponding to the symptom entities; For each of the multiple medical record sentences, sharing the entity attribute corresponding to the target symptom entity in the medical record sentence with at least one other symptom entity in the same medical record sentence, to obtain multiple target medical record sentences; The entity attributes include yin and yang sub-attributes and / or degree sub-attributes, and for each of the multiple medical record sentences, the entity attributes corresponding to the target symptom entity in the medical record sentence are shared with at least one other symptom entity in the same medical record sentence to obtain multiple target medical record sentences, including: based on the position of a first symptom entity in the first medical record sentence, the yin and yang sub-attributes of the first symptom entity are shared with at least one second symptom entity in the first medical record sentence in a first direction, and the first medical record sentence is any medical record sentence in the multiple medical record sentences; and / or, the degree sub-attributes of the first symptom entity are shared with at least one third symptom entity in the first medical record sentence in a second direction, and the second direction is opposite to the first direction; based on the multiple symptom entities obtained after sharing the yin and yang sub-attributes and / or degree sub-attributes, the target medical record sentence corresponding to the first medical record sentence is obtained; Based on the multiple target medical record sentences, a target electronic medical record is obtained.

2. The processing method according to claim 1, characterized in that: Before sharing the yin and yang sub-attributes of the first symptom entity with at least one second symptom entity in the first medical history sentence in a first direction based on the position of the first symptom entity in the first medical history sentence, the method further includes: identifying the yin and yang sub-attributes in the plurality of medical record short sentences; If multiple yin and yang sub-attributes are identified in the second medical record sentence, the second medical record sentence is divided into multiple medical record sentences based on the positions of the symptom entities corresponding to the multiple yin and yang sub-attributes in the second medical record sentence, and the second medical record sentence is any one of the multiple medical record sentences divided by the electronic medical record.

3. The processing method according to claim 1, characterized in that: After sharing the degree sub-attribute of the first symptom entity with at least one third symptom entity in the first medical history sentence according to the second direction, the method further includes: If during the sharing process the third entity attribute corresponding to the third symptom entity and the degree sub-attribute of the first symptom entity are mutually exclusive, the sharing of the degree sub-attribute of the first symptom entity in the second direction is stopped starting from the third symptom entity.

4. The processing method according to claim 1, characterized in that: For each of the multiple medical record sentences, sharing the entity attribute corresponding to the target symptom entity in the medical record sentence with at least one other symptom entity in the same medical record sentence to obtain multiple target medical record sentences includes: Obtaining a symptom category corresponding to each symptom entity in a first medical record short sentence, where the first medical record short sentence is any short sentence among the multiple medical record short sentences; The first entity attribute corresponding to the first symptom entity in the first medical record sentence is shared with the second symptom entity to obtain a target medical record sentence corresponding to the first medical record sentence, wherein the first symptom entity and the second symptom entity are adjacent in the first medical record sentence and belong to the same symptom category.

5. The processing method according to claim 1, characterized in that: The step of obtaining the target electronic medical record based on the multiple target medical record short sentences includes: Based on the multiple target medical record short sentences, multiple target medical record entities are obtained, and the symptom entities in the target medical record entities all correspond to entity attributes; Extract symptom entities and corresponding entity attributes from the multiple target medical record entities to obtain the target electronic medical record.

6. The processing method according to claim 1, characterized in that: Before dividing the electronic medical record into a plurality of medical record short sentences based on the separator in the electronic medical record, the method further includes: identifying connectives in the initial electronic medical record; The connecting word is replaced with the separator to obtain the electronic medical record.

7. An electronic medical record processing device, characterized in that: include: A segmentation module, used for segmenting the electronic medical record into a plurality of medical record short sentences based on the separators in the electronic medical record; A first acquisition module is used to acquire symptom entities in the plurality of medical record short sentences and entity attributes corresponding to the symptom entities; A sharing module, for sharing, for each of the plurality of medical record sentences, an entity attribute corresponding to a target symptom entity in the medical record sentence with at least one other symptom entity in the same medical record sentence, to obtain a plurality of target medical record sentences; The entity attribute includes a yin-yang sub-attribute and / or a degree sub-attribute, and for each of the multiple medical record short sentences, the sharing module is further used to: based on the position of the first symptom entity in the first medical record short sentence, share the yin-yang sub-attribute of the first symptom entity with at least one second symptom entity in the first medical record short sentence in a first direction, wherein the first medical record short sentence is any one of the multiple medical record short sentences; And / or, sharing the degree sub-attribute of the first symptom entity with at least one third symptom entity in the first medical history short sentence in a second direction, wherein the second direction is opposite to the first direction; based on obtaining multiple symptom entities after sharing the yin-yang sub-attributes and / or degree sub-attributes, obtaining a target medical history short sentence corresponding to the first medical history short sentence; The second acquisition module is used to obtain a target electronic medical record based on the multiple target medical record short sentences.

8. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the steps of the method described in any one of claims 1 to 6 are implemented.

9. An electronic device, characterized in that: include: a memory having a computer program stored thereon; A processor, configured to execute the computer program in the memory to implement the steps of the method according to any one of claims 1 to 6.

Images