Handover content generation method and device for nursing work and computer equipment

Abstract

The invention discloses a handover content generation method and device for nursing work and computer equipment. The method comprises the steps that initial medical information of a target object is acquired, and the initial medical information is generated by splicing multiple pieces of statement information; target medical information is determined in the initial medical information, and the target medical information is information in the initial medical information, and the relevancy between the target medical information and target handover content used for nursing work exceeds a preset threshold value; and based on the target medical information, predicting and generating target handover content corresponding to the target object. The technical problem that the processing efficiency and the data accuracy are affected due to the fact that manual content filling is still needed in an existing SBAR handover method is solved.

Description

Technical Field

[0001] The present invention relates to the field of medical care content generation, and in particular to a method, device and computer equipment for generating handover content for nursing work. Background Art

[0002] In the modern healthcare system, nurse handover is a crucial process. Traditional handover methods, primarily paper records and electronic spreadsheets, suffer from inefficient information integration and difficulty ensuring content integrity. Especially when dealing with complex cases, nurses often spend considerable time sifting through scattered nursing records, temperature charts, and other medical documents to compile valuable information into a clear and complete handover report. This process is both time-consuming and prone to human error.

[0003] As a structured communication method, the Situation-Background-Assessment-Recommendation (SBAR) communication tool has been widely adopted in medical institutions, effectively improving the handover process. While the SBAR system integrates statistical information about patients within a ward on a selected date, nurses still need to manually edit key content, including the patient's chief complaint, current medical history, past medical history, auxiliary examination information, nursing records, medical overview, and vital sign data. Although SBAR simplifies some processes, manual editing remains challenging when dealing with large amounts of information and diverse formats, making it difficult to fully utilize SBAR's effectiveness. In particular, bottlenecks exist in information extraction, language conversion, and content integration.

[0004] To address the above-mentioned problems, no effective solutions have been proposed so far. Summary of the Invention

[0005] Embodiments of the present invention provide a method, apparatus, and computer device for generating handover content for nursing work, to at least address the technical problem that the current SBAR handover method still requires manual content entry, which affects processing efficiency and data accuracy.

[0006] According to one aspect of an embodiment of the present invention, a method for generating handover content for nursing work is provided, comprising: obtaining initial medical information of a target object, wherein the initial medical information is generated by splicing multiple sentence information; determining target medical information in the initial medical information, wherein the target medical information is information in the initial medical information whose relevance to target handover content for nursing work exceeds a preset threshold; and predicting and generating target handover content corresponding to the target object based on the target medical information.

[0007] Optionally, the initial medical information includes the target object's medical history information or the target object's nursing record content, wherein the medical history information includes the chief complaint, current medical history, past medical history and auxiliary examination information, and the nursing record content includes vital signs data.

[0008] Optionally, when the initial medical information is medical history information, obtaining the initial medical information of the target object includes: obtaining multiple sentence information corresponding to the medical history information; sorting and splicing the multiple sentence information in the order of the sentence corresponding to the chief complaint, the sentence corresponding to the current medical history, the sentence corresponding to the past history, and the sentence corresponding to the auxiliary examination information to obtain the initial medical information.

[0009] Optionally, when the initial medical information is the content of a nursing record, obtaining the initial medical information of the target object includes: obtaining multiple sentence information corresponding to the content of the nursing record, wherein the multiple sentence information each corresponds to a recording time; sorting and splicing the multiple sentence information in order from small to large according to the time interval between the recording time and the current time to obtain the initial medical information.

[0010] Optionally, in the initial medical information, target medical information is determined, including: inputting the initial medical information into a first preset model to obtain an index of the target medical information, wherein the architecture of the first preset model is a BERT bidirectional encoder architecture; based on the index, the target medical information is located in the initial medical information.

[0011] Optionally, based on the target medical information, the target handover content corresponding to the target object is predicted and generated, including: inputting the target medical information into a second preset model to obtain the target handover content, wherein the architecture of the second preset model is the mT5 multilingual pre-trained text-to-text converter architecture.

[0012] According to another aspect of an embodiment of the present invention, a handover content generation device for nursing work is also provided, including: an acquisition module for acquiring initial medical information of a target object, wherein the initial medical information is generated by splicing multiple sentence information; a determination module for determining target medical information in the initial medical information, wherein the target medical information is information in the initial medical information whose correlation with the target handover content for nursing work exceeds a preset threshold; a prediction module for predicting and generating target handover content corresponding to the target object based on the target medical information.

[0013] According to another aspect of an embodiment of the present invention, a non-volatile storage medium is also provided, which includes a stored program, wherein when the program is running, the device where the non-volatile storage medium is located is controlled to execute any one of the above-mentioned methods for generating handover content for nursing work.

[0014] According to another aspect of an embodiment of the present invention, a computer device is provided. The computer device includes a processor, and the processor is used to run a program. When the program is run, any one of the above-mentioned methods for generating handover content for nursing work is executed.

[0015] According to another aspect of an embodiment of the present invention, a computer program product is provided, including a computer program, which implements any one of the above-mentioned methods for generating handover content for nursing work when executed by a processor.

[0016] In an embodiment of the present invention, a handover content generation method for nursing work is adopted, by obtaining initial medical information of a target object, wherein the initial medical information is generated by splicing multiple sentence information; in the initial medical information, target medical information is determined, wherein the target medical information is information in the initial medical information whose relevance to the target handover content for nursing work exceeds a preset threshold; based on the target medical information, target handover content corresponding to the target object is predicted and generated, thereby achieving the purpose of automatically generating handover content that is highly relevant to nursing work, thereby realizing the technical effect of improving the efficiency and accuracy of the handover of nursing work, and further solving the technical problem that the current SBAR handover method still requires manual filling of content, which affects the processing efficiency and data accuracy. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] The drawings described herein are used to provide a further understanding of the present invention and constitute a part of this application. The exemplary embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute an improper limitation of the present invention. In the drawings:

[0018] Figure 1 A hardware structure block diagram of a computer terminal for implementing a method for generating handover content for nursing work is shown;

[0019] Figure 2 is a flow chart of a method for generating handover content for nursing work according to an embodiment of the present invention;

[0020] Figure 3 is a training flow chart of model 2 (first preset model) provided according to an optional embodiment of the present invention;

[0021] Figure 4 is a training flow chart of model 1 (second preset model) provided according to an optional embodiment of the present invention;

[0022] Figure 5 This is a flowchart of a method for generating intelligent SBAR shift handover content based on an AI model according to an optional embodiment of the present invention;

[0023] Figure 64 is a structural block diagram of a device for generating handover content for nursing work according to an embodiment of the present invention. DETAILED DESCRIPTION

[0024] In order to enable those skilled in the art to better understand the solutions of the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the embodiments described are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts should fall within the scope of protection of the present invention.

[0025] It should be noted that the terms "first", "second", etc. in the description and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that the numbers used in this way can be interchanged where appropriate, so that the embodiments of the present invention described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions. For example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0026] According to an embodiment of the present invention, an embodiment of a method for generating handover content for nursing work is provided. It should be noted that the steps shown in the flowchart of the accompanying drawings can be executed in a computer system such as a set of computer executable instructions, and although a logical order is shown in the flowchart, in some cases, the steps shown or described can be executed in an order different from that shown here.

[0027] The method embodiment provided in the first embodiment of the present application can be executed in a mobile terminal, a computer terminal or a similar computing device. Figure 1 FIG. 1 shows a hardware structure block diagram of a computer terminal for implementing a method for generating handover content for nursing work. Figure 1 As shown, the computer terminal 10 may include one or more (illustrated as 102a, 102b, ..., 102n in the figure) processors (the processor may include but is not limited to a microprocessor MCU or a programmable logic device FPGA and other processing devices), a memory 104 for storing data. In addition, it may also include: a display, an input / output interface (I / O interface), a universal serial bus (USB) port (which may be included as one of the ports of the BUS bus), a network interface, a power supply and / or a camera. It will be understood by those skilled in the art that Figure 1 The structure shown is only for illustration and does not limit the structure of the above electronic device. Figure 1 More or fewer components than shown, or with Figure 1 Different configurations shown.

[0028] It should be noted that the one or more processors and / or other data processing circuits described above may generally be referred to herein as "data processing circuitry." The data processing circuitry may be embodied in whole or in part as software, hardware, firmware, or any other combination thereof. Furthermore, the data processing circuitry may be a single, independent processing module, or may be incorporated in whole or in part into any of the other components of the computer terminal 10. As described in the embodiments of the present application, the data processing circuitry serves as a processor control (e.g., selection of a variable resistor terminal path connected to an interface).

[0029] The memory 104 can be used to store software programs and modules of application software, such as the program instructions / data storage device corresponding to the handover content generation method for nursing work in the embodiment of the present invention. The processor executes various functional applications and data processing by running the software programs and modules stored in the memory 104, that is, implementing the handover content generation method for nursing work of the above-mentioned application program. The memory 104 may include a high-speed random access memory and may also include a non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some instances, the memory 104 may further include a memory remotely located relative to the processor, and these remote memories may be connected to the computer terminal 10 via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0030] The display may be, for example, a touch screen liquid crystal display (LCD) that enables a user to interact with a user interface of the computer terminal 10 .

[0031] Figure 2 FIG. 1 is a flow chart of a method for generating handover content for nursing work according to an embodiment of the present invention. Figure 2 As shown, the method includes the following steps:

[0032] Step S201: Acquire initial medical information of the target object, wherein the initial medical information is generated by splicing multiple sentence information.

[0033] In this step, the target object may refer to the patient corresponding to the work content that needs to be handed over, and the initial medical information covers the patient's comprehensive medical condition at a specific time point or time period, and is the basis for generating the target handover content. This information usually comes from the hospital's electronic health record system, including but not limited to the patient's basic information, chief complaint, current medical history, past history, nursing records, vital signs data, auxiliary examination information, etc. This information is multi-source and scattered, and needs to be integrated and processed to a certain extent before it can be used as model input. This information may be scattered in different record sheets, and each sentence or each fragment of information needs to be spliced ​​in a certain logical order to form a coherent text paragraph as basic data.

[0034] Step S202: determining target medical information in the initial medical information, wherein the target medical information is information in the initial medical information whose relevance to target handover content for nursing work exceeds a preset threshold.

[0035] In this step, the initial medical information covers almost all of the patient's medical records, including but not limited to the chief complaint, current medical history, past medical history, nursing records, vital signs data, auxiliary examination information, etc. Although this information is comprehensive, not all content is directly related to the information points that nurses need to focus on when handing over. In order to improve the efficiency and accuracy of generating handover content, it is necessary to screen these initial medical information and find the most important sentences, that is, to determine the target medical information. Specifically, the AI ​​model can be used to determine the target medical information, and the AI ​​model can be trained based on important sentences based on historical handover content. This screening process not only greatly reduces the input amount for subsequent information generation and ensures the efficiency of model processing, but also ensures the accuracy and professionalism of the generated handover content.

[0036] Step S203: Based on the target medical information, predict and generate target handover content corresponding to the target object.

[0037] In this step, once the target medical information is determined, an accurate and easy-to-understand nursing handover content can be generated based on this information, especially the B (background information) and A (assessment information) modules in the SBAR format. Similarly, AI models can also be used for prediction. The AI ​​model receives the target medical information as input, and then predicts and generates the corresponding target handover content through its internal deep learning mechanism. This process involves the model's understanding, reorganization and language expression of the input information, and the final output is highly personalized and contextualized nursing handover content that meets the requirements of the SBAR format.

[0038] Through the above steps, the goal of automatically generating handover content that is highly relevant to nursing work is achieved, thereby achieving the technical effect of improving the efficiency and accuracy of nursing work handover. This solves the technical problem that the current SBAR handover method still requires manual content filling, which affects processing efficiency and data accuracy.

[0039] As an optional embodiment, the initial medical information includes the target object's medical history information or the target object's nursing record content, wherein the medical history information includes the chief complaint, current medical history, past history and auxiliary examination information, and the nursing record content includes vital signs data.

[0040] Optionally, the initial medical information can be divided into two categories, namely medical history information and nursing record content, which correspond to module B (background information) and module A (assessment information) in the SBAR format respectively.

[0041] Module B focuses on the background information of the patient's current health status. The initial medical information—the medical history—is the most basic and critical part of the nursing handover. This medical history can include the following aspects: the chief complaint, which is the patient's primary reason for seeking medical attention and describes the most prominent symptoms or discomfort, providing the nurse with firsthand information about the patient's immediate needs; the history of present illness, which details the course of the patient's illness, including the development of symptoms, treatment history, and current status; the past medical history, which records the patient's past health information, including previous illnesses, surgical experiences, and drug reactions; and auxiliary examination information, which covers the results of various medical examinations the patient has undergone, such as laboratory tests and imaging studies.

[0042] In contrast, Module A focuses on the patient's current health assessment, as well as changes in vital signs observed and nursing care measures taken during the current shift. Module A's initial medical information may include vital sign data such as temperature, blood pressure, heart rate, and respiratory rate, reflecting the patient's immediate physiological status. The nursing record also includes all nursing activities performed at various times throughout the day, such as medication records, special treatments, and patient activity and responses. These are all real-time data essential for assessing the patient's condition and formulating nursing strategies.

[0043] As an optional embodiment, when the initial medical information is medical history information, obtaining the initial medical information of the target object includes: obtaining multiple sentence information corresponding to the medical history information; sorting and splicing the multiple sentence information in the order of the sentences corresponding to the chief complaint, the sentences corresponding to the current medical history, the sentences corresponding to the past history, and the sentences corresponding to the auxiliary examination information to obtain the initial medical information.

[0044] Alternatively, the target object, or detailed information about the patient's medical history, can be obtained from the hospital's electronic health record system. This information is divided into four main sections: chief complaint, present illness history, past medical history, and auxiliary examination information. Each section consists of a series of statements related to the patient's health status. After obtaining these statements, they can be sorted and integrated according to the logical sequence of Module B in the SBAR format. All statements related to the chief complaint should be placed first. These statements should clearly describe the patient's primary reason for seeking medical attention and the symptoms they present. Following the chief complaint, the statements from the present medical history should be arranged, detailing the patient's medical progression and treatment response from admission to the current time point. Next, statements from the past medical history should be integrated, including records of previous illnesses, surgeries, and allergic reactions. Finally, all statements from auxiliary examination information should be concatenated at the end of the sequence. This information provides objective evidence of the patient's physiological and pathological conditions. This not only captures a comprehensive picture of the patient's medical history, but also organizes and integrates them according to the logical order of information transmission in clinical care, forming a structured initial medical record.

[0045] As an optional embodiment, when the initial medical information is the content of a nursing record, obtaining the initial medical information of the target object includes: obtaining multiple sentence information corresponding to the content of the nursing record, wherein the multiple sentence information each corresponds to a recording time; sorting and splicing the multiple sentence information in order from small to large according to the time interval between the recording time and the current time to obtain the initial medical information.

[0046] Alternatively, in the hospital's electronic health record system, there are a large number of nursing records that detail the patient's (target object's) nursing activities and vital signs since admission. Each record contains one or more sentences describing the patient's condition at a certain point in time or the measures taken by the nursing staff. For example, the record may include "10:00 AM, the patient's heart rate is stable and there are no obvious abnormalities" or "2:30 PM, intravenous infusion treatment was given, and the patient responded well."

[0047] Next, the acquired nursing records are sorted by time. Specifically, the current time is determined, and then the time interval between each nursing record and the current time is calculated based on the timestamp of each nursing record. All statements are then sorted in ascending order of time interval, prioritizing the most recent nursing records. Once sorted, the statements corresponding to all nursing records are concatenated in chronological order to form the initial, time-ordered medical information.

[0048] As an optional embodiment, determining target medical information in the initial medical information includes: inputting the initial medical information into a first preset model to obtain an index of the target medical information, wherein the architecture of the first preset model is a BERT bidirectional encoder architecture; based on the index, locating the target medical information in the initial medical information.

[0049] Optionally, the role of the first preset model is to find the most important sentences in the initial medical information, that is, the sentences most relevant to the handover content, as the target medical information, so that the final generated handover content is generated based on these important sentences. The first preset model can be a BERT (Bidirectional Encoder Representations from Transformers) architecture, which processes the input initial medical information through a deep bidirectional context-aware mechanism. After the model completes processing the information, it generates an output containing the importance score of each sentence or information fragment, which is the so-called "index". These indexes are actually location identifiers pointing to specific sentences or information fragments in the initial medical information. Sentences or information fragments with higher scores mean that they are more relevant and important in the generation of nursing handover content.

[0050] Specifically, when determining the training of the first preset model, it is necessary to prepare a training set first. The data source is composed of n sentences, and different modules select different n sentences. For example, for module B, the chief complaint, current medical history, past history and auxiliary examination information are spliced ​​together as the data source; for module A, the nursing record contents are spliced ​​together in chronological order from the most recent to the most distant as the data source. Let the data source be X = (x1, x2, ..., x n ), where x1,x2,...,x n is n sentences that are ordered together to form the data source X. Label 1 is the actual content filled in, recorded as Y = (y1, y2, ..., y m ), where y1,y2,...,y m is the m sentences of actual content, m≤n. Label 2 is the index of the data source that is closest to the actual content, denoted as Y2, which is a list. The index number starts from 0, so the indexes of the n ordered sentences in the data source are 0, 1, ..., n-1 respectively. You can use the similarity comparison method, set the comparison threshold ε, and compare y1, y2, ..., y m Each sentence in x1,x2,...,x n Calculate the similarity of each sentence in , for example, compare y1 with x1, x2, ..., x nDo similarity calculation. If the similarity between y1 and x2 is greater than the threshold ε, the index corresponding to y1 is considered to be 1. At this time, after removing x2, calculate the similarity between y2 and x1, x3, ..., x n Similarity, get the index corresponding to y2, repeat the above operation until y m The corresponding index.

[0051] For example, the data source is X=(x1,x2,x3,x4,x5,x6,x7,x8,x9,x 10 ), Y1=(y1,y2,y3,y4,y5), where y1,y2,y3,y4,y5 correspond to the data sources with high similarity, specifically x2,x3,x4,x5,x9, then Y2=[1,2,3,4,8]. If Y1=(y1,y2,y3,y4,y5) has some y (assuming it is y2,y3) in X=(x1,x2,x3,x4,x5,x6,x7,x8,x9,x 10 ) can not find any x with similarity greater than ε, and the data sources with high similarity corresponding to y1,y4,y5 are x2,x5,x9, then Y2=[1,4,8]. If any y in Y1=(y1,y2,y3,y4,y5) is in X=(x1,x2,x3,x4,x5,x6,x7,x8,x9,x 10 ) can not find any x with similarity greater than ε, then the index of the entire X is taken as Y2, that is, Y2 = [0,1,2,3,4,5,6,7,8,9].

[0052] After preparing the data source and label pairs, you can start model training. Figure 3 is a training flow chart of model 2 (first preset model) provided according to an optional embodiment of the present invention, such as Figure 3 As shown, the data source X and label 2Y2 are used as the training set, and the data label pair is (X, Y2). One-hot encoding is performed on label 2, that is, when X=(x1,x2,x3,x4,x5,x6,x7,x8,x9,x 10), when Y2 = [1, 2, 3, 4, 8], Y2 is converted to Y2 = (0, 1, 1, 1, 1, 0, 0, 0, 1, 0). The model's predicted value is Y2^, for example, Y2^ = (0.005, 0.2, 0.18, 0.179, 0.166, 0.003, 0.007, 0.002, 0.15, 0.001), which represents the model's score of the importance of each sentence. The loss is then calculated using the binary cross-entropy loss function, and the model parameters are updated through backpropagation. The above example assumes a batch size of 1 for data label pairs, meaning each batch contains only one data label pair. If the batch size is set to another specific number, batch_size = n, then n random data label pairs are fed into the model each time, and the loss is calculated through forward propagation. This results in n loss values, and the model parameters are updated through backpropagation based on the average loss value. Repeat this process until all data label pairs participate in the calculation. Assuming that the total number of data label pairs is N, the model training cycle will have at least N / batch_size parameter updates.

[0053] The first preset model is The model architecture can be BERT, the binary cross entropy loss function is L2, the training period is set to T, the batch size of the data label pairs fed into the model is set to batch_size, and the total amount of data is N. The optimization problem is to minimize the loss, which can be described as:

[0054]

[0055] in, represents the first preset model, Indicates the predicted value Y2 of the first preset model for the batch data source X ∧ , Indicates the predicted value Y2 for this batch ∧ The model updates its parameters once per batch until T cycles of training are completed, at which point parameter updates are stopped to obtain the final first preset model.

[0056] As an optional embodiment, based on the target medical information, the target handover content corresponding to the target object is predicted and generated, including: inputting the target medical information into a second preset model to obtain the target handover content, wherein the architecture of the second preset model is the mT5 multilingual pre-trained text-to-text converter architecture.

[0057] Optionally, in order to convert the target medical information into handover content suitable for the SBAR format, it can be passed as input to a second preset model. The architecture of the second preset model is mT5, the full name of which is multilingual T5. It is a multilingual pre-trained model based on the Transformer architecture. It is particularly good at text-to-text conversion tasks, including semantic understanding, summary generation, question-answering systems, etc. When the target medical information is input into the mT5 model, the model can use its pre-trained multilingual understanding and generation capabilities to perform in-depth analysis and semantic understanding of the input information. Understand the meaning and contextual relationship of each information fragment, and then generate new text output based on these understandings, that is, the target handover content.

[0058] Similar to the first preset model, the model training can be performed using the prepared data source and label pairs. Figure 4 is a training flow chart of model 1 (second preset model) provided according to an optional embodiment of the present invention, such as Figure 4 As shown in the figure, data source X, label 1Y1 and label 2Y2 are used as training sets, and the data label pair is (X, Y1, Y2). Define an index function G(X, Y2) to find the corresponding sentence in the data source according to label 2Y2. That is, assuming that when X=(x1,x2,x3,x4,x5,x6,x7,x8,x9,x 10 ), and Y2 = [1,2,3,4,8], then based on G(X,Y2), we obtain G(X,Y2) = (x2,x3,x4,x5,x9). G(X,Y2) is fed into the second pre-set model to obtain the predicted value Y1^, which is a paragraph consisting of at least one sentence. The cross-entropy loss function is then used to calculate the loss, and the model parameters are updated through backpropagation. The above example assumes that the batch size of data label pairs, batch_size, is set to 1, meaning that each batch contains only one data label pair. If the batch size is set to another specific number, batch_size = n, then n random data label pairs are fed into the model each time, and the loss is calculated through forward propagation. This results in n loss values, and the model parameters are updated through backpropagation based on the average loss value. This process is repeated until all data label pairs are included in the calculation. Assuming that the total number of data label pairs is N, then one cycle of model training involves at least N / batch_size parameter updates.

[0059] The second preset model is The model architecture can be mT5, the cross entropy loss function is L, the training period is set to T, the batch size of the data label pairs fed into the model is set to batch_size, and the total amount of data is N. The optimization problem is to minimize the loss, which can be described as:

[0060]

[0061] in, represents the second preset model, represents the predicted value Y1^ of the second preset model for the input G(X,Y2), The model updates its parameters once per batch until T cycles of training are completed, at which point parameter updates are stopped, resulting in the final second preset model.

[0062] As an optional embodiment, a method for generating intelligent SBAR shift handover content based on an AI model is also provided. Figure 5 is a flow chart of a method for generating intelligent SBAR handover content based on an AI model according to an optional embodiment of the present invention. Figure 5 As shown in the figure, for modules B and A, the corresponding inputs are automatically generated according to relevant information through AI algorithms, and nurses can edit the recommended inputs, which greatly shortens the workload of handover nurses.

[0063] Specifically, for module B, obtain Model 1 and Model 2, which are well-trained and trained using the training set of the module B data source. Obtain a patient's medical desensitized data, such as the chief complaint, current medical history, past medical history, and auxiliary examination information. These data are spliced ​​together in the order of the chief complaint, current medical history, past medical history, and auxiliary examination information as the data source. Send the data source to Model 2 to obtain the index of the important sentences in the data source that the model believes belong to the handover content of module B. According to the index, correspond to the specific important sentences in the data source, and then send these important sentences to Model 1 to obtain the final generated handover content of module B.

[0064] For Module A, obtain Model 1 and Model 2, both well-trained using the Module A data source training set. Obtain a patient's desensitized medical data, such as nursing records and vital signs data such as thermometers. These nursing records are then concatenated in chronological order from most recent to most recent, serving as the data source. This data source is fed into Model 2, which then obtains the indexes of important sentences in the data source that the model deems relevant to the handover content for Module A. Based on these indexes, the model maps these important sentences to specific important sentences in the data source. These sentences are then fed into Model 1 to generate the final handover content for Module A.

[0065] It should be noted that for the aforementioned method embodiments, for simplicity of description, they are all expressed as a series of action combinations. However, those skilled in the art should be aware that the present invention is not limited by the order of the actions described, because according to the present invention, certain steps can be performed in other orders or simultaneously. Secondly, those skilled in the art should also be aware that the embodiments described in this specification are all preferred embodiments, and the actions and modules involved are not necessarily required by the present invention.

[0066] Through the description of the above implementation methods, those skilled in the art can clearly understand that the handover content generation method for nursing work according to the above embodiment can be implemented by means of software plus the necessary general hardware platform. Of course, it can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, can be embodied in the form of a software product. The computer software product is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk), and includes a number of instructions for enabling a terminal device (which can be a mobile phone, computer, server, or network device, etc.) to execute the methods described in each embodiment of the present invention.

[0067] According to an embodiment of the present invention, there is also provided a device for implementing the above-mentioned method for generating handover content for nursing work. Figure 6 is a structural block diagram of a device for generating handover content for nursing work according to an embodiment of the present invention. Figure 6 As shown, the device includes: an acquisition module 61, a determination module 62 and a prediction module 63. The device is described below.

[0068] The acquisition module 61 is used to acquire the initial medical information of the target object, wherein the initial medical information is generated by splicing multiple sentence information.

[0069] The determination module 62 is connected to the acquisition module 61 and is used to determine the target medical information in the initial medical information, wherein the target medical information is information in the initial medical information whose relevance to the target handover content for nursing work exceeds a preset threshold.

[0070] The prediction module 63 is connected to the determination module 62 and is used to predict and generate target handover content corresponding to the target object based on the target medical information.

[0071] It should be noted that the acquisition module 61, determination module 62, and prediction module 63 described above correspond to steps S201 to S203 in the embodiment. The examples and application scenarios implemented by the various modules and corresponding steps are the same, but are not limited to the contents disclosed in the above embodiment. It should be noted that the above modules, as part of the device, can be run in the computer terminal 10 provided in the embodiment.

[0072] An embodiment of the present invention may provide a computer device. Optionally, in this embodiment, the computer device may be located in at least one of a plurality of network devices in a computer network. The computer device includes a memory and a processor.

[0073] Among them, the memory can be used to store software programs and modules, such as the program instructions / modules corresponding to the handover content generation method and device for nursing work in the embodiment of the present invention. The processor executes various functional applications and data processing by running the software programs and modules stored in the memory, that is, realizing the above-mentioned handover content generation method for nursing work. The memory may include a high-speed random access memory, and may also include a non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some instances, the memory may further include a memory remotely arranged relative to the processor, and these remote memories can be connected to the computer terminal via a network. Examples of the above-mentioned network include but are not limited to the Internet, corporate intranet, local area network, mobile communication network and combinations thereof.

[0074] The processor can call the information and application stored in the memory through the transmission device to perform the following steps: obtain the initial medical information of the target object, wherein the initial medical information is generated by splicing multiple sentence information; determine the target medical information in the initial medical information, wherein the target medical information is information in the initial medical information whose correlation with the target handover content used for nursing work exceeds a preset threshold; based on the target medical information, predict and generate the target handover content corresponding to the target object.

[0075] Optionally, the above-mentioned processor can also execute the program code of the following steps: the initial medical information includes the medical history information of the target object or the nursing record content of the target object, wherein the medical history information includes the chief complaint, current medical history, past history and auxiliary examination information, and the nursing record content includes vital signs data.

[0076] Optionally, the processor may also execute the program code of the following steps: when the initial medical information is medical history information, obtaining the initial medical information of the target object, including: obtaining multiple statement information corresponding to the medical history information; sorting and splicing the multiple statement information in the order of the statements corresponding to the chief complaint, the statements corresponding to the current medical history, the statements corresponding to the past history, and the statements corresponding to the auxiliary examination information to obtain the initial medical information.

[0077] Optionally, the processor may also execute the program code of the following steps: when the initial medical information is the content of a nursing record, obtaining the initial medical information of the target object, including: obtaining multiple sentence information corresponding to the content of the nursing record, wherein each of the multiple sentence information corresponds to a recording time; sorting and splicing the multiple sentence information in order from small to large according to the time interval between the recording time and the current time to obtain the initial medical information.

[0078] Optionally, the above-mentioned processor can also execute the program code of the following steps: determining the target medical information in the initial medical information, including: inputting the initial medical information into a first preset model to obtain an index of the target medical information, wherein the architecture of the first preset model is a BERT bidirectional encoder architecture; based on the index, locating the target medical information in the initial medical information.

[0079] Optionally, the above-mentioned processor can also execute the program code of the following steps: based on the target medical information, predict and generate the target handover content corresponding to the target object, including: inputting the target medical information into a second preset model to obtain the target handover content, wherein the architecture of the second preset model is the mT5 multilingual pre-trained text-to-text converter architecture.

[0080] According to an embodiment of the present invention, a method for generating handover content for nursing work is provided. The method obtains initial medical information of a target object, wherein the initial medical information is generated by splicing multiple sentence information; in the initial medical information, target medical information is determined, wherein the target medical information is information in the initial medical information whose relevance to the target handover content for nursing work exceeds a preset threshold; based on the target medical information, target handover content corresponding to the target object is predicted and generated, thereby achieving the purpose of automatically generating handover content that is highly relevant to nursing work, thereby achieving the technical effect of improving the efficiency and accuracy of nursing work handover, and further solving the technical problem that the current SBAR handover method still requires manual content filling, which affects processing efficiency and data accuracy.

[0081] A person skilled in the art will understand that all or part of the steps in the various methods of the above embodiments can be completed by instructing the hardware related to the terminal device through a program, and the program can be stored in a non-volatile storage medium, which may include: a flash drive, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, etc.

[0082] The embodiment of the present invention further provides a non-volatile storage medium. Optionally, in this embodiment, the non-volatile storage medium can be used to store program codes executed by the method for generating handover content for nursing work provided in the embodiment.

[0083] Optionally, in this embodiment, the non-volatile storage medium may be located in any computer terminal in a computer terminal group in a computer network, or in any mobile terminal in a mobile terminal group.

[0084] Optionally, in this embodiment, the non-volatile storage medium is configured to store program code for executing the following steps: obtaining initial medical information of the target object, wherein the initial medical information is generated by splicing multiple sentence information; determining target medical information in the initial medical information, wherein the target medical information is information in the initial medical information whose correlation with the target handover content used for nursing work exceeds a preset threshold; based on the target medical information, predicting and generating target handover content corresponding to the target object.

[0085] Optionally, in this embodiment, the non-volatile storage medium is configured to store program code for executing the following steps: the initial medical information includes the medical history information of the target object or the nursing record content of the target object, wherein the medical history information includes the chief complaint, current medical history, past history and auxiliary examination information, and the nursing record content includes vital signs data.

[0086] Optionally, in this embodiment, the non-volatile storage medium is configured to store program code for executing the following steps: when the initial medical information is medical history information, obtaining the initial medical information of the target object, including: obtaining multiple statement information corresponding to the medical history information; sorting and splicing the multiple statement information in the order of the statements corresponding to the chief complaint, the statements corresponding to the current medical history, the statements corresponding to the past history, and the statements corresponding to the auxiliary examination information to obtain the initial medical information.

[0087] Optionally, in this embodiment, the non-volatile storage medium is configured to store program code for executing the following steps: when the initial medical information is the content of a nursing record, obtaining the initial medical information of the target object, including: obtaining multiple sentence information corresponding to the content of the nursing record, wherein the multiple sentence information each corresponds to a recording moment; sorting and splicing the multiple sentence information in order from small to large according to the time interval between the recording moment and the current moment to obtain the initial medical information.

[0088] Optionally, in this embodiment, the non-volatile storage medium is configured to store program code for performing the following steps: determining target medical information in the initial medical information, including: inputting the initial medical information into a first preset model to obtain an index of the target medical information, wherein the architecture of the first preset model is a BERT bidirectional encoder architecture; based on the index, locating the target medical information in the initial medical information.

[0089] Optionally, in this embodiment, the non-volatile storage medium is configured to store program code for executing the following steps: based on the target medical information, predicting and generating target handover content corresponding to the target object, including: inputting the target medical information into a second preset model to obtain the target handover content, wherein the architecture of the second preset model is the mT5 multilingual pre-trained text-to-text converter architecture.

[0090] An embodiment of the present invention also provides a computer program product, including a computer program. Optionally, in this embodiment, when the computer program is executed by a processor, it can achieve: obtaining initial medical information of the target object, wherein the initial medical information is generated by splicing multiple sentence information; determining target medical information in the initial medical information, wherein the target medical information is information in the initial medical information whose correlation with the target handover content used for nursing work exceeds a preset threshold; based on the target medical information, predicting and generating target handover content corresponding to the target object.

[0091] The serial numbers of the above embodiments of the present invention are for description only and do not represent the advantages or disadvantages of the embodiments.

[0092] In the above embodiments of the present invention, the description of each embodiment has its own focus. For parts that are not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0093] In the several embodiments provided in this application, it should be understood that the disclosed technical content can be implemented in other ways. Among them, the device embodiments described above are only exemplary. For example, the division of the units can be a logical function division. In actual implementation, there may be other division methods, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of units or modules, which can be electrical or other forms.

[0094] The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple units. Some or all of the units may be selected according to actual needs to achieve the purpose of the present embodiment.

[0095] In addition, the functional units in the various embodiments of the present invention may be integrated into a single processing unit, each unit may exist physically separately, or two or more units may be integrated into a single unit. The aforementioned integrated units may be implemented in the form of hardware or software functional units.

[0096] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a non-volatile storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes several instructions for enabling a computer device (which can be a personal computer, server or network device, etc.) to execute all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), mobile hard disk, magnetic disk or optical disk, and other media that can store program code.

[0097] The above is only a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles of the present invention. These improvements and modifications should also be regarded as the scope of protection of the present invention.

Claims

1. A method for generating handover content for nursing work, characterized in that: include: Acquiring initial medical information of a target subject, wherein the initial medical information is generated by concatenating multiple sentence information; Determining target medical information from the initial medical information, wherein the target medical information is information in the initial medical information whose relevance to target handover content for nursing work exceeds a preset threshold; Based on the target medical information, the target handover content corresponding to the target object is predicted and generated.

2. The method according to claim 1, characterized in that The initial medical information includes the medical history information of the target object or the content of the nursing record of the target object, wherein the medical history information includes the chief complaint, current medical history, past medical history and auxiliary examination information, and the content of the nursing record includes vital signs data.

3. The method according to claim 2, characterized in that In a case where the initial medical information is the medical history information, obtaining the initial medical information of the target object includes: Acquire multiple sentence information corresponding to the medical history information; According to the order of the statement corresponding to the chief complaint, the statement corresponding to the current medical history, the statement corresponding to the past history and the statement corresponding to the auxiliary examination information, the multiple statement information are sorted and spliced ​​to obtain the initial medical information.

4. The method according to claim 2, characterized in that In a case where the initial medical information is the content of the nursing record, the step of obtaining the initial medical information of the target object includes: Acquire multiple sentence information corresponding to the nursing record, wherein each of the multiple sentence information corresponds to a recording time; The multiple statement information are sorted and spliced ​​in the order of the time interval between the recording time and the current time from small to large to obtain the initial medical information.

5. The method according to claim 1, characterized in that Determining target medical information from the initial medical information includes: Inputting the initial medical information into a first preset model to obtain an index of the target medical information, wherein the architecture of the first preset model is a BERT bidirectional encoder architecture; Based on the index, the target medical information is located in the initial medical information.

6. The method according to any one of claims 1 to 5, characterized in that The predicting and generating the target handover content corresponding to the target object based on the target medical information includes: The target medical information is input into a second preset model to obtain the target handover content, wherein the architecture of the second preset model is the mT5 multilingual pre-trained text-to-text converter architecture.

7. A handover content generation device for nursing work, characterized in that: include: An acquisition module, configured to acquire initial medical information of a target subject, wherein the initial medical information is generated by concatenating multiple sentence information; a determination module, configured to determine target medical information from the initial medical information, wherein the target medical information is information in the initial medical information whose relevance to target handover content for nursing work exceeds a preset threshold; A prediction module is used to predict and generate the target handover content corresponding to the target object based on the target medical information.

8. A non-volatile storage medium, characterized in that: The non-volatile storage medium includes a stored program, wherein when the program is running, the device where the non-volatile storage medium is located is controlled to execute the handover content generation method for nursing work according to any one of claims 1 to 6.

9. A computer device, characterized in that: include: memory and processor, The memory stores a computer program; The processor is configured to execute a computer program stored in the memory, and when the computer program is run, the processor is configured to execute the method for generating handover content for nursing work according to any one of claims 1 to 6.

10. A computer program product comprising a computer program, characterized in that When the computer program is executed by a processor, the method for generating handover content for nursing work according to any one of claims 1 to 6 is implemented.

Images