Artificial intelligence-based quality control method for electronic medical records

Abstract

The present invention discloses an electronic medical record quality control method based on artificial intelligence, which relates to the field of artificial intelligence technology and includes the following steps: inputting text feature information and vocabulary into a constructed model database, then extracting the text information, obtaining the text feature information in the text information, parsing the text feature information and key vocabulary data in the text information, and importing the text feature information and key vocabulary extracted in S2 into a mining system. The advantages of the present invention are: through medical record mining related technologies and analysis through medical record-related data, doctors view medical record text data through a visualization window, and the medical record-related data assists doctors in making judgments on current medical record text data, making it easier for doctors to view problems in medical records and reducing the time spent by doctors on making judgments on medical records, thereby ensuring doctors' judgments on the condition and improving the quality of electronic medical records.

Description

Technical Field

[0001] The present invention relates to the field of artificial intelligence technology, and in particular to an electronic medical record quality control method based on artificial intelligence. Background Art

[0002] With the promotion and development of medical informatization, electronic medical records have become an indispensable part of the daily work of medical institutions. However, the quality of electronic medical records has always been one of the challenges faced by the medical industry, including incomplete, incorrect, duplicate, and non-standard medical records. In order to improve the quality of electronic medical records, reduce medical risks, and lower the incidence of medical accidents, electronic medical record quality control methods based on artificial intelligence have emerged.

[0003] However, due to incomplete and non-standard medical records, artificial intelligence is currently not fully capable of generating corresponding medical plans based on medical records. Doctors need to work together with artificial intelligence to reduce the problems that arise when artificial intelligence controls the quality of electronic medical records. To this end, we propose an electronic medical record quality control method based on artificial intelligence. Summary of the Invention

[0004] The purpose of the present invention is to provide an electronic medical record quality control method based on artificial intelligence.

[0005] To solve the problems raised in the above background technology, the present invention provides the following technical solution: an electronic medical record quality control method based on artificial intelligence, comprising the following steps:

[0006] S1. Establishing a model database: Inputting text feature information and vocabulary into the constructed model database;

[0007] S2. Obtaining medical record text data: extracting text information, obtaining text feature information in the text information, and parsing the text feature information and key word data in the text information;

[0008] S3, privacy protection: import the text feature information and key words extracted in S2 into the mining system;

[0009] S3.1. Enable the MCU and retrieve the key word data identified in S2 and match it with the model database in S1;

[0010] S3.2. After the data matching in S3.1 is completed, the data is transmitted to the mining system. After the mining system completes the processing, the cached data in the microcontroller is cleared;

[0011] S4, medical record mining: parsing the medical record data in S3;

[0012] S4.1. Mining medical record data by setting up data mining technology in the mining system to parse the medical record text data related to the medical record;

[0013] S4.2. Match the medical record associated data with the medical record text data, and generate list data of the medical record text data and the medical record associated data;

[0014] S5, treatment suggestion: by judging the medical record text data and the medical record-related data, by calling the treatment plan of the medical record-related data in the model database, and analyzing the medical record-related data, and combining the analysis results with the medical record text data, the treatment plan in the current medical record text data is parsed based on the medical record text data and the medical record-related data, the generated data is transmitted to S6, and an instruction to clear the cached data is sent to the single-chip microcomputer;

[0015] S6, treatment decision: Obtain the data in S5 and S4 and transfer them to the visualization window. The doctor then compares the medical record text data with the relevant data, and then checks the treatment plan associated with the medical record data and the treatment plan generated by the mining system. The doctor uses his or her own qualifications to compare the two treatment plans to ensure that the treatment plan is feasible. At the same time, the doctor can enter the doctor's modification suggestions and modify the treatment plan in the visualization window based on the comparison of the medical record text data and the relevant data, and at the same time transfer the data entered by the doctor to the model database in S1;

[0016] S7. Medical record tracking: By real-time monitoring of medical record text data, the quality of the medical record text can be ensured. At the same time, by recording abnormal data in S2, S3, S4 and S5, medical staff can be reminded to modify and improve the medical record text.

[0017] As a further solution of the present invention: the specific steps in S1 are as follows:

[0018] S1.1. Establish a model database based on different departments;

[0019] S1.2. Establish a tree diagram based on the medical record data, and connect the model databases of different departments in series based on the tree diagram to ensure that hidden diseases can be deeply analyzed.

[0020] As a further solution of the present invention: the specific steps in S2 are as follows:

[0021] S2.1. Scanning the text on the medical record using image processing technology to obtain text features on the medical record, and transmitting the text feature information to the next step;

[0022] S2.2. Utilize natural speech processing technology to analyze and understand text features. Simultaneously, natural speech processing technology analyzes whether the text feature data identified by image processing technology is abnormal.

[0023] S2.3. Through the interaction of natural speech processing technology and image processing technology, abnormal text feature data is identified and corrected, and omissions are supplemented to ensure the integrity of text information and the accuracy of medical record recognition.

[0024] S2.4. Ensure the integrity of basic medical information, medical history, and diagnosis through natural speech processing technology.

[0025] As a further solution of the present invention: in said S3.1, by identifying the department data on the medical record text data, the corresponding department data in the model database is retrieved, and a temporary interactive channel is established between the single-chip microcomputer to facilitate the single-chip microcomputer to retrieve the corresponding department data in the model database.

[0026] As a further solution of the present invention: in S4.1, data mining technology is combined through text mining, rule mining and text classification. Key data in the text is extracted through text mining, and the key data is transferred to rule mining and text classification. Rule mining uses the Apriori algorithm to mine the correlation between medical record text data and the model database, and divides the medical record text data and instance status through text classification, and obtains the data structure between the medical record text data and the model database through a decision tree.

[0027] As a further solution of the present invention: after obtaining the data in S4.1, S4.2 transfers the data to the next step when the results of the rule mining and text classification technologies are consistent; when some structures are inconsistent, the results of the rule mining and text classification technologies are color-marked, and the text colors of the rule mining and text classification are contrasting colors, while the colors of the same parts remain unchanged.

[0028] As a further solution of the present invention: after S5 obtains the medical record data through S4, it combines and matches the medical record text data with the medical record-related data. When the medical record-related data is missing, it returns to S3 to re-acquire the medical record data in the model database, and then improves the analysis results of the medical record text data, and then converts the medical record-related data and the medical record text data into a visual list.

[0029] As a further solution of the present invention: in S6, the modification suggestions and treatment plans input by the doctor in the visualization window are transmitted to the model database, and the administrator in the model database decides whether to store the modified suggestions and treatment plans and plans the department to process them.

[0030] As a further solution of the present invention: in said S7, the medical record text data is monitored in real time by the audit system to ensure the quality of the medical record text data in each link and to ensure the subsequent doctor's follow-up of the patient's condition.

[0031] By adopting the above technical solution, compared with the prior art, the beneficial effects of the present invention are:

[0032] 1. The present invention uses medical record mining related technologies and analyzes medical record-related data. When the analysis of medical record-related data is incomplete, it returns to S3 and re-retrieves the data in the model database to improve the medical record-related data and then analyzes the treatment plan again. The doctor views the medical record text data through a visualization window. The medical record-related data assists the doctor in judging the current medical record text data and the feasibility of the treatment plan given in S6. The doctor is then able to modify the medical record, making it easier for the doctor to view problems in the medical record and reducing the time the doctor spends judging the medical record. This ensures the doctor's judgment of the condition and improves the quality of electronic medical records.

[0033] 2. The present invention mines medical record text data through a mining system. After the mining system mines the data, it uses data mining technology to analyze the medical record text data and medical record-related data, and transmits the data to S5. S5 analyzes the medical record-related data. If the data integrity of the medical record-related data analysis is insufficient, the medical record-related data and the treatment plan of the related data in the model database are retrieved through a temporary interactive channel. After the analysis is completed, the cached data in the single-chip computer is cleared, which can protect the patient's privacy and the security of quality control.

[0034] 3. The present invention uses the audit system to monitor each link of the medical record text data in real time, and reminds the abnormal data and information in each link, so as to facilitate medical staff to modify and improve the data in the medical record text data, thereby ensuring the quality of the medical record text data;

[0035] 4. The present invention displays the medical record data associated with the current medical record text data through a visual window, which is convenient for doctors to analyze the causes of diseases appearing in the current medical record text data, thereby reducing the risk of misdiagnosis by doctors. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] Figure 1Schematic diagram of the steps of the electronic medical record quality control method in an embodiment of the present invention;

[0037] Figure 2 Schematic diagram of privacy protection steps in an embodiment of the present invention;

[0038] Figure 3 Schematic diagram of the medical record mining steps in an embodiment of the present invention. DETAILED DESCRIPTION

[0039] The following is a further description of specific embodiments of the present invention in conjunction with the accompanying drawings. It should be noted that the description of these embodiments is intended to facilitate understanding of the present invention and does not constitute a limitation of the present invention. In addition, the technical features involved in the various embodiments of the present invention described below may be combined with each other as long as they do not conflict with each other.

[0040] The electronic medical record quality control method based on artificial intelligence of the present invention comprises the following steps:

[0041] S1. Establishing a model database: Inputting text feature information and vocabulary into the constructed model database;

[0042] S2. Obtaining medical record text data: extracting text information, obtaining text feature information in the text information, and parsing the text feature information and key word data in the text information;

[0043] S3, privacy protection: import the text feature information and key words extracted in S2 into the mining system;

[0044] S3.1. Enable the MCU and retrieve the key word data identified in S2 and match it with the model database in S1;

[0045] S3.2. After the data matching in S3.1 is completed, the data is transmitted to the mining system. After the mining system completes the processing, the cached data in the microcontroller is cleared;

[0046] S4, medical record mining: parsing the medical record data in S3;

[0047] S4.1. Mining medical record data by setting up data mining technology in the mining system to parse the medical record text data related to the medical record;

[0048] S4.2. Match the medical record associated data with the medical record text data, and generate list data of the medical record text data and the medical record associated data;

[0049] S5, treatment suggestion: by judging the medical record text data and the medical record-related data, by calling the treatment plan of the medical record-related data in the model database, and analyzing the medical record-related data, and combining the analysis results with the medical record text data, the treatment plan in the current medical record text data is parsed based on the medical record text data and the medical record-related data, the generated data is transmitted to S6, and an instruction to clear the cached data is sent to the single-chip microcomputer;

[0050] S6, treatment decision: Obtain the data in S5 and S4 and transfer them to the visualization window. The doctor then compares the medical record text data with the relevant data, and then checks the treatment plan associated with the medical record data and the treatment plan generated by the mining system. The doctor uses his or her own qualifications to compare the two treatment plans to ensure that the treatment plan is feasible. At the same time, the doctor can enter the doctor's modification suggestions and modify the treatment plan in the visualization window based on the comparison of the medical record text data and the relevant data, and at the same time transfer the data entered by the doctor to the model database in S1;

[0051] S7. Medical record tracking: By real-time monitoring of medical record text data, the quality of the medical record text can be ensured. At the same time, by recording abnormal data in S2, S3, S4 and S5, medical staff can be reminded to modify and improve the medical record text.

[0052] In one embodiment of the present invention, the specific steps in S1 are as follows:

[0053] S1.1. Establish a model database based on different departments;

[0054] S1.2. Establish a tree diagram based on the medical record data, and connect the model databases of different departments in series based on the tree diagram to ensure that hidden diseases can be deeply analyzed.

[0055] In one embodiment of the present invention, the specific steps in S2 are as follows:

[0056] S2.1. Scanning the text on the medical record using image processing technology to obtain text features on the medical record, and transmitting the text feature information to the next step;

[0057] S2.2. Utilize natural speech processing technology to analyze and understand text features. Simultaneously, natural speech processing technology analyzes whether the text feature data identified by image processing technology is abnormal.

[0058] S2.3. Through the interaction of natural speech processing technology and image processing technology, abnormal text feature data is identified and corrected, and omissions are supplemented to ensure the integrity of text information and the accuracy of medical record recognition.

[0059] S2.4. Ensure the integrity of basic medical information, medical history, and diagnosis through natural speech processing technology.

[0060] In one embodiment of the present invention: In S3.1, by identifying the department data on the medical record text data, the corresponding data of the department in the model database is retrieved, and a temporary interactive channel is established between the single-chip microcomputer to facilitate the single-chip microcomputer to retrieve the data of the corresponding department in the model database.

[0061] In one embodiment of the present invention: In S4.1, data mining technology is combined through text mining, rule mining and text classification. Key data in the text is extracted through text mining, and the key data is transferred to rule mining and text classification. Rule mining uses the Apriori algorithm to mine the correlation between medical record text data and the model database, and divides the medical record text data and instance status through text classification, and obtains the data structure between the medical record text data and the model database through a decision tree.

[0062] In one embodiment of the present invention: After S4.2 obtains the data in S4.1, when the results of the rule mining and text classification technologies are consistent, the data is transferred to the next step; when some structures are inconsistent, the results of the rule mining and text classification technologies are color-marked, and the text colors of the rule mining and text classification are contrasting colors, while the colors of the same parts remain unchanged.

[0063] In one embodiment of the present invention: After S5 obtains the medical record data through S4, it combines and matches the medical record text data with the medical record-related data. When the medical record-related data is missing, it returns to S3 to re-acquire the medical record data in the model database, and then improves the analysis results of the medical record text data, and then converts the medical record-related data and the medical record text data into a visual list.

[0064] In one embodiment of the present invention: In S6, the doctor's modification suggestions and treatment plans are input in the visualization window, so that the doctor's modification suggestions and treatment plans are transmitted to the model database, and the administrator in the model database decides whether to store the modified suggestions and treatment plans and plans the department to process them.

[0065] In one embodiment of the present invention: In S7, the medical record text data is monitored in real time by the audit system to ensure the quality of the medical record text data in each link and to ensure the subsequent doctor's follow-up of the patient's condition.

[0066] Example 1

[0067] A model database is established through different departments, and the relevant medical records are connected in series through the tree diagram in the model database. Then, the current medical record text data and medical record-related data are analyzed through the decision tree and text classification in data mining technology. The decision tree and tree diagram are used to quickly obtain data related to the medical record text data, and obtain the treatment plan of the medical record-related data. At the same time, the correlation between the treatment plan and the medical record text is analyzed. Finally, the doctor judges the feasibility of the current medical record text treatment plan. At the same time, the audit system monitors each link of the medical record text data in real time to ensure the overall quality of the medical record text.

[0068] As attached Figure 1 -Attached Figure 3 As shown, through the medical record mining related technology, and through the medical record related data analysis, at the same time, when the medical record related data analysis is incomplete, it returns to S3 and re-retrieves the data in the model database, improves the medical record related data and then analyzes the treatment plan again. The doctor views the medical record text data through the visualization window, and the medical record related data assists the doctor in judging the current medical record text data, and at the same time judges the feasibility of the treatment plan given in S6 and modifies it, so as to facilitate the doctor to view the problems in the medical record, reduce the time spent by the doctor on judging the medical record, thereby ensuring the doctor's judgment on the condition and improving the quality of electronic medical records;

[0069] The medical record text data is mined through the mining system. After the mining system mines the data, the medical record text data and medical record related data are analyzed using data mining technology and transmitted to S5. S5 analyzes the medical record related data. If the data integrity of the medical record related data analysis is insufficient, the medical record related data and the treatment plan of the related data in the model database are retrieved through a temporary interactive channel. After the analysis is completed, the cached data in the single-chip computer is cleared, which can protect the patient's privacy and the security of quality control.

[0070] Through the audit system, each link of the medical record text data is monitored in real time, and abnormal data and information in each link are reminded, which is convenient for medical staff to modify and improve the data in the medical record text data, ensure the quality of the medical record text data, and display the medical record data related to the current medical record text data through a visual window, which is convenient for doctors to analyze the causes of the diseases in the current medical record text data, thereby reducing the risk of misdiagnosis by doctors.

[0071] The embodiments of the present invention are described in detail above with reference to the accompanying drawings, but the present invention is not limited to the described embodiments. It is apparent to those skilled in the art that various changes, modifications, substitutions, and variations to these embodiments may be made without departing from the principles and spirit of the present invention, and these changes and modifications still fall within the scope of protection of the present invention.

[0072] Throughout the specification, references to terms such as "one embodiment," "example," or "specific example" indicate that the specific features, structures, materials, or characteristics described in conjunction with that embodiment or example are included in at least one embodiment or example of the present invention. In this specification, schematic representations of these terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in any one or more embodiments or examples.

[0073] The above contents are merely examples and explanations of the present invention. Any modifications, additions or replacements in similar ways made by those skilled in the art to the specific embodiments described shall fall within the scope of protection of the present invention as long as they do not deviate from the invention or exceed the scope defined by the claims.

[0074] The above description is only a preferred specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any technician familiar with the technical field, within the technical scope disclosed by the present invention, who makes equivalent replacements or changes based on the technical solution and inventive concept of the present invention, should be covered by the scope of protection of the present invention.

Claims

1. An artificial intelligence-based electronic medical record quality control method, characterized in that: The following steps are involved: S1. Establishing a model database: Inputting text feature information and vocabulary into the constructed model database; S2. Obtaining medical record text data: extracting text information, obtaining text feature information in the text information, and parsing the text feature information and key word data in the text information; S3, privacy protection: import the text feature information and key words extracted in S2 into the mining system; S3.

1. Enable the MCU and retrieve the key word data identified in S2 and match it with the model database in S1; S3.

2. After the data matching in S3.1 is completed, the data is transmitted to the mining system. After the mining system completes the processing, the cached data in the microcontroller is cleared; S4, medical record mining: parsing the medical record data in S3; S4.

1. Mining medical record data by setting up data mining technology in the mining system to parse the medical record text data related to the medical record; S4.

2. Match the medical record associated data with the medical record text data, and generate list data of the medical record text data and the medical record associated data; S5, treatment suggestion: by judging the medical record text data and the medical record-related data, by calling the treatment plan of the medical record-related data in the model database, and analyzing the medical record-related data, and combining the analysis results with the medical record text data, the treatment plan in the current medical record text data is parsed based on the medical record text data and the medical record-related data, the generated data is transmitted to S6, and an instruction to clear the cached data is sent to the single-chip microcomputer; S6, treatment decision: Obtain the data in S5 and S4 and transfer them to the visualization window. The doctor then compares the medical record text data with the relevant data, and then checks the treatment plan associated with the medical record data and the treatment plan generated by the mining system. The doctor uses his or her own qualifications to compare the two treatment plans to ensure that the treatment plan is feasible. At the same time, the doctor can enter the doctor's modification suggestions and modify the treatment plan in the visualization window based on the comparison of the medical record text data and the relevant data, and at the same time transfer the data entered by the doctor to the model database in S1; S7, Medical Record Tracking: By monitoring medical record text data in real time, the quality of medical record text can be ensured. At the same time, by recording abnormal data in S2, S3, S4 and S5, medical staff can be reminded to modify and improve the medical record text; The specific steps in S1 are as follows: S1.

1. Establish a model database based on different departments; S1.

2. Build a tree diagram based on medical record data and connect different department model databases in series based on the tree diagram to ensure in-depth analysis of hidden conditions. In S3.1, by identifying the department data in the medical record text data, the corresponding department data in the model database is retrieved, and a temporary interactive channel is established between the microcontroller and the microcontroller to facilitate the microcontroller to retrieve the corresponding department data in the model database; In S4.1, data mining technology combines text mining, rule mining, and text classification. Text mining extracts key data from the text and transfers the key data to rule mining and text classification. Rule mining uses the Apriori algorithm to mine the association between medical record text data and the model database. Text classification divides the medical record text data and instance status, and obtains the data structure between the medical record text data and the model database through a decision tree. S4.2 After obtaining the data in S4.1, if the results of the rule mining and text classification techniques are consistent, the data is transferred to the next step. If some structures are inconsistent, the rule mining and text classification technology results are color-coded, and the text colors of the rule mining and text classification are contrasting, while the colors of the identical parts remain unchanged. After S5 obtains the medical record data from S4, it combines and matches the medical record text data with the medical record-related data. When the medical record-related data is missing, it returns to S3 to re-acquire the medical record data in the model database, and then improves the analysis results of the medical record text data, and then converts the medical record-related data and the medical record text data into a visual list.

2. The artificial intelligence-based electronic medical record quality control method according to claim 1, characterized in that: In S6, the doctor's modification suggestions and treatment plans entered in the visualization window are transmitted to the model database, and the administrator in the model database decides whether to store the modified suggestions and treatment plans and plans the department to process them.

3. The artificial intelligence-based electronic medical record quality control method according to claim 1, characterized in that: In S7, the medical record text data is monitored in real time through the audit system to ensure the quality of the medical record text data in each link and to ensure subsequent doctors' follow-up on the patient's condition.

4. The artificial intelligence-based electronic medical record quality control method according to claim 1, characterized in that: The specific steps in S2 are as follows: S2.

1. Scanning the text on the medical record using image processing technology to obtain text features on the medical record, and transmitting the text feature information to the next step; S2.

2. Utilize natural speech processing technology to analyze and understand text features. Simultaneously, natural speech processing technology analyzes whether the text feature data identified by image processing technology is abnormal. S2.

3. Through the interaction of natural speech processing technology and image processing technology, abnormal text feature data is identified and corrected, and omissions are supplemented to ensure the integrity of text information and the accuracy of medical record recognition. S2.

4. Ensure the integrity of basic medical information, medical history, and diagnosis through natural speech processing technology.

Images