CN115759060A - Structured medical record generation method, terminal and readable storage medium - Google Patents
Structured medical record generation method, terminal and readable storage medium Download PDFInfo
- Publication number
- CN115759060A CN115759060A CN202211021521.6A CN202211021521A CN115759060A CN 115759060 A CN115759060 A CN 115759060A CN 202211021521 A CN202211021521 A CN 202211021521A CN 115759060 A CN115759060 A CN 115759060A
- Authority
- CN
- China
- Prior art keywords
- medical record
- information
- structured
- word segmentation
- character information
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Abstract
The invention discloses a method for generating a structured medical record, which comprises the following steps: when receiving the medical record voice information, converting the medical record voice information into medical record character information; identifying characteristic keywords in the medical record text information; dividing the characteristic keywords according to preset medical record subject items; and outputting corresponding characteristic keywords aiming at each medical record topic item to generate a structured medical record. The invention also provides a terminal and a readable storage medium. The structured medical record generation method obtains the medical record voice information by means of voice inputting of the medical record information of the patient, further converts the medical record voice information into the medical record character information, improves the medical record inputting efficiency, in addition, identifies the characteristic key words in the medical record character information, divides the key words aiming at each medical record topic item in the preset medical record topic items and outputs the key words corresponding to the medical record topic items, so that the formed structured medical record has high quality, can be directly retrieved through the key words, is convenient to browse, and has good ductility.
Description
Technical Field
The invention relates to the technical field of electronic medical records, in particular to a structured medical record generation method, a terminal and a computer readable storage medium.
Background
At present, medical records are widely applied in the fields of clinical scientific research, public health big data, hospital data intelligent management, intelligent follow-up visit and the like. The traditional electronic patient medical record entry is usually stored in a computer in an unstructured medical record form by adopting a keyboard entry, or a blank space to be filled in an electronic medical record template is selected by a mouse through a mouse and a keyboard, and the patient medical record entry is input through the keyboard, so that the operation is complicated.
The above is only for the purpose of assisting understanding of the technical aspects of the present invention, and does not represent an admission that the above is prior art.
Disclosure of Invention
The invention mainly aims to provide a structured medical record generation method, a terminal and a readable storage medium, and aims to solve the problem that the traditional electronic medical record entry for patients is complicated to operate through manual entry of a keyboard or combined entry of a mouse and a keyboard.
In order to achieve the above object, the present invention provides a method for generating a structured medical record, wherein the method for generating a structured medical record comprises:
when receiving the medical record voice information, converting the medical record voice information into medical record character information;
identifying characteristic keywords in the medical record text information;
dividing the characteristic keywords according to preset medical record subject items;
and outputting corresponding characteristic keywords for each medical record topic item to generate the structured case.
Optionally, the step of identifying the feature keyword in the medical record text information includes:
performing word segmentation on the medical record text information;
and identifying characteristic keywords from the segmented medical record character information.
Optionally, the step of segmenting the medical record text information includes:
identifying the medical record character information to generate word segmentation character information to be selected;
when the word segmentation text information to be selected is multiple, respectively obtaining the weighted values of the word segmentation text information to be selected;
and taking the word segmentation text information with the highest weight value to be selected as the medical record text information after word segmentation.
Optionally, after the step of outputting a corresponding feature keyword for each medical record topic item to generate the structured case, the method for generating a structured medical record further includes:
when a modification instruction is detected, determining a subject item of a medical record to be modified and a target feature keyword according to the modification instruction;
and modifying the characteristic keywords of the medical record subject items to be modified according to the target characteristic keywords.
Optionally, after the step of modifying the medical record text information of the medical record subject item to be modified according to the target feature keyword, the method for generating a structured medical record further includes:
acquiring medical record text information corresponding to the medical record subject item to be modified;
when the target characteristic keywords are matched with the medical record character information, obtaining word segmentation character information to be selected corresponding to the medical record character information;
and when the target characteristic keywords are matched with the participles in the participle character information to be selected, increasing the weight value of the participle character information to be selected.
Optionally, when the target feature keyword is matched with the medical record text information, after the step of obtaining the word segmentation text information to be selected corresponding to the medical record text information, the method for generating the structured medical record includes:
when the target characteristic keywords are not matched with the participles in the participle character information to be selected, dividing the medical record character information according to the characteristic keywords corresponding to the modified medical record subject item to generate the participle character information to be selected, and increasing the weight value of the participle character information to be selected.
Optionally, after the step of modifying the medical record text information of the medical record subject item to be modified according to the target feature keyword, the structured medical record generating method includes:
when the target characteristic keywords are not matched with the medical record text information, acquiring recognition candidate words in the process of converting the medical record voice information into the medical record text information;
and when the target characteristic key words are matched with any one of the identification candidate words, increasing the weight value of the matched identification candidate words.
Optionally, after the step of identifying the to-be-selected word in the process of acquiring and identifying the medical record voice information as the medical record text information, the method for generating the structured medical record includes:
and outputting voice recognition error verification prompt information when the target characteristic key words are not matched with the recognition candidate words.
In addition, to achieve the above object, the present invention also provides a terminal, including: the system comprises a memory, a processor and a generation program of the structured medical record stored in the memory and capable of running on the processor, wherein the generation program of the structured medical record realizes the steps of the generation method of the structured medical record when being executed by the processor.
In addition, to achieve the above object, the present invention further provides a readable storage medium, which stores a generation program of a structured medical record, and the generation program of the structured medical record realizes the steps of the generation method of the structured medical record as described above when executed by the processor.
The method for generating the structured medical record can obtain the medical record voice information by a mode of inputting the medical record information of a patient through voice, further convert the medical record voice information into the medical record character information and improve the efficiency of inputting the medical record, in addition, the characteristic key words in the medical record character information are identified, the key words are divided aiming at each medical record theme item in the preset medical record theme items and the key words are output corresponding to the medical record theme items, and the formed structured medical record content is high in concise quality, can be directly retrieved through the key words, is convenient to browse, and is good in ductility.
Drawings
Fig. 1 is a block diagram of a terminal according to various embodiments of a method for generating a structured medical record of the present invention;
FIG. 2 is a flowchart illustrating a first embodiment of a method for generating a structured medical record according to the present invention;
FIG. 3 is a flowchart illustrating a process of segmenting textual information of a medical record according to a first embodiment of a method for generating a structured medical record of the present invention;
FIG. 4 is a flowchart illustrating a method for generating a structured medical record according to a second embodiment of the present invention;
FIG. 5 is a flowchart illustrating a method for generating a structured medical record according to a third embodiment of the present invention;
fig. 6 is a flowchart illustrating a method for generating a structured medical record according to a fourth embodiment of the present invention.
The implementation, functional features and advantages of the present invention will be further described with reference to the accompanying drawings.
Detailed Description
It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.
In the following description, suffixes such as "module", "component", or "unit" used to denote elements are used only for facilitating the explanation of the present invention, and have no specific meaning in itself. Thus, "module", "component" or "unit" may be used mixedly.
The terminal may be implemented in various forms. For example, the terminal device described in the present invention may include a mobile terminal such as a mobile phone, a tablet computer, a notebook computer, a palmtop computer, a Personal Digital Assistant (PDA), and a Portable Media Player (PMP).
It will be understood by those skilled in the art that the configuration according to the embodiment of the present invention can be applied to a fixed type mobile terminal, in addition to elements particularly used for mobile purposes.
Referring to fig. 1, fig. 1 is a block diagram of a terminal according to various embodiments of a method for generating a structured medical record of the present invention, where the terminal may include: memory 101, processor 102, and display unit 103. Those skilled in the art will appreciate that the block diagram of the terminal shown in fig. 1 does not constitute a limitation of the terminal, and that the terminal may include more or less components than those shown, or may combine certain components, or a different arrangement of components. The memory 101 stores a central control system and a program for generating a structured medical record. The processor 102 is a control center of the terminal, and the processor 102 executes a program for generating a structured medical record stored in the memory 101 to implement the steps of the embodiments of the method for generating a structured medical record of the present invention. The Display unit 103 includes a Display panel, which may be configured in the form of a Liquid Crystal Display (LCD), an Organic Light-Emitting Diode (OLED), and the like, for displaying an interface viewed by a user. The display unit 103 may be integrated with a touch panel, and when the touch panel detects a touch operation of a finger on or near the touch panel, the touch panel transmits the touch operation to the processor 102 to determine the type of the touch event, and then the processor 102 correspondingly implements a function according to the type of the touch event.
Based on the structural block diagram of the terminal, embodiments of the method for generating the structured medical record are provided.
Referring to fig. 2, fig. 2 is a schematic flowchart illustrating a first embodiment of a method for generating a structured medical record according to the present invention. In this embodiment, the method for generating a structured medical record includes the following steps:
step S10, when medical record voice information is received, converting the medical record voice information into medical record character information;
the medical record voice information is the patient medical record information recorded in a voice form. The medical record text information is patient medical record information recorded in a text form. The received medical record voice information can be directly input into patient medical record information through a microphone, or the patient medical record information can be recorded in advance through voice and stored, and the stored patient medical record information is sent through a network or received in a copying mode through an external memory, and the method is not limited herein. The medical record voice information is converted into medical record text information, the medical record voice information can be recognized through the voice recognition text software carried by the terminal to obtain the medical record text information, the medical record voice information can also be sent to the server, the medical record voice information is recognized through the voice recognition text software on the server to obtain the voice text information, and then the medical record text information returned from the server is received.
Step S20, identifying characteristic keywords in the medical record text information;
step S30, dividing the characteristic keywords according to preset medical record subject items;
and S40, outputting corresponding characteristic keywords for each medical record topic item to generate the structured medical record.
The characteristic keywords are keywords which belong to medical record character information and are matched with a pre-stored medical record characteristic word bank. The preset medical record subject items are subject items required to be contained in the structured medical record of the patient, such as: the medical staff can clearly know the medical history information of the patient. Optionally, the medical record subject items are preset to be multiple items. The characteristic keywords in the medical record character information can be identified through a pre-stored medical record characteristic word bank, namely, the keywords matched with the medical record characteristic word bank in the medical record character information are used as the characteristic keywords.
The medical record feature word library can store the keywords in a secondary storage form, for example: the primary storage is used for storing preset medical record subject items, the preset medical record subject items include a plurality of medical record subject items, and the secondary storage is a feature keyword corresponding to each medical record subject item, wherein the secondary storage can be set in a nested manner, and specifically includes the following steps:
1. the current medical history;
1.1 course of treatment; 1.1.1 3 days; 1.1.2 1 week;
1.2, usage; 1.2.1 oral/administration; 1.2.2 injections;
2, physical examination;
2.1 body temperature; 2.1.1 ℃;
2.2 consciousness; 2.2.1 clear; 2.2.2 blur. The primary storage is serial numbers 1 and 2, and the secondary storage is understood to be serial numbers 1.1, 1.2, 2.1, 2.2, 1.1.1, 1.2.1, and the like.
The characteristic keywords in the medical record text information can be obtained through the identification of the characteristic keywords stored in the second stage, and then the obtained characteristic keywords are divided through the medical record subject items stored in the first stage so as to finally obtain the characteristic keywords corresponding to each medical record subject item, wherein the characteristic keywords corresponding to one medical record subject item can be single or multiple; in addition, the medical record text information obtained by identification may also be divided by the feature keywords corresponding to the medical record subject items stored in the first stage, that is, a specific paragraph or sentence in the medical record text information corresponding to each medical record subject item may be determined according to the feature keywords corresponding to the medical record subject item, and the topic item may be labeled on the paragraph or sentence and associated with the corresponding medical record subject item in the structured medical record in a topic item labeling form, which is not limited in this embodiment.
Referring to fig. 3, fig. 3 is a schematic flow chart illustrating word segmentation of medical record text information according to a first embodiment of a method for generating a structured medical record of the present invention. Optionally, step S20 includes:
step S21, performing word segmentation on the medical record character information;
and S22, identifying characteristic keywords from the medical record character information after word segmentation.
Step S21 includes: identifying the medical record character information to generate word segmentation character information to be selected; when the word segmentation information to be selected is multiple, respectively obtaining the weighted values of the word segmentation information to be selected; and taking the word segmentation text information with the highest weight value to be selected as the medical record text information after word segmentation. It should be noted that the word segmentation text information to be selected is word segmentation text information divided according to the association relationship between preceding words and following words in the medical record text information, wherein the word segmentation text information to be selected is one or more kinds of word segmentation text information. For example, the word segmentation text message identifying the text message as "how the student played basketball" may have the following two cases: A. this/student/meeting/basketball/Do; or, b. the/student/basketball/do; wherein, A and B can be regarded as word segmentation information to be selected. It is easy to understand that the participle character information with a high weight value is determined as the participle character information by comparing the weight values of the participle character information corresponding to the a and the B, that is, when the participle character information to be selected is multiple, the weight values of the participle character information to be selected are respectively obtained; and taking the word segmentation text information with the highest weight value to be selected as the medical record text information after word segmentation.
Optionally, after step S40, the structured medical record is output. The structured medical record information of the patient is output and displayed to be checked and confirmed, so that the quality of the structured medical record is improved. In addition, before the step of outputting the structured medical record, the generated structured medical record can be detected, and the medical record subject item with empty content in the structured medical record of the patient is labeled, for example, the medical record subject item and/or the region position of the input content information corresponding to the medical record subject item are highlighted through character color or background color, and then the remarked structured medical record is output to output prompt information for supplementing medical record information.
In the technical scheme disclosed in this embodiment, the medical record voice information can be obtained by inputting the medical record information of a patient through voice, and then the medical record voice information is converted into medical record text information, so that the medical record inputting efficiency is improved.
Referring to fig. 4, fig. 4 is a flowchart illustrating a method for generating a structured medical record according to a second embodiment of the present invention. In this embodiment, step S40 is followed by:
s50, when a modification instruction is detected, determining a subject item of a medical record to be modified and a target feature keyword according to the modification instruction;
and S60, modifying the characteristic keywords of the medical record subject items to be modified according to the target characteristic keywords.
The medical record subject items to be modified are medical record subject items in preset medical record subject items corresponding to the modification operation, wherein the medical record subject items to be modified are one or more medical record subject items. The modification mode of the feature keywords of the medical record subject item to be modified includes, but is not limited to, feature keyword replacement, feature keyword supplement and feature keyword deletion operation, and the target feature keywords can be all the feature keywords corresponding to the medical record subject item to be modified after modification instructions are executed, namely after modification operation; the feature keywords corresponding to the medical record subject item to be modified may also be modified, that is, the original feature keywords corresponding to the medical record subject item to be modified before modification and the new feature keywords corresponding to the medical record subject item to be modified after modification are compared, and the feature keywords in which the original feature keywords do not match with the new feature keywords are used as the target feature keywords, which is not limited here.
The modification instruction may be triggered by clicking a modification button or clicking a modification option, or may be triggered by detecting a preset touch instruction corresponding to the modification instruction, which is not limited to this. It is easy to understand that, corresponding to a preset medical record subject item of a structured medical record and a feature keyword corresponding to any one of the preset medical record subject items, a medical record subject item to be modified can be uniquely determined from the preset medical record subject item through a modification instruction, a target feature keyword after modification operation can be obtained by detecting the modification operation performed on the medical record subject item to be modified by a user in the modification process, and then at least one of feature keyword replacement, feature keyword supplement and feature keyword deletion operation is performed on the feature keyword corresponding to the medical record subject item to be modified, so as to modify the feature keyword of the medical record subject item to be modified.
In the technical scheme disclosed in this embodiment, after the structured medical record is generated, the generated structured medical record can be modified through the modification instruction, the structured medical record is perfected and confirmed to obtain a complete and effective structured medical record, and the quality of the structured medical record is improved.
Referring to fig. 5, fig. 5 is a flowchart illustrating a method for generating a structured medical record according to a third embodiment of the present invention. In this embodiment, step S60 is followed by:
step S70, acquiring medical record character information corresponding to the medical record subject item to be modified;
step S80, when the target characteristic keywords are matched with the medical record character information, obtaining word segmentation character information to be selected corresponding to the medical record character information;
step S90, when the target characteristic keywords are matched with the participles in the participle character information to be selected, increasing the weight value of the participle character information to be selected;
and step S100, when the target characteristic keywords are not matched with the participles in the word information to be selected, increasing the weighted values of the participles which are the same as the target characteristic keywords.
The medical record character information corresponding to the medical record subject item to be modified can be obtained by obtaining the subject item label associated with the medical record subject item to be modified, so that a specific paragraph or sentence corresponding to the medical record subject item to be modified in the medical record character information can be obtained according to the subject item label. In the practical application process, one or more word segmentation character information can be obtained as the word segmentation character information to be selected by segmenting according to the postamble connection before the word in the medical record character information, and the word segmentation character information to be selected with the highest weight value in the multiple word segmentation character information is used as the medical record character information after word segmentation.
However, even if the most preferable information of the word to be selected is obtained, it cannot be guaranteed that the medical record character information after word segmentation completely meets the word segmentation requirements, that is, the feature keywords matched with the pre-stored medical record feature word library are obtained from the medical record character information after word segmentation as much as possible, and whether the most preferable information of the word to be selected has word segmentation errors can be known by comparing the target feature keywords with the medical record character information before word segmentation, that is, the target feature keywords cannot be obtained due to word segmentation errors because the most preferable information of the word to be selected is used for word segmentation of the medical record character information, so that the most preferable information of the analysis to be selected meeting the current word segmentation requirements can be verified.
The method is easy to understand, when a target characteristic keyword is matched with medical record character information, namely, most preferable word segmentation information to be selected has word segmentation errors, word segmentation character information to be selected corresponding to the medical record character information is obtained, wherein the word segmentation character information to be selected is one or more kinds of word segmentation character information which is divided according to the preceding and following relations of words in the medical record character information, when the word segmentation character information to be selected is various kinds of word segmentation character information, the target characteristic keyword can be respectively matched with the various kinds of word segmentation character information, when the target characteristic keyword is matched with words in the word segmentation character information of the various kinds of word segmentation character information, the weight value of the word segmentation character information to be selected, namely the matched word segmentation character information is increased, the weight value of the matched word segmentation character information is increased to meet word segmentation requirements, accordingly, the medical record character information is optimally identified to generate word segmentation character information to be selected, and the word segmentation character information with the largest weight and capable of meeting the word segmentation requirements is obtained from the word segmentation character information to be selected. The weight value of the word segmentation information to be selected can be increased by adding a preset unit weight value on the basis of the original weight value of the word segmentation information to be selected, which is not limited in this embodiment.
When the target characteristic keyword is not matched with the participle in the participle character information to be selected, namely the participle of any participle character information in the participle character information to be selected is not matched with the target characteristic keyword, so that the weight value of the participle same as the target characteristic keyword can be increased. The weight value of the participle which is the same as the target feature keyword is added, and a preset unit weight value is added on the basis that the original weight value of the participle is 0, which is not limited in this embodiment.
To facilitate understanding of this embodiment, the following example is used for illustration.
Corresponding to the step S70, acquiring medical record text information corresponding to the medical record subject item to be modified;
and modifying the medical record character information corresponding to the medical record subject item to be 'amoxicillin eaten'.
The first condition is as follows: and identifying the medical record character information as the word segmentation character information to be selected has errors.
After speech recognition, the amoxicillin can be recognized as 'eating';
the word information for identifying the medical record to generate the word segmentation to be selected can be 'eating amoxicillin', and the user modifies 'amoxicillin' into a target characteristic keyword 'amoxicillin' after modifying.
Step S80, when the target characteristic keywords are matched with the medical record character information, obtaining word segmentation character information to be selected corresponding to the medical record character information;
the target feature keyword is not matched with the medical record text information, if corresponding to the condition one, the target feature keyword ' amoxicillin ' is matched with ' amoxicillin ' in the medical record text information ' amoxicillin ' eaten '. Acquiring word segmentation text information to be selected, namely 'eating amoxicillin', corresponding to medical record text information 'eating amoxicillin'.
Step S90, when the target characteristic keywords are matched with the participles in the participle character information to be selected, increasing the weight values of the participle character information to be selected;
for example, if the medical record text information is recognized to generate word segmentation text information to be selected as 'eaten/amoxicillin', the target feature keyword 'amoxicillin' is matched with one of the word segmentation in the word segmentation text information to be selected as 'eaten' and 'amoxicillin', and the weight value of the word segmentation 'amoxicillin' and/or the word segmentation 'eaten' in the word segmentation text information to be selected is increased.
And step S100, when the target characteristic keywords are not matched with the participles in the word information to be selected, increasing the weighted values of the participles which are the same as the target characteristic keywords.
The target characteristic keyword amoxicillin is not matched with any one of the word segmentations 'eating A' and 'moxacillin' in the word segmentation information to be selected, the weight value of the word segmentations 'eating A' in the word segmentation information to be selected is reduced, and/or the weight value of the word segmentations corresponding to the target characteristic keyword amoxicillin is increased, so that when the subsequent execution of recognizing the medical record word information as the word segmentation information to be selected is carried out, the recognition accuracy is improved.
In the technical scheme disclosed in this embodiment, a target feature keyword is determined by a modification operation performed on a medical record subject item to be modified, whether a word segmentation error exists in the most preferable word segmentation information to be selected, which is obtained from the word segmentation information to be selected corresponding to the medical record subject item to be modified, is determined by comparing the target feature keyword with the medical record word information corresponding to the medical record subject item to be modified, that is, the most preferable word segmentation information to be selected is used for performing word segmentation on the medical record word information, the target feature keyword cannot be obtained due to a word segmentation error, when the word segmentation error exists, the word segmentation word information to be selected is generated by optimizing and identifying the medical record word information by increasing the weighted value of the word segmentation information to be selected, and the purpose of obtaining the word segmentation word information with the largest weighted value and capable of meeting the word segmentation requirement from the word segmentation word information to be selected can be achieved by repeatedly executing an optimization process, that the feature keyword matched with a pre-stored feature word library is obtained from the medical record word information after word segmentation as much as possible.
Based on the second embodiment or the third embodiment, a fourth embodiment of the method for generating a structured medical record of the present invention is provided, please refer to fig. 6, and fig. 6 is a flowchart illustrating the fourth embodiment of the method for generating a structured medical record of the present invention. In this embodiment, step S60 is followed by:
step S70, acquiring medical record character information corresponding to the medical record subject item to be modified;
step S110, when the target characteristic keywords are not matched with the medical record text information, acquiring identification candidate words in the process of converting the medical record voice information into the medical record text information in the step S10;
step S120, when the target characteristic key words are matched with any recognition candidate word in the recognition candidate words, increasing the weight value of the matched recognition candidate words;
step S130, outputting voice recognition error verification prompt information when the target characteristic keywords are not matched with the recognition candidate words.
The medical record character information corresponding to the medical record subject item to be modified can be obtained by obtaining the subject item label associated with the medical record subject item to be modified, so that a specific paragraph or sentence corresponding to the medical record subject item to be modified in the medical record character information can be obtained according to the subject item label. The target characteristic keywords can be all characteristic keywords corresponding to the subject items of the medical records to be modified after the modification instruction is executed, namely after the modification operation is carried out; or the modified characteristic keyword corresponding to the medical record subject item to be modified, that is, the characteristic keyword in which the original characteristic keyword does not match with the new characteristic keyword is taken as the target characteristic keyword by comparing the original characteristic keyword corresponding to the medical record subject item to be modified before modification with the new characteristic keyword corresponding to the modified medical record subject item to be modified.
In the practical application process, in the process of generating the structured medical record, after the medical record voice information is input through the microphone, the medical record voice information is converted into medical record text information, and then the characteristic keywords are identified based on the medical record text information and are divided according to preset medical record subject items so as to output corresponding characteristic keywords aiming at each medical record subject item, when the target characteristic keywords are not matched with the medical record text information, namely when the target characteristic keywords cannot be acquired due to word segmentation errors in generating the word segmentation text information to be selected for the medical record text information, the medical record voice information is acquired and identified as the identification word to be selected in the process of generating the medical record text information, wherein, the method comprises the steps of identifying a single or a plurality of words to be selected, identifying medical record voice information as medical record character information, namely identifying the words to be selected with the highest weight value as the medical record character information, and identifying the medical record voice information as the medical record character information, wherein the target characteristic keyword is a modified characteristic keyword, when the words to be selected are identified as a plurality of words to be selected and the target characteristic keyword is matched with any one of the identified words to be selected, namely the identified words with the highest weight value do not meet identification requirements, and then increasing the weight value of the matched identified words to be selected so as to improve the selection probability of the matched identified words to be selected, so that the medical record voice information is converted into the medical record character information to be optimized so as to meet the identification requirements. It can be understood that when the target feature keyword and the recognition candidate word are not matched, the voice recognition error verification prompt information is output, and the prompt information can be at least one of characters, voice and vibration.
It should be noted that, the to-be-selected word is identified as a selectable identifying word for converting the medical record text information into the medical record text information, for example, when the medical record voice information is "enriching", the identified medical record text information may be "taking", or "multiplexing", and then "taking" and "multiplexing" are identified as the to-be-selected word.
To facilitate understanding of this embodiment, the following example is used for illustration.
Corresponding to the step S70, acquiring medical record text information corresponding to the medical record subject item to be modified;
and modifying the medical record character information corresponding to the medical record subject item to be 'amoxicillin eaten'.
Case two: the medical record voice information is identified as medical record character information with errors.
After speech recognition, "riding amoxicillin" can also be recognized;
the word information for identifying the medical record to generate the word segmentation to be selected can be 'riding/amoxicillin', and the user modifies 'riding' into a target feature keyword 'eating' after correcting.
Step S110, when the target characteristic keywords are not matched with the medical record text information, acquiring identification candidate words in the process of converting the medical record voice information into the medical record text information in the step S10;
the target characteristic keyword is not matched with the medical record text information, if corresponding to the second condition, any word in the target characteristic keyword 'eaten' and the medical record text information 'riding amoxicillin' is not matched.
In the step S10 of acquiring, the identification candidate in the process of converting the medical record voice information into the medical record text information includes, but is not limited to, "eat", "ride" and "hesitation".
Wherein, the weight value for identifying the word to be selected is the highest weight value in all the words to be selected.
Step S120, when the target characteristic key words are matched with any recognition candidate word in the recognition candidate words, increasing the weight value of the matched recognition candidate words;
the target feature keyword 'eaten' is matched with the recognized candidate words 'eaten', 'ridden' and 'stumbled' and 'eaten', and the weight value of the matched recognized candidate words is increased, namely the weight value of the recognized candidate words 'eaten' is increased. Optionally, the weight value for identifying the word to be selected is decreased while the weight value for identifying the word to be selected is increased.
Matching the target characteristic keyword with medical record text information, if corresponding to the condition I, matching the target characteristic keyword 'amoxicillin' with 'amoxicillin' in the medical record text information 'eating amoxicillin', and acquiring word segmentation text information to be selected corresponding to the medical record text information, namely 'eating amoxicillin';
step S130, outputting voice recognition error verification prompt information when the target characteristic keywords are not matched with the recognition candidate words.
For example, if the candidate word is recognized as "riding" and "being out of date", it indicates that the target feature keyword "eating" does not match the candidate word, and the voice recognition error may be prompted by outputting a voice recognition error verification prompt message, alternatively, "eating" may be added to the recognized candidate word, and a weight value, such as an increased unit weight value, may be assigned to the recognized candidate word "eating".
In the technical scheme disclosed in this embodiment, a target feature keyword is determined by a modification operation performed on a medical record subject item to be modified, and by comparing the target feature keyword with medical record text information corresponding to the medical record subject item to be modified, when the target feature keyword is not matched with the medical record text information corresponding to the medical record subject item to be modified, it is determined whether an error exists in a process of converting medical record voice information into medical record text information, where the identification word with the highest weight value is adopted to identify whether an error exists in the selected word to be selected, that is, the target feature keyword cannot be obtained by converting the medical record voice information into the medical record text information because the selected word with the highest weight value is adopted to identify the selected word, and when the error exists, the medical record voice information is optimally converted into the medical record text information by increasing the weight value of the identified selected word matched with the target feature keyword, and by repeatedly executing an optimization process, the purpose of obtaining the largest weight value from the identified selected word and meeting the voice conversion requirement can be obtained, that the feature keyword matched with a pre-stored feature word library is obtained from the identified selected word as much as possible.
The invention further provides a terminal, which comprises a memory, a processor and a generating program of the structured medical record, wherein the generating program of the structured medical record is stored in the memory and can run on the processor, and when being executed by the processor, the generating program of the structured medical record realizes the steps of the generating method of the structured medical record in any embodiment.
The invention further provides a readable storage medium, wherein a structured medical record generation program is stored on the readable storage medium, and when being executed by a processor, the steps of the structured medical record generation method in any embodiment of the invention are realized.
In the embodiments of the terminal and the readable storage medium provided by the present invention, all technical features of the embodiments of the method for generating a structured medical record are included, and the contents of the expansion and the explanation of the specification are basically the same as those of the embodiments of the method for generating a structured medical record, and are not described herein again.
It should be noted that, in this document, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or system that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or system. Without further limitation, an element defined by the phrase "comprising a … …" does not exclude the presence of another identical element in a process, method, article, or system that comprises the element.
The above-mentioned serial numbers of the embodiments of the present invention are merely for description and do not represent the merits of the embodiments.
Through the above description of the embodiments, those skilled in the art will clearly understand that the method of the above embodiments can be implemented by software plus a necessary general hardware platform, and certainly can also be implemented by hardware, but in many cases, the former is a better implementation manner. Based on such understanding, the technical solution of the present invention may be embodied in the form of a software product, which is stored in a storage medium (such as ROM/RAM, magnetic disk, optical disk) and includes instructions for enabling a mobile terminal (such as a mobile phone, a computer, a server, a controlled terminal, or a network device) to execute the method of each embodiment of the present invention.
The above description is only a preferred embodiment of the present invention, and not intended to limit the scope of the present invention, and all modifications of equivalent structures and equivalent processes, which are made by using the contents of the present specification and the accompanying drawings, or directly or indirectly applied to other related technical fields, are included in the scope of the present invention.
Claims (10)
1. A structured medical record generation method is characterized by comprising the following steps:
when receiving the medical record voice information, converting the medical record voice information into medical record character information;
identifying characteristic keywords in the medical record text information;
dividing the characteristic keywords according to preset medical record subject items;
and outputting corresponding characteristic keywords aiming at each medical record topic item to generate the structured medical record.
2. The method for generating structured medical records according to claim 1, wherein the step of identifying the feature keyword in the text information of the medical record comprises:
performing word segmentation on the medical record text information;
and identifying characteristic keywords from the segmented medical record character information.
3. The method for generating structured medical records according to claim 2, wherein the step of segmenting the textual information of the medical record comprises:
identifying the medical record character information to generate word segmentation character information to be selected;
when the word segmentation information to be selected is multiple, respectively obtaining the weighted values of the word segmentation information to be selected;
and taking the word segmentation text information with the highest weight value to be selected as the medical record text information after word segmentation.
4. The structured medical record generation method according to claim 3, wherein after the step of outputting the corresponding feature keyword for each of the medical record topic items to generate the structured case, the structured medical record generation method further comprises:
when a modification instruction is detected, determining a subject item of a medical record to be modified and a target feature keyword according to the modification instruction;
and modifying the characteristic keywords of the medical record subject items to be modified according to the target characteristic keywords.
5. The method for generating a structured medical record according to claim 4, wherein after the step of modifying the medical record text information of the subject item of the medical record to be modified according to the target feature keyword, the method for generating a structured medical record further comprises:
acquiring medical record text information corresponding to the medical record subject item to be modified;
when the target characteristic keywords are matched with the medical record character information, obtaining word segmentation character information to be selected corresponding to the medical record character information;
and when the target characteristic keywords are matched with the participles in the participle character information to be selected, increasing the weight value of the participle character information to be selected.
6. The method for generating a structured medical record according to claim 5, wherein after the step of obtaining the word-to-be-selected text information corresponding to the medical record text information when the target feature keyword matches the medical record text information, the method for generating a structured medical record comprises:
and when the target characteristic key words are not matched with the word segmentation in the word segmentation information to be selected, increasing the weight values of the word segmentation same as the target characteristic key words.
7. The method for generating structured medical records according to claim 4, wherein after the step of modifying the feature keyword of the subject item of the medical record to be modified according to the target feature keyword, the method for generating structured medical records comprises:
acquiring medical record text information corresponding to the medical record subject item to be modified;
when the target characteristic keywords are not matched with the medical record character information, acquiring identification candidate words in the process of converting the medical record voice information into the medical record character information, wherein the identification candidate words with the highest weight value in the identification candidate words are used as the medical record character information corresponding to the medical record voice information;
and when the target feature keywords are matched with any recognition candidate word in the recognition candidate words, increasing the weight value of the matched recognition candidate word.
8. The method as claimed in claim 7, wherein after the step of identifying the candidate word in the process of acquiring and identifying the medical record voice information as the medical record text information, the method for generating the structured medical record comprises:
and outputting voice recognition error verification prompt information when the target characteristic keywords are not matched with the recognition candidate words.
9. A mobile terminal, characterized in that the mobile terminal comprises: a memory, a processor, and a structured medical record generation program stored in the memory and executable on the processor, the structured medical record generation program when executed by the processor implementing the steps of the structured medical record generation method of any of claims 1-8.
10. A computer readable storage medium, wherein a structured medical record generation program is stored on the computer readable storage medium, and when executed by a processor, the structured medical record generation program implements the steps of the structured medical record generation method according to any one of claims 1-8.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211021521.6A CN115759060A (en) | 2022-08-24 | 2022-08-24 | Structured medical record generation method, terminal and readable storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211021521.6A CN115759060A (en) | 2022-08-24 | 2022-08-24 | Structured medical record generation method, terminal and readable storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115759060A true CN115759060A (en) | 2023-03-07 |
Family
ID=85349280
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211021521.6A Pending CN115759060A (en) | 2022-08-24 | 2022-08-24 | Structured medical record generation method, terminal and readable storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115759060A (en) |
-
2022
- 2022-08-24 CN CN202211021521.6A patent/CN115759060A/en active Pending
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US11101024B2 (en) | Medical coding system with CDI clarification request notification | |
| US9659055B2 (en) | Structured searching of dynamic structured document corpuses | |
| US8700589B2 (en) | System for linking medical terms for a medical knowledge base | |
| US20200226481A1 (en) | Methods and systems for managing medical information | |
| US11573954B1 (en) | Systems and methods for processing natural language queries for healthcare data | |
| US7672987B2 (en) | System and method for integration of medical information | |
| US11183300B2 (en) | Methods and apparatus for providing guidance to medical professionals | |
| US20040102971A1 (en) | Method and system for context-sensitive recognition of human input | |
| EP2523126A2 (en) | Information processing apparatus, information processing method, program, and information processing system | |
| US20220254507A1 (en) | Knowledge graph-based question answering method, computer device, and medium | |
| US11222031B1 (en) | Determining terminologies for entities based on word embeddings | |
| JP2013543999A (en) | Structured search of dynamic structured document corpus | |
| JP2007034871A (en) | Character input apparatus and character input apparatus program | |
| US20080109400A1 (en) | Method and device for configuring a variety of medical information | |
| CN109871440B (en) | Intelligent prompting method, device and equipment based on semantic analysis | |
| Reeves et al. | Adaptation of an NLP system to a new healthcare environment to identify social determinants of health | |
| CN112825111A (en) | Natural language processing method and computing device thereof | |
| CN113555105A (en) | Method and device for recommending medical products | |
| CN116719840A (en) | Medical information pushing method based on post-medical-record structured processing | |
| CN115759060A (en) | Structured medical record generation method, terminal and readable storage medium | |
| CN112700862B (en) | Determination method and device of target department, electronic equipment and storage medium | |
| US11392628B1 (en) | Custom tags based on word embedding vector spaces | |
| US20170212990A1 (en) | System and method for optimizing electronic medical terminology post-coordination coding | |
| US20220351868A1 (en) | Systems and Methods for Machine Learning From Medical Records | |
| Tsai et al. | Application of Named Entity Recognition by Self-Attention BiLSTM-CRF to Chinese Prescription Document |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |