CN109637605B - Electronic medical record structuring method and computer-readable storage medium - Google Patents

Abstract

The invention provides an electronic medical record structuring method and a computer readable storage medium. Wherein, the method comprises the following steps: loading a first medical knowledge base; the first electronic medical record is divided into sentences according to special symbols to obtain a plurality of text sentences; matching each of the plurality of text sentences with attributes in the first medical knowledge base using a matching scoring algorithm; and storing the matching result. By the method and the device, the problem that the electronic medical record cannot be completely structured in the related technology is solved, and the electronic medical record is completely structured.

Description

Electronic medical record structuring method and computer readable storage medium

technical field

The present invention relates to the medical field, in particular, to an electronic medical record structuring method and a computer-readable storage medium.

Background technique

With the electronic, networked and intelligent medical system, the patient's medical data is stored in the electronic medical record, including all-round information such as chief complaint, medical history, examination, diagnosis, treatment plan, and disposal. In the context of big data, these raw data provide new possibilities for medical diagnosis and decision-making, which makes people consider mining information, extracting rules, and designing intelligent systems from these medical record data to further improve medical level and medical quality.

However, the electronic medical record database often saves the original text entered by the doctor. Although it is written according to some specified templates, it still has some freedom and flexibility of natural language expression. Therefore, such data is not fully structured, but only semi-structured data, which is not suitable for deeper scientific research tasks and smart medical projects. This creates a requirement for us to structure raw text data.

Due to the diversity of natural language expressions and the professionalism of medical terminology, there are certain difficulties in the structuring method of electronic medical record texts, and the current domestic research work is still insufficient. For the electronic medical record structuring method, the results of domestic research work are mainly based on the use of semantic positive and negative disease information to make positive or negative judgments on electronic medical records. This method can solve the disease information calibrated by binary logic, but for numerical, Information such as the degree of disease cannot be extracted; in addition, the current research results on the occurrence site of patient-related disease information have not proposed a corresponding solution. The incompleteness of this information extraction has formed certain limitations for medical research, the development of diagnostic decision-making intelligent systems, and so on.

The purpose of the present invention is to perform complete information extraction on the electronic medical record for different types of disease information and medical treatment information, so as to realize the complete structuring of the text of the electronic medical record.

SUMMARY OF THE INVENTION

The present invention provides a method for structuring an electronic medical record and a computer-readable storage medium, so as to at least solve the problem that the electronic medical record cannot be completely structured in the related art.

In a first aspect, an embodiment of the present invention provides a method for structuring an electronic medical record, including: loading a first medical knowledge base; segmenting the first electronic medical record according to special symbols to obtain multiple text sentences; using a matching scoring algorithm , matching the attributes in the first medical knowledge base for each text sentence in the plurality of text sentences; and saving the matching result.

In a second aspect, an embodiment of the present invention provides a computer-readable storage medium on which computer program instructions are stored, and when the computer program instructions are executed by a processor, the method described in the first aspect is implemented.

According to the electronic medical record structuring method and the computer-readable storage medium provided by the embodiment of the present invention, the first medical knowledge base is loaded; the first electronic medical record is sentenced according to special symbols to obtain a plurality of text sentences; the matching scoring algorithm is used , matches the attributes in the first medical knowledge base for each text sentence in the multiple text sentences; saves the matching results, solves the problem that the electronic medical record cannot be completely structured in the related technology, and realizes the complete structuring of the electronic medical record.

Description of drawings

The accompanying drawings described herein are used to provide a further understanding of the present invention and constitute a part of the present application. The exemplary embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute an improper limitation of the present invention. In the attached image:

1 is a flowchart of a method for structuring an electronic medical record according to an embodiment of the present invention;

2 is a schematic diagram of a hardware structure of an electronic medical record structuring device according to an embodiment of the present invention;

3 is a flowchart of a method for structuring an electronic medical record according to a preferred embodiment of the present invention;

4 is a schematic diagram of an example of the structure of a first medical knowledge base in the field of oral prosthodontics according to a preferred embodiment of the present invention;

5 is a schematic diagram of an example of an electronic medical record according to a preferred embodiment of the present invention;

6 is a schematic diagram of a structural matching result of an electronic medical record according to a preferred embodiment of the present invention;

FIG. 7 is a statistical chart of matching frequency of attributes in the structural matching result of electronic medical records according to a preferred embodiment of the present invention.

Detailed ways

The features and exemplary embodiments of various aspects of the present invention will be described in detail below. In order to make the objects, technical solutions and advantages of the present invention more clear, the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention, but not to limit the present invention. It will be apparent to those skilled in the art that the present invention may be practiced without some of these specific details. The following description of the embodiments is only intended to provide a better understanding of the present invention by illustrating examples of the invention.

It should be noted that, in this document, relational terms such as first and second are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any relationship between these entities or operations. any such actual relationship or sequence exists. Moreover, the terms "comprising", "comprising" or any other variation thereof are intended to encompass a non-exclusive inclusion such that a process, method, article or device that includes a list of elements includes not only those elements, but also includes not explicitly listed or other elements inherent to such a process, method, article or apparatus. Without further limitation, an element defined by the phrase "comprises" does not preclude the presence of additional identical elements in a process, method, article, or device that includes the element.

A method for structuring an electronic medical record is provided in this embodiment. FIG. 1 is a flowchart of a method for structuring an electronic medical record according to an embodiment of the present invention. As shown in FIG. 1 , the process includes the following steps:

Step S101, loading the first medical knowledge base;

Step S102, segmenting the first electronic medical record according to special symbols to obtain a plurality of text sentences;

Step S103, using the matching scoring algorithm to match the attributes in the first medical knowledge base to each text sentence in the plurality of text sentences;

Step S104, save the matching result.

Through the above steps, the text sentence can be well matched with the attributes in the first medical knowledge base by using the matching scoring algorithm, and the matched keywords can not only include disease information calibrated by binary logic, but also can match numerical values, diseases It can solve the problem that the electronic medical record cannot be completely structured in the related technology, and realize the complete structuring of the electronic medical record.

Optionally, the first medical knowledge base includes multiple parts, each part includes one or more attributes, one or more keywords corresponding to the attributes, and each attribute at least includes: attribute name, attribute value and location, each A keyword also includes the keyword's score. For example, in the first medical knowledge base, its basic unit is an attribute, which consists of three parts: attribute name, attribute value and location. The attribute name can be the symptoms, physical characteristics or treatment methods of a certain disease; its corresponding attribute The value can be the presence or absence of symptoms, the severity of the symptoms, the specific manifestation of physical characteristics or the specific method of treatment, etc.; the location can be a body part with corresponding attributes. A set of attributes collectively belongs to a section (such as examination, treatment plan, etc.), and each section constitutes the entire knowledge base.

Due to the complexity of medical diagnosis and treatment measures themselves, in order to describe the medical knowledge in detail and preserve the information of the original medical record as much as possible in the structuring process, in this embodiment, the first medical knowledge base can perform the following steps: Improvements in aspects: a) expand the value type of attribute values; b) add "location" to each attribute to describe the body part of the corresponding attribute; c) add description of time series information; d) classify attributes based on medical knowledge , forming a hierarchical expression of medical knowledge.

The specific instructions are as follows:

a) The attribute value types of the first medical knowledge base include real number type, Boolean type, discrete classification type, etc., and the value method of attribute value includes judgment, single selection, number, multiple selection and various combinations of these methods . This diverse expression form can realize the value expression of various attributes that appear in medicine.

b) Since most of the attributes in the first medical knowledge base relate to a specific body part, such as the occurrence part of disease information, the implementation part of medical measures, etc., in this embodiment, a corresponding body part description is added to the attributes. At the same time, after adding the "location" description, it is necessary to add the extraction of the "location" information in the structuring method, which will be further described in this embodiment.

c) Since the medical behavior itself is a procedural behavior, rather than a simple static combination of various medical measures, especially the treatment plan and disposal measures specified for the patient's symptoms, there is a sequential relationship between different medical measures. In order to preserve the sequential dependencies between different medical measures, a description of time series information is added to the first medical knowledge base. For example, the serialized expression of attributes can be realized by adding two members, step and substep, to attributes that need to express time series to describe the order in which the attributes appear in the treatment process.

d) Based on medical considerations, the first medical knowledge base involved in this embodiment is divided into eight parts: chief complaint, follow-up, history of present illness, past history, examination, diagnosis, treatment plan, and disposal, each part describing specific needs Attributes are designed and graded in the medical field. For example, in the field of prosthodontics, the inspection part includes the inspection results of the teeth and the oral cavity. The inspection results of the oral part are divided into two sub-parts according to whether they are related to the tooth position. Each of the above parts includes several attributes. Disease information is described in detail.

The above-mentioned first medical knowledge base can appropriately implement the structured expression of the original medical record text.

Optionally, the special symbols include at least one of the following: Chinese and English commas, periods, line breaks, and tabs.

Optionally, before loading the first medical knowledge base, the method further includes: loading a second medical knowledge base; extracting keywords and their scores according to the second medical knowledge base and the second electronic medical record; according to the second medical knowledge base And the extracted keywords and their scores to build the first medical knowledge base. In each embodiment, the structure of the first medical knowledge base needs to have corresponding specifications, and in this embodiment, a second medical knowledge base is provided, and this second medical knowledge base is equivalent to a standard template of the first medical knowledge base; Similar to the first medical knowledge base, the second medical knowledge base also includes a plurality of parts, each part includes one or more attributes; each attribute at least includes: attribute name, attribute value and location. Different from the first medical knowledge base, the second medical knowledge base does not have one or more keywords corresponding to the attributes and score information of the keywords. These keywords and their score information are extracted from the second electronic medical record. The first medical knowledge base is constructed by adding one or more keywords and their scores for each attribute in the second medical knowledge base.

Optionally, extracting keyword names and keyword scores according to the second medical knowledge base and the second electronic medical record includes: performing word segmentation on text sentences in the second electronic medical record according to attribute names and attribute values to obtain a plurality of keywords, and The synonyms and synonyms of the keyword are also used as keywords; according to the importance of the keyword (whether it is a common word), negativity (whether it is a negative word) and the weight of the logical relationship (and, or, not), give its different scores.

Optionally, using a matching scoring algorithm, matching the attributes in the first medical knowledge base to each text sentence in the multiple text sentences includes: matching each text sentence to the keywords of all attributes and their scores to obtain each text sentence. The text sentence corresponds to the total score of all attributes; the text sentence whose total score of the attribute is higher than the preset threshold is matched with the keywords in the attribute and their scores, and the attribute value and position in the text sentence corresponding to the attribute are obtained. Attribute value score and position score; the attribute value, position and corresponding attribute with the highest attribute value score and position score are used as the matching result of this text sentence. Through the above matching scoring algorithm, the matching between text sentences and attributes is realized.

Optionally, the matching result includes: a text sentence, and the attribute, attribute value, position, part to which the text sentence corresponds, and the position of the text sentence in the first electronic medical record. When saving the matching results, you can save the matching results of each text sentence as a row of data, and arrange the matching results of all text sentences in sequence according to the time series and the part of the text sentence, and save them in . query and processing.

Optionally, the method further includes: extracting and saving text sentences that are not correctly matched by any attribute (including text sentences that match attributes but do not match attribute values). In the above manner, the matching degree of text sentences can be grasped. Among them, for each part of the text sentence that does not match the attribute value can be saved as: text sentence, text start position, text end position, medical record folder number, medical record number; for each part of the text sentence that does not match the attribute value can be Save as: text sentence, matched attributes, text start position, text end position, medical record folder number, medical record number. The saved format is preferably .xls format.

After extracting text sentences that are not correctly matched by any attribute, these text sentences can also be processed by word segmentation, sorting, manual screening, etc. to find the lack of keyword or attribute classification in the second medical knowledge base, and pass The operations such as adding/deleting/adjusting the score of keywords are performed to realize the iterative optimization of the second medical knowledge base, thereby further improving the matching rate and accuracy of the second medical knowledge base to the text sentences of the electronic medical record.

From the description of the above embodiments, those skilled in the art can clearly understand that the method according to the above embodiment can be implemented by means of software plus a necessary general hardware platform, and of course can also be implemented by hardware, but in many cases the former is better implementation. Based on this understanding, the technical solutions of the present invention essentially or the parts that contribute to the prior art can be embodied in the form of software products, and the computer software products are stored in a storage medium (such as ROM/RAM, magnetic disk, CD-ROM), including several instructions to make a terminal device (which may be a mobile phone, a computer, a server, or a network device, etc.) to execute the methods described in the various embodiments of the present invention.

The electronic medical record structuring method of the embodiment of the present invention described in conjunction with FIG. 1 may be implemented by an electronic medical record structuring device. FIG. 2 shows a schematic diagram of a hardware structure of an electronic medical record structuring device provided by an embodiment of the present invention.

The electronic medical record structuring device may include a processor 21 and a memory 22 storing computer program instructions.

Specifically, the above-mentioned processor 21 may include a central processing unit (CPU), or a specific integrated circuit (Application Specific Integrated Circuit, ASIC), or may be configured as one or more integrated circuits implementing the embodiments of the present invention.

Memory 22 may include mass storage for data or instructions. By way of example and not limitation, memory 22 may include a Hard Disk Drive (HDD), a floppy disk drive, flash memory, optical disk, magneto-optical disk, magnetic tape or Universal Serial Bus (USB) drive or two or more A combination of more than one of the above. Memory 22 may include removable or non-removable (or fixed) media, as appropriate. Where appropriate, memory 22 may be internal or external to the data processing device. In certain embodiments, memory 22 is non-volatile solid state memory. In particular embodiments, memory 22 includes read only memory (ROM). Where appropriate, the ROM may be a mask programmed ROM, programmable ROM (PROM), erasable PROM (EPROM), electrically erasable PROM (EEPROM), electrically rewritable ROM (EAROM) or flash memory or A combination of two or more of the above.

The processor 21 reads and executes the computer program instructions stored in the memory 22 to implement any one of the electronic medical record structuring methods in the foregoing embodiments.

In one example, the electronic medical record structuring device may also include a communication interface 23 and a bus 20 . Among them, as shown in FIG. 2 , the processor 21 , the memory 22 , and the communication interface 23 are connected through the bus 20 to complete the mutual communication.

The communication interface 23 is mainly used to implement communication between modules, apparatuses, units and/or devices in the embodiments of the present invention.

The bus 20 includes hardware, software, or both, coupling the components of the electronic medical record structuring device to each other. By way of example and not limitation, the bus may include an Accelerated Graphics Port (AGP) or other graphics bus, Enhanced Industry Standard Architecture (EISA) bus, Front Side Bus (FSB), HyperTransport (HT) Interconnect, Industry Standard Architecture (ISA) Bus, Infiniband Interconnect, Low Pin Count (LPC) Bus, Memory Bus, Microchannel Architecture (MCA) Bus, Peripheral Component Interconnect (PCI) Bus, PCI-Express (PCI-X) Bus, Serial Advanced Technology Attachment (SATA) bus, Video Electronics Standards Association Local (VLB) bus or other suitable bus or a combination of two or more of these. Where appropriate, bus 20 may include one or more buses. Although embodiments of the present invention describe and illustrate a particular bus, the present invention contemplates any suitable bus or interconnect.

The electronic medical record structuring device may execute the electronic medical record structuring method in the embodiment of the present invention based on the acquired data, thereby realizing the electronic medical record structuring method described in conjunction with FIG. 1 .

In addition, in combination with the electronic medical record structuring method in the foregoing embodiments, the embodiments of the present invention may provide a computer-readable storage medium for implementation. Computer program instructions are stored on the computer-readable storage medium; when the computer program instructions are executed by the processor, any one of the electronic medical record structuring methods in the foregoing embodiments is implemented.

In order to make the description of the embodiments of the present invention clearer, the following descriptions and explanations are made with reference to the preferred embodiments.

This preferred embodiment provides a method for structuring an electronic medical record. FIG. 3 is a flowchart of a method for structuring an electronic medical record according to a preferred embodiment of the present invention. As shown in FIG. 3 , the flowchart includes the following steps:

Step 1: Build the first medical knowledge base.

In this preferred embodiment, building the first medical knowledge base based on the second medical knowledge base includes the following steps:

1. The format of the second medical knowledge base is defined in Table 1, the detailed description of the "requirements" in Table 1 is given in Table 2, and Figure 4 is a schematic diagram of an example of the structure of the first medical knowledge base in the field of prosthodontics.

Table 1 A second medical knowledge base format

Table 2 Details of "Requirements" in Table 1

Require illustrate radio The default is "unknown", when the attribute name score >= 1, select the option with the highest score single choice * Choose the option with the highest score Multiple choice Select all options with a score >=1 judge The default is "None", when the attribute name score >=1, and there is no negative word, select "Yes" number Choose the option (unit) with the highest score and find the number before the unit in the sentence time Choose the option (unit) with the highest score and find the time word before the unit in the sentence radio/numeric When attribute name score >=1, choose the option with the highest score, or find the number

2. Perform manual word segmentation on all the phrases in Figure 5, and screen the medical records by sampling to obtain frequently occurring keywords (including synonyms, synonyms, abbreviations, typos, etc.);

3. Add keywords to the back of each matching object in the second medical knowledge base to form the first medical knowledge base. And give different scores according to the different importance and part of speech corresponding to the keywords (for example, professional terms are positive, negative words are negative, and common words are 0). At the same time, the "and or not" relationship is also realized through scores: for example, since the score is greater than or equal to 1, the matching is successful, so if two keywords are required to appear at the same time, the scores of the two words can be set to 0.5 respectively. as shown in Table 3.

Table 3 A format of the first medical knowledge base

Step 2: Clause the first electronic medical record.

In most cases, a short sentence (delimited by commas) in the electronic medical record corresponds to a set of "attribute-attribute-value". Therefore, the electronic medical records are divided according to punctuation.

1. The entire medical record text is divided into sentences according to Chinese and English commas, periods, line breaks and tabs.

2. Handle special cases of division (such as decimal point, serial number, etc.)

Step 3: Define the structured format.

1. The basic format of the structured target is: text sentence, attribute, attribute value, position, part, and the corresponding position of the text in the electronic medical record. For the attributes that need to be added to the time series, the target format is: text sentence, attribute, attribute value, position, step, substep, part, and the corresponding position of the text in the electronic medical record.

2. Take this as a line, arrange the entire medical record file by sentence, and save it in .csv format.

Step 4: Match text sentences with the first medical knowledge base.

1. For each text sentence, iterate over all attributes. For each attribute, set the matching score of the attribute name and the matching score of each option of the attribute value to an initial value of 0.

2. Match the attribute name, attribute value, and location of the attribute. The specific matching process is as follows:

a) attribute name matching

According to the keyword group corresponding to the attribute name and the text sentence, if the matching is successful, the score will be accumulated (positive keywords plus, negative keywords minus), and the total score of all keywords matching the attribute name will be obtained. If the score exceeds a certain threshold, it is considered that the text sentence matches the attribute name of this attribute successfully, and the attribute value is matched.

b) Option attribute value matching

For each option of the attribute value, match the corresponding keyword group with the text sentence, and if successful, accumulate scores to obtain the total score of all keyword matches of the option. For single-choice attributes, the option with the highest cumulative score is taken as the attribute value of the attribute; for multiple-choice attributes, all options with cumulative scores exceeding a certain threshold are taken as the attribute value of the attribute; for judgmental attributes, if the cumulative score of the option is If it exceeds a certain threshold, the attribute value is considered to be successfully matched.

c) Numeric attribute value matching

Loop judgment on each character in the text sentence, find out the continuous string that expresses the numerical value, and convert it into a numerical type as the attribute value of the attribute.

d) Position matching

If the attribute is related to the tooth position, use the regular expression to match the tooth position in the text sentence (three consecutive '/' as the feature) as the value of the attribute position. If there are multiple options in the position, the same method of option attribute value matching is adopted to match each option of the position, and the option whose cumulative score meets the requirements is selected as the position value according to the different requirements of the position value.

3. According to the score matching criteria corresponding to different requirements shown in Table 3, determine whether the "attribute-attribute value" meets the requirements. If it is satisfied, save the text sentence and the corresponding "attribute-attribute value" pair according to the format in step 3; if not, enter the next attribute for matching.

4. For a situation where multiple attributes may be successfully matched, save each successful matching result.

5. Extract time series information in text information.

Since the different operations in the treatment planning section are sequenced, they need to be reflected in the structured results. For each text sentence in the treatment plan section, find the sequence number of the step at the beginning of the text sentence as the operation order of the corresponding attribute of the sentence.

Since there will be multiple options in each step, they also need to be reflected in the structured results. For each such text sentence, it is judged whether there is a word representing an "or" relationship in the text sentence, and if so, it is separated, and attribute matching is performed respectively.

6. The matching based on the matching scoring algorithm can fully extract the information in the text sentence. In most cases, one text sentence corresponds to one attribute; for the case that one text sentence corresponds to multiple attributes, these attributes can also be matched according to the algorithm logic. Since the medical knowledge base involved in the present invention includes value descriptions of various types such as Boolean type, real number type, classification type, etc., and words with positive and negative semantics are added to the keyword group, the matching algorithm can not only correctly identify disease information At the same time, it can also extract the specific numerical information of the disease information (indicating the severity of the disease, measurement values, etc.), which cannot be achieved by other structural methods at present.

Using keyword groups to match text sentences can identify various types of information in text sentences, including positive and negative semantics, different options for attribute values, numerical values, etc., which greatly expands the applicability of this method.

Step 5: Save the text sentences that are not completely matched during the matching process.

1. The format of the incompletely matched file is: text sentence, matched attribute*, text start position, text end position, medical record folder number, medical record file number.

2. Use this as a line of content, arrange all the unmatched sentences in the medical record files by sentence, and save them in .xls format.

3. For each text sentence in each medical record file, check its match. If the text sentence does not meet the conditions for successful matching, it will be saved in the .xls table of the corresponding part.

Structured Results Analysis

In this embodiment, a set of tools for structuring medical record texts are developed by using the python language for the above electronic medical record structuring method, and more than 3,000 electronic medical record texts are structured. The display and analysis statistics of this result will be given below.

The medical record text processed in this embodiment comes from the related medical records of dentition defects in the prosthodontics department, and the medical knowledge base used is sorted out based on the relevant knowledge in the field of prosthodontics. Part of the knowledge base is shown in FIG. 4 . An example of medical record text is shown in Figure 5, and the structured result is shown in Figure 6.

From the structured results, this method achieves the following beneficial effects:

1. Be able to accurately identify the position information that appears in the medical record text, including the position and tooth position information in text such as "upper jaw" and "lower jaw".

2. Attributes and corresponding attribute values in the medical record text can be effectively marked, and different types of attribute values can be effectively identified.

3. The sequence between different treatment measures in the text can be effectively extracted.

Compared with several existing medical record structuring methods, the method involved in the embodiment of the present invention builds a more comprehensive first medical knowledge base, which can be more suitable for the medical record text, and can also more completely extract the medical record text. Information. However, the existing methods, such as the structured methods based on semantic pros and cons, can only give positive/negative judgments to the medical professional words described in the knowledge base according to the text, but cannot give more comprehensive information (such as the onset of the disease) to this attribute. location, extent, etc.).

The first medical knowledge base used in this embodiment includes 12 parts and a total of 389 attributes. The attribute value can be of multiple choice, single choice judgment, numerical value and other types, and the attribute position value can be of single choice, tooth position and other types. Figure 7 shows the frequency statistics of some attributes in the structured result of this example, and the statistics about the difference of different values of attributes are not reflected in the data. As can be seen from Figure 7, among the more than 3,000 medical records, there is a large gap in the frequency of occurrence of different attributes, which reflects that in addition to some common diseases in the medical records, it also provides a statistical method for us to understand diseases. method.

By randomly extracting a certain number of medical records, manually labeling them against the first medical knowledge base, and using this as a standard to measure the effect of the structured results given by this method, it shows that the electronic medical record structuring method provided by the embodiment of the present invention can complete the first medical Structured tasks required in the knowledge base.

The above descriptions are only preferred embodiments of the present invention, and are not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and changes. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included within the protection scope of the present invention.

Claims (8)

1. An electronic medical record structuring method is characterized by comprising the following steps:

loading a first medical knowledge base, wherein the first medical knowledge base comprises a plurality of parts, and each part comprises one or more attributes and one or more keywords corresponding to the attributes; each attribute includes at least: attribute name, attribute value, position and value type/mode description, each keyword further includes: a score for the keyword;

the first electronic medical record is divided into sentences according to special symbols to obtain a plurality of text sentences;

matching, using a matching scoring algorithm, attributes in the first medical knowledge base for each of the plurality of text sentences, comprising:

matching the attribute name keywords and the scores of the attribute names of each text sentence to obtain the total score of the attribute names of each text sentence corresponding to each attribute;

matching the text sentences with the attribute name total scores of the attributes higher than the preset threshold value with the keywords and the scores of the keywords and the positions in the attributes to obtain the attribute value scores and the position scores of the attributes corresponding to the attributes in the text sentences;

taking the attribute value, the position and the corresponding attribute corresponding to the highest attribute value score and the highest position score as the matching result of the text sentence;

and storing the matching result.

2. The method of claim 1, wherein the type of attribute value comprises at least one of: real number type, boolean type, discrete classification type; the attribute value is selected in at least one of the following manners: judgment, single selection, digit, multiple selection.

3. The method of claim 1, wherein the special symbol comprises at least one of: chinese and English commas, periods, line feed symbols and tab symbols.

4. The method of claim 1, wherein prior to loading the first medical knowledge base, the method further comprises:

loading a second medical knowledge base, wherein the second medical knowledge base comprises a plurality of parts, and each part comprises one or more attributes; each attribute includes at least: attribute name, attribute value and location;

extracting keywords and scores thereof according to the second medical knowledge base and the second electronic medical record;

and constructing the first medical knowledge base according to the second medical knowledge base, the extracted keywords and the scores of the keywords.

5. The method of claim 4, wherein extracting keyword names and keyword scores from the second medical knowledge base and the second electronic medical record comprises:

performing word segmentation on the second electronic medical record according to the attribute names and the attribute values to obtain a plurality of keywords, and taking the similar words and the synonyms of the keywords as the keywords;

different scores are given to the keywords according to the importance, the negativity and the weight of the logical relationship.

6. The method of claim 1, wherein the matching result comprises: the text sentence, and the corresponding attribute, attribute value, position, belonging part and position of the text sentence in the first electronic medical record.

7. The method according to claim 4 or 5, characterized in that the method further comprises:

extracting and storing text sentences which are not correctly matched by any attribute;

performing word segmentation, sequencing and manual screening on the extracted text sentences, and comparing to find the defects of keyword or attribute classification in the second medical knowledge base;

and performing addition/deletion/score adjustment operation on the keywords to realize iterative optimization of the second medical knowledge base.

8. A computer-readable storage medium having computer program instructions stored thereon, which when executed by a processor implement the method of any one of claims 1-7.

Images