Electronic nursing record automatic input method and device adopting generative artificial intelligence

Abstract

The method for automatically inputting an electronic care record according to the present invention may comprise: a step of selecting a user interface section by an electronic care record system; inputting a text into a text input box of the user interface unit in accordance with items of a care record; performing cleaning and word segmentation on the input text; a step of performing fine tuning on the text after word segmentation; and inputting the fine-tuned text into the electronic nursing recording system. Wherein the step of fine tuning can screen at least more than one piece of nursing record data corresponding to the input text from a predefined nursing record database, so that the input text can be automatically completed or corrected.

Description

Technical Field

[0001] This invention relates to an automatic input method and apparatus for electronic nursing records using generative artificial intelligence, and more specifically, to a medical record (nursing record) system using generative artificial intelligence that can be used in classrooms and training sessions for aspiring medical professionals. Background Technology

[0002] Electronic medical records (EMR) refer to the results of computerized processing of medical records used in hospital clinical support services. Relatedly, when hospital staff input patient information into the hospital's internal electronic medical record system, they can view various records related to the patient's illness.

[0003] While most hospitals and clinics have now introduced and are using electronic medical record (EMR) systems, classroom teaching at health and medical universities has not kept pace with the changes in hospital healthcare systems.

[0004] Therefore, in many classroom-based medical record training sessions, computer systems are often unavailable, and training is conducted using the traditional handwritten record method. While there are cases of creating teaching versions of electronic medical records (EMR) systems, especially electronic nursing records (ENR), used in hospitals and clinics, to address this issue, these only replace handwritten records with typing; inputting nursing records still requires a significant amount of time.

[0005] Therefore, there is a growing demand for technologies that enable the automatic input of nursing records, such as databases storing nursing records of virtual patients generated based on teaching, generative AI, and artificial intelligence modeling.

[0006] In other words, previous technologies related to electronic nursing record systems required a considerable amount of time to complete nursing records. Consequently, prospective medical personnel faced difficulties in completing these records when working in actual clinical hospitals, leading to a prolonged adaptation period. Furthermore, the use of clinically-based electronic nursing record systems for teaching purposes necessitated the direct use of actual patients' personal information. Therefore, significant technological advancements have not yet been achieved in electronic nursing record systems for prospective nurses' educational purposes.

[0007] Therefore, a specific method is needed to solve the above problems.

[0008] Existing technical documents Patent documents (Patent Document 0001) Korean Patent Publication No. 10-2022-0164439 (Publication Date: December 13, 2022) Summary of the Invention Technical issues The purpose of the automatic input method and apparatus for electronic nursing records using generative artificial intelligence according to the present invention is to enable prospective medical personnel to undergo pre-training in electronic nursing records and electronic medical records before entering actual clinical hospitals, especially advanced general hospitals, so as to be able to quickly adapt to different hospital environments.

[0009] In addition, the purpose of this invention is to shorten the time required for filling out electronic nursing records and to automatically correct errors and typos that occur during record filling based on context, thereby improving the accuracy and speed of nursing record filling.

[0010] Problem-solving methods To address the problems described above, the method for automatically inputting electronic nursing records according to the present invention may include: selecting a user interface section through an electronic nursing record system; inputting text in a text input box of the user interface section according to pre-defined nursing record items; cleaning and segmenting the input text; fine-tuning the segmented text; and inputting the fine-tuned text into the electronic nursing record system; wherein the fine-tuning step may involve filtering at least one nursing record data corresponding to the input text from a predefined nursing record database for automatic completion or correction of the input text.

[0011] The steps of cleaning and segmenting the input text, and fine-tuning the segmented text, can be implemented using a generative artificial intelligence module.

[0012] The information entered in the text input box may include at least one of the following: medical records, disease codes, vital signs, fall risk level, blood test results, fluid imbalance test results, and prescription details.

[0013] The nursing record may include a nursing system consisting of at least one of NANDA, SOAPIE, Focus DAR, ICNP, narrative record, and nursing process. The electronic nursing record automatic input method may also include a step of inputting the nursing record while changing it through the interface change unit in the user interface unit.

[0014] The nursing record data used for the automatic completion or correction of the input text from at least one nursing record data selected from the predefined nursing record database can be determined based on pre-learned user preferences or pre-determined priority benchmarks.

[0015] The electronic nursing record automatic input device according to the present invention may include: a user interface unit, the user interface unit including a text input box matching a pre-specified nursing record item; a nursing record database, the nursing record database storing nursing record data about a virtual patient and nursing record data about a patient; a control unit, the control unit automatically completing or correcting the text entered in the text input box according to the context; and a communication unit, the communication unit being able to send and receive data about the text with the nursing record database so that automatic completion or correction can be realized in the control unit; wherein, the automatic completion and correction can be realized by a generative artificial intelligence module within the control unit.

[0016] This may include an interface modification unit, which can modify the input of nursing record data according to the input environment; wherein, the nursing record data may include a nursing system consisting of at least one of NANDA, SOAPIE, Focus DAR, ICNP, narrative records, and nursing processes.

[0017] Invention Effects The automatic input method and apparatus for electronic nursing records using generative artificial intelligence according to the present invention can improve the accuracy and speed of filling in nursing records.

[0018] The automatic input method and apparatus for electronic nursing records using generative artificial intelligence according to the present invention can be applied to nurse work support and nursing education, thereby improving the clinical adaptability of nursing students.

[0019] The automatic input method and apparatus for electronic nursing records using generative artificial intelligence according to the present invention can be applied to electronic health record systems such as electronic nursing record (ENR) or electronic medical record (EMR), thereby enabling the upgrading of electronic nursing records and electronic medical records.

[0020] The effects of the present invention are not limited to the contents disclosed above, and more diverse effects are included in this specification. Attached Figure Description

[0021] The embodiments in this specification can be better understood by referring to the following description connected with the accompanying drawings, wherein similar reference numerals denote the same or functionally similar elements.

[0022] Figure 1 This is a sequence diagram of an automatic input method for electronic nursing records according to an embodiment of the present invention.

[0023] Figure 2This is a sequence diagram of an automatic input method for electronic nursing records using generative artificial intelligence according to an embodiment of the present invention.

[0024] Figure 3 This is a sequence diagram illustrating the content to be filled in a text input box in an automatic input method for electronic nursing records using generative artificial intelligence according to an embodiment of the present invention.

[0025] Figure 4 This is a sequence diagram illustrating the nursing record content and input method in an automatic electronic nursing record input method employing generative artificial intelligence according to an embodiment of the present invention.

[0026] Figure 5 To show Figure 4 Example diagram.

[0027] Figure 6 This is a conceptual diagram illustrating an automatic input device for electronic nursing records employing generative artificial intelligence according to an embodiment of the present invention.

[0028] Figure 7 A conceptual diagram illustrating the implementation principle of an automatic electronic nursing record input device employing generative artificial intelligence according to an embodiment of the present invention.

[0029] The figures referenced above are not necessarily drawn to scale and should be understood as simplified representations of various preferred features illustrating the basic principles of this disclosure. For example, specific design features of this disclosure, including specific dimensions, orientations, positions, and shapes, will be partly determined by the specific intended application and environment of use.

[0030] (Explanation of reference numerals in the attached image) 1: Electronic nursing record automatic input device; 10: User interface unit 100: Interface Change Department; 110: Text Input Box 20: Control Unit; 21: Generative Artificial Intelligence Module 200: Data Cleaning Department; 210: Fine-tuning Department 220: Correlation Measurement Department; 30: Communication Department 40: Electronic Nursing Record System (ENR) 50: Encryption Department 60: Nursing record database Detailed Implementation The following describes some embodiments of this disclosure in detail with reference to the exemplary accompanying drawings. When affixing reference numerals to the constituent elements of the drawings, identical constituent elements are identified by the same numerals whenever possible, even when shown in different drawings. Furthermore, in describing this embodiment, detailed descriptions of related well-known structures or functions may be omitted if it is determined that such detailed descriptions might obscure the essence of the technical concept. The terms "comprising," "having," and "constituting," as used in this specification, may be supplemented with other parts unless "only" is used. When a constituent element is expressed in the singular, it may include a plural unless otherwise expressly stated.

[0031] In addition, in describing the constituent elements of this disclosure, terms such as 1, 2, A, B, (a), and (b) may be used. Such terms are used only to distinguish the constituent element from other constituent elements, and the nature, order, sequence, or number of the corresponding constituent elements are not limited by the terms.

[0032] In describing the positional relationship between constituent elements, when two or more constituent elements are described as "connected," "combined," or "continuous," it should be understood that the two or more constituent elements can be directly "connected," "combined," or "continuous," or that the two or more constituent elements can be "connected," "combined," or "continuous" through other constituent elements. Furthermore, other constituent elements can also be included in one or more of the two or more constituent elements that are mutually "connected," "combined," or "continuous."

[0033] In describing temporal relationships related to constituent elements, operating methods, or production methods, for example, when using phrases such as "after," "immediately following," "after," or "before" to describe temporal or procedural sequences, discontinuous cases may also be included unless "immediately" or "directly" is used.

[0034] On the other hand, when referring to the numerical values ​​of constituent elements or their corresponding information (e.g., levels), even if not explicitly stated separately, the numerical values ​​or their corresponding information can be interpreted as including the range of errors that may arise due to various factors (e.g., process factors, internal or external impacts, noise, etc.).

[0035] The core technical feature of this invention lies in the use of classification models such as Top-k recommendation or Top-k accuracy during fine-tuning. The frequency of the input text with sentences in the nursing database is used as weights for scoring, which is then applied to sentence generation and recommendation. The invention will be described in detail below with reference to the accompanying drawings.

[0036] Figure 1This is a sequence diagram of an automatic electronic nursing record input method according to an embodiment of the present invention. Figure 2 This is a sequence diagram of an automatic input method for electronic nursing records using generative artificial intelligence according to an embodiment of the present invention. Figure 3 This is a sequence diagram illustrating the content to be entered into a text input box in an automatic electronic nursing record input method employing generative artificial intelligence according to an embodiment of the present invention. Figure 4 This is a sequence diagram illustrating the nursing record content and input method in an automatic electronic nursing record input method employing generative artificial intelligence according to an embodiment of the present invention. Figure 5 To show Figure 4 Example diagram.

[0037] refer to Figure 1 According to an embodiment of the present invention, the automatic input method for electronic nursing records implemented by the electronic nursing record automatic input device 1 may include step S100 of selecting the user interface unit 10 through the electronic nursing record system (ENR) 40. As an example, the user interface unit 10 may be configured with an input system based on, but is not limited to, NANDA or ICNP.

[0038] As an example, NANDA stands for NANDA International, the international standard used for classifying nursing diagnoses. NANDA International focuses on defining and standardizing nursing diagnoses in the field of nursing. A nursing diagnosis refers to one of the nurse's roles or nursing processes in assessing and understanding a patient's condition, used to describe the patient's health status and develop preventative or situation-based care plans.

[0039] On the other hand, ICNP stands for International Classification for Nursing Practice. ICNP is a standardized international classification system for nursing practice used internationally by the World Nursing Organization. In this context, ICNP provides nurses, nursing scholars, healthcare information specialists, and other health-related professionals with standardized language and structures for classifying and understanding nursing issues.

[0040] Then, step S200 can begin, where text is entered into the text input boxes of the user interface unit 10 according to the items in the nursing record. As an example, the content entered into the text input boxes may include, but is not limited to, at least one of the following: medical records, disease codes, vital signs, fall risk, blood test results, fluid imbalance test results, and prescription details. Other content may also be entered depending on the settings.

[0041] Then, the steps of cleaning and tokenizing the input text (S300) and fine-tuning the tokenized text (S400) can be started.

[0042] The steps of segmenting and / or fine-tuning can be viewed as performing a cleaning process on the input text, which is to remove noise from the text and standardize the text in a consistent format.

[0043] As an example of performing the above cleaning process, a process of converting English tags to uppercase and / or lowercase can be performed. Furthermore, frequently occurring but almost meaningless words can be filtered and removed from the text to highlight the truly important information. Alternatively, cleaning can be performed by removing special characters from the text, or by using regular expressions, or by performing stemming and lemmatization, removing duplicate characters, frequency-based filtering, etc.

[0044] On the other hand, tokenizing refers to the process of breaking down given text into smaller units, which can typically be words, sentences, or paragraphs. That is, it can be viewed as breaking down a sentence into its smallest meaningful words. Therefore, the input text can be split into pre-defined units such as words, sentences, or paragraphs through the tokenizing process. In this process, besides simply splitting the input text into pre-defined units such as words, sentences, or paragraphs, it is also possible to remove all or part of recurring meaningless words from the input text, or to convert the input text or its components into a consistent format.

[0045] Tokenization is a crucial preprocessing step in Natural Language Processing (NLP) and machine learning. It's the initial stage of NLP, transforming text data into a form that can be input into a model. This segmented data is typically used as model input or for text-based tasks. Technically, it's performed using regular expressions, string manipulation functions, or specialized libraries.

[0046] The fine-tuning refers to the automatic completion or correction of the text using a pre-trained neural network model or similar method to improve its accuracy and clarity. During this process, data corresponding to the text can be sent and received from a predefined nursing record database, and this data can be used to automatically complete or correct the text.

[0047] On the other hand, the steps of cleaning and segmenting the input text (S300) and fine-tuning the segmented text (S400) can be implemented using the generative artificial intelligence module 21 included in the control unit 20.

[0048] Generative Artificial Intelligence (CAI) module 21 is a type of AI primarily focused on data analysis and generating new data. It learns meaningful patterns or information from given input data and generates or completes new content based on these patterns. Therefore, by segmenting and / or fine-tuning the input text using CAI, cleaned new text can be derived for input into an electronic nursing record system. Examples of CAI modules include, but are not limited to, chatGPT, Bard, and HyperCLOVA X.

[0049] On the other hand, to apply the generative artificial intelligence module 21 as described above and measure its performance, the Top K-Accuracy classification model and the Top K-Recommendation recommendation (preference diagnosis name measurement) model can be used. Top K-accuracy is one of the metrics used to evaluate model performance in multi-class classification problems, measuring the proportion of the actual class included in the top K model predictions.

[0050] In Top K-accuracy, K represents how many of the preceding predictions are considered. As an example, Top 3-accuracy measures the number of actual classes included in the first three model predictions. This is particularly useful when the order of the preceding predictions is not important.

[0051] Most deep learning frameworks provide functions or libraries that can compute Top K-accuracy, and you can usually specify the K value to be considered when evaluating the model.

[0052] On the other hand, Top-K Recommendation refers to a system that recommends the top K items to a user, while Preference Diagnosis is a process of providing recommendation models (e.g., text) based on user preferences and behaviors, and it represents the process of improving or describing them.

[0053] During the fine-tuning process, classification models such as Top-K Recommendation or Top-K Accuracy can also be used. For example, when fine-tuning the input text, at least one piece of data (or nursing record data) corresponding to the input text can be filtered from a predefined nursing record database. This data can then be used as a benchmark for automatic completion or correction of the input text. In this case, a Top-K Recommendation classification model can be used to filter more than one piece of data (or nursing record data) from the predefined nursing record database to be used as the benchmark.

[0054] In addition, for each of the large amounts of data (or nursing record data) recorded in the predefined nursing record database, a weight can be assigned to each data (or nursing record data) based on its usage frequency. The data (or nursing record data) with the highest assigned weight or the K data (or nursing record data) arranged in descending order of weight are then selected as the data (or nursing record data) corresponding to the input text. Thus, the most frequently used data (or nursing record data) in the nursing record database can be used to automatically complete or correct the input text.

[0055] Additionally, the data (or nursing record data) used for the automatic completion or correction of the input text from at least one data (or nursing record data) filtered from the predefined nursing record database can be determined based on pre-learned user preferences or pre-determined priority benchmarks.

[0056] On the other hand, the finely tuned text that has undergone automatic completion or correction processing can be recommended to the user as diagnostic data input into the electronic nursing record system.

[0057] On the other hand, if users continuously receive diagnostic data recommendations through this system, various factors such as user behavior records, evaluations, and filling history can be considered during this process to collect diagnostic data on user preferences. At this time, fine-tuning of the input text can also be achieved using the collected preference diagnostic data.

[0058] In addition, to enhance the interpretability of preference diagnostic data, the function of describing the reasons for recommending the corresponding diagnostic data can be added, thereby increasing user trust and providing transparency to the recommendation system. Furthermore, by receiving real-time feedback, the preference diagnostic data can be updated as new feedback is collected.

[0059] In addition, personalized preference diagnosis can help understand each user's preferences and add the ability to make customized recommendations or descriptions for each user, thereby enabling personalized recommendation results.

[0060] As an example, machine learning (ML) and deep learning can be applied to systems and software in an embedded artificial intelligence manner to realize the process of building an LLM (Large Language Model).

[0061] Embedded AI refers to the technology of implementing artificial intelligence models in small, lightweight devices, using methods optimized for running artificial intelligence algorithms in mobile devices, sensors, embedded systems, IoT (Internet of Things) devices, etc.

[0062] Since most embedded systems have limited resources, memory and computing resources can be used efficiently through model lightweighting. Techniques such as model compression, quantization, and weight pruning can be used to reduce the model size, and optimization techniques that utilize hardware accelerators (e.g., GPUs, TPUs) can be used to run the model efficiently.

[0063] Embedded AI typically runs in edge computing environments, which allows data to be processed and analyzed in real time at the point of generation (i.e., the edge). Additionally, hardware acceleration optimized for specific embedded platforms can be employed.

[0064] The lightweight model achieved through this process can be effectively applied to embedded environments, thereby providing effective artificial intelligence capabilities even in real-time or resource-constrained environments.

[0065] As an example, word segmentation can be performed using NLP. NLP is an abbreviation for "Natural Language Processing." NLP can be used for text classification, which involves categorizing text documents into multiple categories or classes; filtering; sentiment analysis; topic classification; and machine translation.

[0066] In addition, in order to perform information retrieval and text classification in the nursing record database 60 and the electronic nursing record system 40, Named Entity Recognition (NER) can be implemented, and appropriate text generation confirmed by the relevance measurement unit 220 can be achieved through automatic summarization and cleaning processes. The automatic summarization extracts only the important content of long documents or texts and provides a concise summary.

[0067] Then, in order to enable the text to be automatically completed or corrected according to the context, step S500 of sending and receiving data related to the text between the nursing record database and the electronic nursing record automatic input device, and step S600 of transmitting the nursing record filled in the nursing record details in the electronic nursing record system to the electronic nursing record automatic input device can be started.

[0068] As an example, the nursing record may include a nursing system consisting of at least one of NANDA, SOAPIE, Focus DAR, ICNP, narrative record, and nursing process.

[0069] Therefore, step S700 can be further initiated, allowing for input of changes to the nursing record simultaneously via the interface modification unit 100 within the user interface unit 10. For example, the nursing record can be modified according to the environment of a hospital or university.

[0070] Figure 6 This is a conceptual diagram illustrating an automatic input device for electronic nursing records employing generative artificial intelligence according to an embodiment of the present invention. Figure 7 A conceptual diagram illustrating the implementation principle of an automatic electronic nursing record input device employing generative artificial intelligence according to an embodiment of the present invention.

[0071] An electronic nursing record automatic input device 1 corresponding to an embodiment of the present invention may include: a user interface unit 10, which includes a text input box matching a pre-specified nursing record item; a nursing record database 60, which stores nursing record data about a virtual patient and nursing record data about a patient; a control unit 20, which automatically completes or corrects the text entered in the text input box according to the context; and a communication unit 30, which can send and receive data about the text with the nursing record database 60 so that automatic completion or correction can be achieved in the control unit 20. Furthermore, the automatic completion and correction can be achieved by a generative artificial intelligence module 21 within the control unit 20.

[0072] As an example, the nursing record data in the nursing record database 60 may include a nursing system consisting of at least one of NANDA, SOAPIE, Focus DAR, ICNP, narrative records, and nursing processes. The electronic nursing record automatic input device 1 may include an interface modification unit 100 in the user interface unit 10, which is capable of modifying and inputting the nursing record data according to the input environment.

[0073] Additionally, refer to Figures 6 to 7It is understood that the aforementioned method for automatically inputting electronic nursing records using generative artificial intelligence according to an embodiment of the present invention can be implemented by an application stored in the storage medium of a device or computer. The computer may include an automatic electronic nursing record input system using generative artificial intelligence. The computer's operating system may be a Windows or Macintosh operating system installed on a regular PC such as a desktop or laptop, or a mobile-specific operating system such as iOS or Android installed on a mobile terminal such as a smartphone or tablet.

[0074] The aforementioned method for automatically inputting electronic nursing records using generative artificial intelligence according to embodiments of the present invention can be implemented by an application program (i.e., a computer program) that is installed by default on a computer or installed by the user, and can be stored (recorded) in a computer-readable storage medium. That is, it can be implemented through hardware components, software components, and / or a combination of hardware components and software components.

[0075] For example, the apparatus and components described in the embodiments can be implemented using more than one general-purpose or special-purpose computer, such as a processor, controller, ALU (arithmetic logic unit), digital signal processor, microcomputer, FPGA (field programmable gate array), PLU (programmable logic unit), microprocessor, or any other device that can execute instructions and respond to them.

[0076] For ease of understanding, although sometimes described as using a single processing device, those skilled in the art will understand that the processing device may include multiple processing elements and / or various types of processing elements.

[0077] For example, the processing device may include multiple processors, or one processor and one controller. Alternatively, other processing configurations may be used, such as a parallel processor.

[0078] An application program for implementing the automatic input method for electronic nursing records using generative artificial intelligence according to an embodiment of the present invention, and stored and running in a computer storage medium, can perform the following steps: selecting the user interface unit 10 in the intelligent ENR program; receiving text input in the text input box 110 that matches the nursing record item in the user interface; cleaning and segmenting the words or sentences in the text input box 110; fine-tuning the segmented words using generative artificial intelligence; exchanging data with the database to automatically complete or correct the data; and transmitting the nursing record filled in the nursing record details in the intelligent ENR program.

[0079] Thus, in order for the computer to read the program recorded in the storage medium to execute the electronic nursing record automatic input method using generative artificial intelligence according to this embodiment, the application program may contain code written in computer languages ​​such as C, C++, JAVA, and machine language, which is readable by the computer's processor (CPU).

[0080] This code may include function code related to the functions that define the aforementioned functions, or it may include execution process control code required for the computer's processor to execute the aforementioned functions according to a predetermined program.

[0081] In addition, this code may also contain memory reference related code, which instructs the computer processor on where (address number) the additional information or media required to perform the aforementioned functions should be referenced in the computer's internal or external memory.

[0082] On the other hand, when the computer's processor needs to communicate with any other computer or server remotely in order to perform the aforementioned functions, the code may further include communication-related code to instruct the computer's processor how to use the computer's communication module (e.g., wired and / or wireless communication module) to communicate with any other computer or server remotely, and what information or media needs to be sent and received during communication.

[0083] Furthermore, the functional program and related code and code segments used to implement this embodiment can also be derived or modified by programmers skilled in the art, considering the system environment of a computer that reads storage media and executes the program.

[0084] Furthermore, the computer-readable storage medium containing the program as described above can be distributed across computer systems connected via a network, storing and executing the computer-readable code in a distributed manner.

[0085] At this time, any one or more of the distributed computers can execute a portion of the functions described above and transmit the execution result to one or more of the other distributed computers. The computer that receives the result can also execute a portion of the functions described above and provide its result to the other distributed computers in the same way.

[0086] As described above, a computer-readable storage medium that records an application for executing an automatic input method for electronic nursing records employing generative artificial intelligence according to an embodiment of the present invention may, as one embodiment, include a read-only memory (ROM), a random access memory (RAM), an optical disc read-only memory (CD-ROM), a magnetic tape, a floppy disk, an optical media storage device, etc.

[0087] Additionally, the program for executing the automatic input method of electronic nursing records using generative artificial intelligence according to embodiments of the present invention is recorded in a computer-readable storage medium containing the application. This medium can be a storage medium (e.g., a hard disk) included in an application provider server, which may include an application store server, a web server associated with the application or corresponding service, or the application provider server itself, or other computers or their storage media containing the program.

[0088] In summary, a readable record is a computer that serves as the storage medium for programs, i.e., applications, that execute electronic nursing record automatic input methods employing generative artificial intelligence. This includes not only general PCs such as desktops or laptops, but also mobile terminal devices such as smartphones, tablets, PDAs (Personal Digital Assistants), and mobile communication terminals. Furthermore, it should be interpreted as including all devices capable of computing.

[0089] The above description is merely an illustrative description of the technical concept of the present invention. Anyone skilled in the art can make various modifications and variations without departing from the essential characteristics of the present invention.

[0090] Therefore, the embodiments disclosed in this invention are not intended to limit but to describe the technical concept of the invention, and the scope of the technical concept of the invention is not limited by such embodiments.

[0091] The scope of protection of this invention should be interpreted according to the following claims, and all technical ideas within the equivalent scope should be interpreted as being included in the scope of the invention.

[0092] Industrial applicability The automatic input method and apparatus for electronic nursing records using generative artificial intelligence of the present invention can be used in various automatic input methods and apparatus for electronic nursing records.

Claims

1. A method for automatically inputting electronic nursing records, the method comprising: Select the steps in the user interface section using the electronic nursing record system; In the text input box of the user interface section, enter text according to the items of the pre-specified nursing record; The steps of cleaning and segmenting the input text; The steps of fine-tuning the segmented text; and The step of inputting the finely adjusted text into the electronic nursing record system; The fine-tuning step involves filtering at least one nursing record from a predefined nursing record database that corresponds to the input text, for use in the automatic completion or correction of the input text.

2. The method for automatically inputting electronic nursing records according to claim 1, characterized in that, The steps of cleaning and segmenting the input text, and fine-tuning the segmented text, are implemented using a generative artificial intelligence module.

3. The method for automatically inputting electronic nursing records according to claim 2, characterized in that, The information to be entered in the text input box includes at least one of the following: medical records, disease code, vital signs, fall risk level, blood test results, fluid imbalance test results, and prescription details.

4. The method for automatically inputting electronic nursing records according to claim 2, characterized in that, The nursing records include a nursing system consisting of at least one of NANDA, SOAPIE, Focus DAR, ICNP, narrative records, and nursing processes. The electronic nursing record automatic input method further includes a step of inputting the nursing record while modifying it via an interface modification unit within the user interface unit.

5. The method for automatically inputting electronic nursing records according to claim 1, wherein, The nursing record data used for the automatic completion or correction of the text of the input, selected from at least one nursing record data from the predefined nursing record database, is determined based on pre-learned user preferences or a pre-determined priority benchmark.

6. An automatic input device for electronic nursing records, characterized in that, include: The user interface section includes text input boxes that match pre-specified nursing record items; A nursing record database, which stores nursing record data about virtual patients and nursing record data about patients; A control unit that automatically completes or corrects the text entered in the text input box based on the context; and The communication unit is capable of receiving or sending data corresponding to the input text from or to the nursing record database, so that automatic completion or correction can be achieved in the control unit. The automatic completion and correction are achieved through the generative artificial intelligence module within the control unit.

7. The electronic nursing record automatic input device according to claim 6, characterized in that, include: The interface modification unit is capable of modifying the nursing record data input according to the input environment. The nursing record data includes a nursing system consisting of at least one of NANDA, SOAPIE, Focus DAR, ICNP, narrative records, and nursing processes.

Images