Clinical nursing multifunctional electronic medical record generation method and system

Abstract

The invention belongs to the technical field of internet medical treatment, and discloses a multifunctional electronic medical record generation method and system for clinical nursing, and the method comprises the steps: generating and storing an event ID uniquely corresponding to a nursing event; collecting multi-modal care data associated with the event ID; verifying the validity of the standard scale reference object in the nursing image; inputting the nursing image passing the verification into a multi-label classification model; performing structured information extraction on the supplementary record data; performing associated storage on the event ID, the nursing image, the AI tag and the structured nursing text; according to a comparison report generation request of the target patient, analyzing the acquired nursing images at the at least two different time points; and generating an image-text combined nursing comparison report. According to the invention, a dynamic and continuous electronic nursing file is constructed for each patient, so that a traditional discrete recording mode which takes text as a main part is thoroughly changed, and the objectivity, integrity and specialization level of nursing records are improved.

Description

Technical Field

[0001] This invention relates to the field of internet healthcare technology, and in particular to a method and system for generating multifunctional electronic medical records for clinical nursing. Background Technology

[0002] In recent years, with the deep integration of artificial intelligence and mobile internet technology in the medical field, electronic medical record systems have gradually become an important tool for improving diagnostic and treatment efficiency and standardizing the quality of medical documents. Several AI-based electronic medical record generation solutions have emerged in the existing technology. For example, patent document CN 120199392 A discloses a method for generating electronic medical records for skin consultations based on a deep learning model. By combining doctor-patient dialogue content with multimodal data such as skin images and medical reports, it utilizes a large language model to generate electronic medical records, thereby improving the quality of dermatology medical record generation.

[0003] However, current technologies primarily focus on generating medical record texts for outpatient or online consultation scenarios, and have not yet fully met the urgent needs of clinical nursing, especially bedside nursing, for comprehensive, multimodal, comparable, and structured recording of the entire course of a patient's condition. In clinical nursing practice, nurses need to continuously observe and record patients' wounds, skin conditions, and tubing status. Traditional methods, relying on paper records or simple electronic documents, have the following prominent problems: The information format is too limited to fully record the condition: Existing systems are mostly text-based, making it difficult to easily integrate information such as wound photos, videos, and voice descriptions, and thus failing to intuitively reflect the true state of the condition.

[0004] Lack of continuity and comparability: Nursing records are often scattered across different time points, lacking effective timeline views and comparison tools, making it difficult to intuitively show the process of wound healing and disease progression, which is not conducive to efficacy assessment and nursing decisions.

[0005] The recording process is inefficient and lacks intelligence: nurses need to manually input a large amount of descriptive text, and photos and text are separated, making it impossible to achieve automated annotation, key information extraction, and intelligent report generation, which increases their workload. Summary of the Invention

[0006] In order to solve the technical problems existing in the prior art, the purpose of this invention is to provide a multifunctional electronic medical record generation method and system for clinical nursing, which is suitable for nursing scenarios that require continuous observation, recording and evaluation of changes in the condition of wounds or specific sites, and can be executed collaboratively by a nurse's handheld mobile terminal and a backend server.

[0007] To achieve this objective, the present invention adopts the following technical solution: In a first aspect, the present invention provides a method for generating multifunctional electronic medical records for clinical nursing, comprising the following steps: S1. Create a nursing record creation request for the target patient, generate and store an event ID that uniquely corresponds to this nursing event; S2. Collect multimodal nursing data associated with the event ID; the multimodal nursing data includes at least nursing images containing standard scale references and supplementary record data associated with the nursing images; S3. Verify the validity of the standard scale reference in the nursing images; S4. Input the verified nursing image into a multi-label classification model to obtain at least one AI label for the nursing image; S5. Extract structured information from the supplementary record data to obtain structured nursing text; S6. The event ID, nursing image, AI tag and structured nursing text are associated and stored, and displayed in chronological order in the timeline view of the target patient; S7. For the target patient, generate a comparison report request and obtain nursing images and associated structured nursing texts from at least two different time points; S8. Analyze the nursing images acquired at least two different time points to generate visual comparison analysis results of changes in the nursing site; S9. Based on the visual comparison analysis results and the structured nursing text at the corresponding time points, generate a nursing comparison report combining text and images.

[0008] As a preferred method for generating multifunctional electronic medical records in clinical nursing, step S3 specifically includes: S31. Identify whether a standard scale reference object exists in the nursing image using a target detection model; S32. Verify whether the integrity, clarity, and position of the standard scale reference object in the nursing image conform to the preset shooting specifications.

[0009] As a preferred embodiment of a multifunctional electronic medical record generation method for clinical nursing, the supplementary recording data includes at least voice recording data, and step S5 specifically includes: S51. Input the voice recording data in the supplementary recording data into the voice recognition model and convert it into initial text; S52. Use a medical named entity recognition model to extract key medical entities and numerical information from the initial text; S53. Fill the extracted information into the corresponding fields of the predefined nursing record structure template to form structured nursing text.

[0010] As a preferred method for generating multifunctional electronic medical records in clinical nursing, step S8 specifically includes: S81. Use an image segmentation model to segment the standard scale reference area and the target nursing area in each nursing image; S82. Based on the pixel size of the segmented standard scale reference area and its known actual physical size, calculate the scaling factor of the current nursing image; S83. Calculate the pixel area covered by the pixel mask of the target care area; S84. Multiply the pixel area by the square of the scaling factor to obtain the actual physical area of ​​the target care area; S85. Calculate the actual physical area change rate of nursing images at two different time points; S86. Extract the color and texture features of the target nursing area from each nursing image; S87. Calculate the color feature distance and texture feature distance between target care areas at different time points; S88. Generate visual contrast analysis results based on the actual physical area change rate, color feature distance, and texture feature distance.

[0011] As a preferred method for generating multifunctional electronic medical records in clinical nursing, it also includes access control: Based on the logged-in user's role information, the scope of patient data that the user can access and the functional modules that can be operated are dynamically determined; the role information includes at least nurses, head nurses, and doctors, and nurses can only access the patient data they are responsible for.

[0012] Secondly, the present invention provides a multifunctional electronic medical record generation system for clinical nursing, comprising: The record creation module is used to generate and store an event ID that uniquely corresponds to the nursing event in response to the nursing record creation request of the target patient. A multimodal data acquisition module is used to acquire multimodal nursing data associated with the event ID; the multimodal nursing data includes at least nursing images containing standard scale references and supplementary record data associated with the nursing images; The reference object verification module is used to verify the validity of standard scale reference objects in the nursing images; An image classification module is used to input the nursing image into an AI analysis model to obtain at least one AI label for the nursing image; the AI ​​label includes at least the nursing site, the type of nursing problem, and the severity level. The text structuring module is used to extract structured information from the supplementary record data to obtain structured nursing text; The storage and display module is used to associate and store the event ID, nursing images, AI tags and structured nursing text, and display them in chronological order in the timeline view of the target patient; The comparison analysis module is used to generate comparison reports for target patients by acquiring nursing images and associated structured nursing text at at least two different time points, analyzing the acquired nursing images, and generating visual comparison analysis results of changes in nursing sites. The report generation module is used to generate a nursing comparison report combining text and images based on the visual comparison analysis results and the structured nursing text at the corresponding time points.

[0013] As a preferred embodiment of a multifunctional electronic medical record generation system for clinical nursing, the reference verification module specifically includes: The detection unit is used to identify whether a standard scale reference object exists in the nursing image through a target detection model; The verification unit is used to verify whether the integrity, clarity, and position of the standard scale reference object in the nursing image conform to the preset shooting specifications.

[0014] As a preferred embodiment of a multifunctional electronic medical record generation system for clinical nursing, the text structuring module specifically includes: The speech recognition unit is used to input the speech recording data from the supplementary recording data into the speech recognition model and convert it into initial text. An information extraction unit is used to extract key medical entities and numerical information from the initial text using a medical named entity recognition model. The template filling unit is used to fill the extracted information into the corresponding fields of a predefined nursing record structure template to form structured nursing text.

[0015] As a preferred embodiment of a multifunctional electronic medical record generation system for clinical nursing, the comparative analysis module specifically includes: The image segmentation unit is used to segment the standard scale reference area and the target nursing area in each nursing image; The image calculation unit is used to calculate the actual physical area change rate, color feature distance, and texture feature distance of nursing images at two different time points.

[0016] Thirdly, the present invention provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that the processor executes the computer program to implement the above-described method for generating multifunctional electronic medical records for clinical care.

[0017] Compared with the prior art, the beneficial effects of the present invention are as follows: (1) This invention uses event ID as the core to associate multimodal data such as nursing images, voice recordings, and text descriptions, and uses AI models to automatically perform image analysis, tagging, and text structuring, so that each nursing event can form a complete nursing record and be clearly displayed in the form of a timeline. This creates a dynamic and continuous electronic nursing file for each patient, which completely changes the traditional text-based discrete recording mode and improves the objectivity, completeness, and professionalism of nursing records.

[0018] (2) By integrating functions such as mobile terminal photography, AI-assisted reference object recognition, voice recognition and automatic text structuring, the system automatically processes and replaces a large amount of tedious manual input, classification and summarization work. Nurses can quickly complete patient record work at the bedside, which improves the efficiency of nurses' work.

[0019] (3) This invention utilizes computer vision technology to achieve precise quantitative measurement and cross-time point comparison of the area, color, and texture features of wounds and other nursing sites. The system generates a graphic comparison report that transforms the patient's healing process into intuitive images and quantitative data, providing a visual basis for nursing effect evaluation, treatment plan adjustment, and communication between medical staff and patients, thereby improving the scientific nature and precision of nursing. Attached Figure Description

[0020] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0021] Figure 1 This is a flowchart illustrating the multifunctional electronic medical record generation method for clinical nursing as described in Embodiment 1 of the present invention.

[0022] Figure 2 This is a schematic diagram of the standard scale reference verification process described in Embodiment 1 of the present invention.

[0023] Figure 3 This is a schematic diagram of the structured information extraction process described in Embodiment 1 of the present invention.

[0024] Figure 4 This is a schematic diagram of the visual contrast analysis process described in Embodiment 1 of the present invention.

[0025] Figure 5 This is a schematic diagram of the functional modules of the multifunctional electronic medical record generation system for clinical nursing described in Embodiment 2 of the present invention.

[0026] Explanation of reference numerals in the attached figures: 1. Record creation module; 2. Multimodal data acquisition module; 3. Reference object verification module; 4. Image classification module; 5. Text structuring module; 6. Storage and display module; 7. Comparative analysis module; 8. Report generation module. Detailed Implementation

[0027] The technical solution of the present invention will be further described below with reference to the accompanying drawings and specific embodiments.

[0028] The accompanying drawings are for illustrative purposes only and are schematic diagrams, not actual images. They should not be construed as limiting the scope of this patent. To better illustrate the embodiments of the present invention, some parts in the drawings may be omitted, enlarged, or reduced, and do not represent the actual dimensions of the product. It is understandable to those skilled in the art that some well-known structures and their descriptions may be omitted in the drawings.

[0029] In the accompanying drawings of the embodiments of the present invention, the same or similar reference numerals correspond to the same or similar components. In the description of the present invention, it should be understood that if terms such as "upper," "lower," "left," "right," "inner," and "outer" indicate the orientation or positional relationship based on the orientation or positional relationship shown in the drawings, they are only for the convenience of describing the present invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, the terms used to describe positional relationships in the drawings are only for illustrative purposes and should not be construed as limiting the present patent. For those skilled in the art, the specific meaning of the above terms can be understood according to the specific circumstances.

[0030] In the description of this invention, unless otherwise explicitly specified and limited, the term "connection" or similar designation indicating a connection between components should be interpreted broadly. For example, it can refer to a fixed connection, a detachable connection, or an integral part; it can be a mechanical connection or an electrical connection; it can be a direct connection or an indirect connection through an intermediate medium; it can refer to the internal communication between two components or the interaction between two components. Those skilled in the art can understand the specific meaning of the above terms in this invention based on the specific circumstances.

[0031] Example 1: like Figure 1 As shown, this embodiment provides a multifunctional electronic medical record generation method for clinical nursing. This method is particularly suitable for nursing scenarios that require continuous observation, recording, and evaluation of changes in the condition of wounds or specific areas. It can be executed collaboratively by a nurse's handheld mobile terminal and a backend server. The method mainly includes the following steps: Step S1: Create a nursing record request for the target patient, generate and store an event ID that uniquely corresponds to this nursing event; When a nurse needs to perform a nursing procedure on a patient and record it, such as changing a dressing for a pressure ulcer on the sacrum and coccyx, the nurse selects the target patient in the mobile application and clicks the "New Record" function to trigger a nursing record creation request. The backend server responds to the request and immediately generates a globally unique event ID.

[0032] Preferably, this embodiment can use the Snowflake ID algorithm to generate this event ID, for example, the event ID is 2025121510000001, thereby ensuring its uniqueness and timeliness in the distributed system. Simultaneously, the system records the current precise timestamp, for example, 2025-04-25 10:30:05, and binds this timestamp to the event ID. This event ID will become the core index for all subsequent related data.

[0033] Step S2: Collect multimodal nursing data associated with the event ID; Specifically, while performing nursing procedures at the bedside, nurses collect various forms of data through mobile terminals, including taking nursing images and entering supplementary record data. All data is associated with the aforementioned event ID when uploaded.

[0034] More specifically, multimodal nursing data includes at least nursing images containing standard scale references and supplementary record data associated with the nursing images; the standard scale references are preferably medical color cards with a clear 1cm×1cm grid; the supplementary record data includes at least voice recording data, and nurses can add supplementary information simultaneously or subsequently by voice recording wound descriptions, filling out structured forms, or inputting free text.

[0035] When taking nursing images, the application will call the device's camera and guide the nurse to take a picture of the nursing area. The shooting interface will prompt the nurse to place a standard scale reference on the same plane as the patient's nursing area.

[0036] When entering supplementary record data, nurses can supplement the record in one or more of the following ways on the same or related screen as the captured image: Voice input: Press and hold the record button and verbally describe the observation results, for example: "The wound area is about 4cm², dark red in color, with a large amount of yellow exudate." Structured form completion: Select or fill in items in the standardized nursing assessment form provided in the program interface, such as vital signs, dressing type, nursing measures, etc.

[0037] Free text notes: In the text box next to the image preview area, enter your operational experience, special findings, or follow-up precautions for this nursing care.

[0038] Step S3: Verify the validity of the standard scale reference in the captured nursing images; like Figure 2 As shown, step S3 specifically includes: S31. Identify whether a standard scale reference object exists in nursing images using a target detection model; S32. Verify whether the integrity, clarity, and position of the standard scale reference object in the nursing image conform to the preset shooting specifications.

[0039] After the nursing images in step S2 are uploaded, the system first calls the YOLO or Faster R-CNN object detection model to quickly identify whether the image contains a standard scale reference object that meets the requirements, and verifies its integrity, clarity and relative position in the picture.

[0040] The purpose of this step is to ensure that the nursing images meet the basic requirements for subsequent accurate quantification. If verification fails, the system prompts the nurse to retake the image; if verification succeeds, the nursing image is marked as "valid" and proceeds to the next step.

[0041] Step S4: Input the verified nursing images into the multi-label classification model to obtain at least one AI label for the nursing images; Specifically, the nursing images that have passed the verification in step S3 are input into a pre-trained multi-label image classification model, which outputs one or more AI labels. This step completes the qualitative identification of nursing problems.

[0042] Among them, the multi-label image classification model preferably uses ResNet or VisionTransformer pre-trained on ImageNet as the base model, performs transfer learning architecture, and is fine-tuned using a large number of desensitized clinical wound images and expert-annotated labels.

[0043] After a nurse uploads a new nursing image, the system calls the trained multi-label image classification model to perform forward inference. The model outputs the probability that the image belongs to each label, and takes the top few labels with the highest probabilities as the automatic labeling results. For example, location: right hand; type: indwelling needle.

[0044] Understandably, the labeling system can be multi-level categorized. For example, the primary label could be the location: sacrococcygeal region; and the secondary label could be the type: pressure ulcer / stage III.

[0045] Step S5: Extract structured information from the supplementary record data to obtain structured nursing text; like Figure 3 As shown, step S5 specifically includes: S51. Input the voice recording data from the supplementary recording data into the speech recognition model and convert it into initial text; S52. Use a medical named entity recognition model to extract key medical entities and numerical information from the initial text; S53. Fill the extracted information into the corresponding fields of the predefined nursing record structure template to form structured nursing text.

[0046] In practical applications, the speech recording data in the supplementary recording data of step S2 is converted into text using a medical-adaptive speech recognition model; the speech recognition model undergoes acoustic model adaptation and language model adaptation to achieve the purpose of medical-adaptation.

[0047] Acoustic model adaptation: Based on a general Chinese speech model, fine-tuning is performed using a large amount of medical scenario speech data spoken by medical staff to adapt to clinical environmental noise and the pronunciation of medical professional terms.

[0048] Language model adaptation: The language model is trained using text corpora such as medical documents and nursing records, which greatly improves the recognition accuracy of medical terms such as "sacral and coccygeal region", "drainage fluid", and "piperacillin".

[0049] Real-time processing: The front end collects audio streams and transmits them to the back end speech recognition service in real time via the WebSocket protocol. The server performs streaming recognition and returns the results to the front end for display in real time, realizing "speaking and converting simultaneously".

[0050] Then, using a named entity recognition model fine-tuned based on medical BERT, key medical entities are extracted, such as "size: 4". 3cm, Exudate: Medium, Color: Yellow, etc. Fill these entities into the structured nursing template to generate structured nursing text.

[0051] Step S6: Link and store the event ID, nursing image, AI tag and structured nursing text, and display them in chronological order in the timeline view of the target patient; Specifically, the backend server associates and stores all data, including event IDs, timestamps, nursing images, AI tags, and structured nursing text. On the patient details page on the mobile device, the system visualizes all nursing events in chronological order in a timeline format. Clicking on any event card allows you to view all associated multimodal records for that event.

[0052] Step S7: Generate a comparison report request for the target patient and obtain nursing images and associated structured nursing texts from at least two different time points; When it is necessary to assess the effectiveness of nursing care, nurses manually select nursing event records from two different time points in the timeline view and trigger the "Generate Comparison Report" command.

[0053] Step S8: Analyze the nursing images obtained at least two different time points to generate visual comparison analysis results of changes in the nursing site; like Figure 4 As shown, step S8 specifically includes: S81. Use an image segmentation model to segment the standard scale reference area and the target nursing area in each nursing image; S82. Based on the pixel size of the segmented standard scale reference area and its known actual physical size, calculate the scaling factor of the current nursing image; S83. Calculate the pixel area covered by the pixel mask of the target care area; S84. Multiply the pixel area by the square of the scaling factor to obtain the actual physical area of ​​the target care area; S85. Calculate the actual physical area change rate of nursing images at two different time points; S86. Extract the color and texture features of the target nursing area from each nursing image; S87. Calculate the color feature distance and texture feature distance between target care areas at different time points; S88. Generate visual contrast analysis results based on the actual physical area change rate, color feature distance, and texture feature distance.

[0054] More specifically, the image segmentation model is preferably implemented using a U-Net network based on an encoder-decoder structure, which is specifically trained to simultaneously segment multiple targets in medical images, such as standard-scale reference objects, wound areas, and healthy skin. The encoder is responsible for extracting and compressing multi-level features from the input image. The encoder is usually composed of a backbone network of a pre-trained deep convolutional neural network such as ResNet-50 or VGG16. Through a series of convolutional and pooling layers, it gradually transforms the high-resolution image into a low-resolution feature map containing rich semantic information.

[0055] The decoder is responsible for progressively upsampling the deep, abstract feature maps extracted by the encoder to restore them to the original image resolution, thus achieving pixel-level classification. The decoder performs upsampling through deconvolution or interpolation.

[0056] The U-Net network can concatenate high-resolution feature maps containing more spatial details from each layer of the encoder with upsampled feature maps from the middle layer of the decoder. This structure effectively integrates deep semantic information and shallow contour information, making the model more accurate in locating target boundaries.

[0057] Calculation of actual physical area change rate: In practical applications, two selected contrasting nursing images are first input into the image segmentation model. After training, the model can simultaneously and accurately segment two key regions in the image: the standard-scale reference region and the target nursing region. The output is a pixel-level mask for both regions.

[0058] Because the shooting angle may not be directly facing the subject, perspective distortion may occur in the reference object. The system will use the four corner points of the detected reference object to perform perspective transformation, correcting the reference object area and the target nursing area in the nursing image to a standard front view, thereby eliminating measurement errors caused by angle.

[0059] For each nursing image, the pixel size of the reference object in the image is calculated based on the segmented reference object region mask. Combined with the known actual physical size of the reference object, a pixel-to-physical-size ratio coefficient specific to that nursing image is calculated. Scale factor = Actual physical size of reference object / Pixel size of reference object; Using the scaling factor calculated in the previous step, the mask of the segmented target care region in the same image is converted, the pixel area of ​​the region is calculated, and then the actual physical area of ​​the target care region in the care image is calculated based on the pixel area: Actual physical area = pixel area × (scale factor)²; Using the actual physical area of ​​the target care area calculated in the previous step, calculate the rate of change of area between the two comparative care images.

[0060] Color feature evolution analysis: The target treatment area of ​​two contrasting treatment images is converted from RGB color space to HSV color space, its color histogram is extracted, and the Bartholomew's distance between the color histograms of the target treatment area and the healthy skin area is calculated. By comparing the Bartholomew's distance of treatment images at different time points, the degree of color changes such as redness, necrosis, and granulation tissue growth can be quantified.

[0061] Texture feature evolution analysis: The Local Binary Pattern (LBP) algorithm was used to extract texture features from the target care area, generating LBP feature histograms. By calculating the chi-square distance of the LBP histograms at different time points, the texture changes of the wound surface from erosion to smoothness, or from wetness to dryness, were quantified.

[0062] Step S9: Based on the visual comparison analysis results and the structured nursing text at the corresponding time points, generate a nursing comparison report combining text and images.

[0063] This step is achieved through an integrated intelligent report generation engine. The core logic of this engine includes rule matching, template filling, and natural language generation. The specific process is as follows: The system pre-configures a report template library containing various clinical scenarios and an associated decision rule library. Each rule consists of a "condition" and an "output description template." For example: Rule 1: IF area change rate < -10% THEN area description template = "wound area significantly reduced by approximately {abs(change rate)}%".

[0064] Rule 2: IF color feature distance decreases > 0.2 THEN color description template = "wound color significantly improved".

[0065] Rule 3: IF the exudate text entity changes from "[large amount]" to "[small amount]" THEN the exudate description template = "exudate significantly reduced".

[0066] The report generation engine receives all the quantitative data output from step S8 and the key descriptive entities extracted from the structured care texts at two time points. It then matches the data with the conditions in the rule base and filters out all triggered rules.

[0067] The report generation engine employs template-based natural language generation technology, assembling the output description templates corresponding to the triggering rules and injecting specific quantitative values ​​or textual descriptions into the corresponding placeholders in the templates. Simultaneously, the system integrates key vital signs or nursing observation summaries from two time points into the report context, such as "body temperature dropped from 38.5°C to 36.8°C".

[0068] Finally, the report generation engine combines the generated comprehensive natural language description paragraphs, comparative quantitative indicators, and side-by-side or overlapping comparative images into a structured "Dynamic Comparison Report of Nursing Effects" graphic template, generating the final structured nursing comparison report.

[0069] It should be noted that the system in this embodiment will dynamically determine the range of patient data that the logged-in user can access and the functional modules that can be operated based on the user's role information, and perform dual verification on the backend API interface and the frontend interface; all transmitted data is encrypted and stored data is desensitized, in accordance with medical information security standards.

[0070] Role information includes at least nurses, head nurses, and doctors, and nurses can only access the data of the patients they are responsible for.

[0071] The specific encryption method is as follows: Encrypted transmission: All front-end and back-end communications are encrypted throughout using the HTTPS protocol (TLS 1.2 / 1.3).

[0072] Database encryption: Sensitive structured data such as patient names and hospital numbers are encrypted at the field level within the database.

[0073] File encryption: Images, audio files, and other files stored in object storage are encrypted using server-side encryption (SSE-S3), with the key managed by the cloud service provider to ensure the security of static data.

[0074] Example 2: like Figure 5 As shown, this embodiment provides a multifunctional electronic medical record generation system for clinical nursing, including: Record creation module 1 is used to generate and store an event ID that uniquely corresponds to the nursing event in response to the nursing record creation request of the target patient. This module is deployed on the backend server and is triggered when a nurse selects a patient and clicks "Create Record" on the mobile terminal. Its core function is to call a distributed ID generator to generate a globally unique event ID and record the precise creation timestamp. This module returns the event ID to the mobile terminal as the root identifier for all data in this instance.

[0075] Multimodal data acquisition module 2 is used to acquire multimodal nursing data associated with event IDs; the multimodal nursing data includes at least nursing images containing standard scale references and supplementary record data associated with the nursing images; The reference object verification module 3 specifically includes a detection unit and a verification unit. The detection unit uses a target detection model to quickly identify whether a standard-scale reference object exists in the image. The verification unit performs logical verification on the detected standard-scale reference objects, including their completeness, clarity, and whether their relative position in the image is reasonable (e.g., whether they are too far to the edge).

[0076] Image classification module 4 is used to input nursing images into the AI ​​analysis model to obtain at least one AI label for the nursing image; the AI ​​label includes at least the nursing site, the type of nursing problem, and the severity level; The text structuring module 5 specifically includes a speech recognition unit, an information extraction unit, and a template filling unit. The speech recognition unit integrates a domain-adaptive speech recognition service, inputting the speech recording data from the supplementary recording data into the speech recognition model and converting it into initial text. The information extraction unit uses a medical named entity recognition model to extract key medical entities and numerical information from the initial text. The template filling unit fills the extracted information into the corresponding fields of a predefined nursing record structuring template to form structured nursing text.

[0077] The storage and display module 6 is used to associate and store event IDs, nursing images, AI tags and structured nursing texts, and display them in chronological order in the timeline view of the target patient; The comparison analysis module 7, in response to the comparison report generation request of the target patient, acquires nursing images and associated structured nursing texts at at least two different time points, analyzes the acquired nursing images, and generates visual comparison analysis results of changes in nursing sites. The comparative analysis module 7 specifically includes an image segmentation unit and an image calculation unit. The image segmentation unit is used to segment the standard scale reference area and the target nursing area in each nursing image. The image calculation unit is used to calculate the actual physical area change rate, color feature distance and texture feature distance of the nursing images at two different time points.

[0078] The report generation module 8 is used to generate a nursing comparison report combining text and images based on the results of visual comparison analysis and the structured nursing text at the corresponding time points.

[0079] Example 3: This embodiment provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. The device is characterized in that the processor executes the computer program to implement a multifunctional electronic medical record generation method for clinical nursing procedures.

[0080] In some embodiments, the processor may be composed of integrated circuits, such as a single packaged integrated circuit or multiple integrated circuits with the same or different functions, including combinations of one or more central processing units (CPUs), microprocessors, digital processing chips, graphics processors, and various control chips. The processor is the control core of the electronic device in this embodiment, connecting various components of the device through various interfaces and lines. It executes computer programs stored in memory, such as executing a multi-functional electronic medical record generation program, and calls data stored in memory to perform various functions of the electronic device and process data.

[0081] The memory includes at least one type of readable storage medium, including flash memory, portable hard drives, multimedia cards, card-type memory (e.g., SD or DX memory), magnetic storage, magnetic disks, optical disks, etc. In some embodiments, the memory can be an internal storage unit of the device, such as the device's portable hard drive. In other embodiments, the memory can be an external storage device of the device, such as a plug-in portable hard drive, smart memory card, secure digital card, flash memory card, etc. Furthermore, the memory can include both internal and external storage units of the device. The memory can be used not only to store application software and various types of data installed on the device, such as the code of a multi-functional electronic medical record generation program, but also to temporarily store data that has been output or will be output.

[0082] The electronic device may also include a power supply for powering the various components. Preferably, the power supply can be logically connected to the at least one processor via a power management device, thereby enabling functions such as charging management, discharging management, and power consumption management. The power supply may also include one or more DC or AC power sources, a recharging device, a power fault detection circuit, a power converter or inverter, a power status indicator, and other arbitrary components. The electronic device may also include various sensors, Bluetooth modules, Wi-Fi modules, etc., which will not be elaborated further here.

[0083] It should be understood that the embodiments described are for illustrative purposes only and are not limited to this structure in the scope of the patent application.

[0084] In the several embodiments provided by this invention, it should be understood that the disclosed devices, apparatuses, and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of modules is only a logical functional division, and other division methods may be used in actual implementation.

[0085] The modules described as separate components may or may not be physically separate. The components shown as modules may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs.

[0086] Furthermore, the functional modules in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or in the form of hardware plus software functional modules.

[0087] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the present invention can be implemented in other specific forms without departing from the spirit or essential characteristics of the present invention.

[0088] Therefore, the embodiments should be considered exemplary and non-limiting in all respects, and the scope of the invention is defined by the appended claims rather than the foregoing description. Thus, all variations falling within the meaning and scope of equivalents of the claims are intended to be embraced within the invention. No appended diagram markings in the claims should be construed as limiting the scope of the claims.

[0089] The embodiments of this application can acquire and process relevant data based on artificial intelligence technology. Artificial intelligence is the theory, method, technology, and application system that uses digital computers or machines controlled by digital computers to simulate, extend, and expand human intelligence, perceive the environment, acquire knowledge, and use that knowledge to obtain optimal results.

[0090] Furthermore, it is clear that the word "comprising" does not exclude other units or steps, and the singular does not exclude the plural. Multiple units or devices recited in the system claims may also be implemented by a single unit or device through software or hardware. The terms "first," "second," etc., are used to indicate names and do not indicate any specific order.

[0091] It should be stated that the above-described specific embodiments are merely preferred embodiments of the present invention and the technical principles employed. Those skilled in the art should understand that various modifications, equivalent substitutions, and variations can be made to the present invention. However, such variations, as long as they do not depart from the spirit of the present invention, should be within the scope of protection of the present invention. Furthermore, some terminology used in this specification and claims is not limiting, but merely for ease of description.

Claims

1. A method for generating multifunctional electronic medical records for clinical nursing, characterized in that, Includes the following steps: S1. Create a nursing record creation request for the target patient, generate and store an event ID that uniquely corresponds to this nursing event; S2. Collect multimodal nursing data associated with the event ID; The multimodal nursing data includes at least nursing images containing standard scale references and supplementary record data associated with the nursing images; S3. Verify the validity of the standard scale reference in the nursing images; S4. Input the verified nursing image into a multi-label classification model to obtain at least one AI label for the nursing image; S5. Extract structured information from the supplementary record data to obtain structured nursing text; S6. The event ID, nursing image, AI tag and structured nursing text are associated and stored, and displayed in chronological order in the timeline view of the target patient; S7. For the target patient, generate a comparison report request and obtain nursing images and associated structured nursing texts from at least two different time points; S8. Analyze the nursing images acquired at least two different time points to generate visual comparison analysis results of changes in the nursing site; S9. Based on the visual comparison analysis results and the structured nursing text at the corresponding time points, generate a nursing comparison report combining text and images.

2. The method for generating multifunctional electronic medical records for clinical nursing according to claim 1, characterized in that, Step S3 specifically includes: S31. Identify whether a standard scale reference object exists in the nursing image using a target detection model; S32. Verify whether the integrity, clarity, and position of the standard scale reference object in the nursing image conform to the preset shooting specifications.

3. The method for generating multifunctional electronic medical records for clinical nursing according to claim 1, characterized in that, The supplementary recorded data includes at least voice recording data, and step S5 specifically includes: S51. Input the voice recording data in the supplementary recording data into the voice recognition model and convert it into initial text; S52. Use a medical named entity recognition model to extract key medical entities and numerical information from the initial text; S53. Fill the extracted information into the corresponding fields of the predefined nursing record structure template to form structured nursing text.

4. The method for generating multifunctional electronic medical records for clinical nursing according to claim 1, characterized in that, Step S8 specifically includes: S81. Use an image segmentation model to segment the standard scale reference area and the target nursing area in each nursing image; S82. Based on the pixel size of the segmented standard scale reference area and its known actual physical size, calculate the scaling factor of the current nursing image; S83. Calculate the pixel area covered by the pixel mask of the target care area; S84. Multiply the pixel area by the square of the scaling factor to obtain the actual physical area of ​​the target care area; S85. Calculate the actual physical area change rate of nursing images at two different time points; S86. Extract the color and texture features of the target nursing area from each nursing image; S87. Calculate the color feature distance and texture feature distance between target care areas at different time points; S88. Generate visual contrast analysis results based on the actual physical area change rate, color feature distance, and texture feature distance.

5. The method for generating multifunctional electronic medical records for clinical nursing according to claim 1, characterized in that, It also includes access control: Based on the logged-in user's role information, the scope of patient data that the user can access and the functional modules that can be operated are dynamically determined; the role information includes at least nurses, head nurses, and doctors, and nurses can only access the patient data they are responsible for.

6. A multifunctional electronic medical record generation system for clinical nursing, characterized in that, include: The record creation module is used to generate and store an event ID that uniquely corresponds to the nursing event in response to the nursing record creation request of the target patient. A multimodal data acquisition module is used to acquire multimodal nursing data associated with the event ID; The multimodal nursing data includes at least nursing images containing standard scale references and supplementary record data associated with the nursing images; The reference object verification module is used to verify the validity of standard scale reference objects in the nursing images; An image classification module is used to input the nursing image into an AI analysis model to obtain at least one AI label for the nursing image; the AI ​​label includes at least the nursing site, the type of nursing problem, and the severity level. The text structuring module is used to extract structured information from the supplementary record data to obtain structured nursing text; The storage and display module is used to associate and store the event ID, nursing images, AI tags and structured nursing text, and display them in chronological order in the timeline view of the target patient; The comparison analysis module is used to generate comparison reports for target patients by acquiring nursing images and associated structured nursing text at at least two different time points, analyzing the acquired nursing images, and generating visual comparison analysis results of changes in nursing sites. The report generation module is used to generate a nursing comparison report combining text and images based on the visual comparison analysis results and the structured nursing text at the corresponding time points.

7. The multifunctional electronic medical record generation system for clinical nursing according to claim 6, characterized in that, The reference object verification module specifically includes: The detection unit is used to identify whether a standard scale reference object exists in the nursing image through a target detection model; The verification unit is used to verify whether the integrity, clarity, and position of the standard scale reference object in the nursing image conform to the preset shooting specifications.

8. The multifunctional electronic medical record generation system for clinical nursing according to claim 6, characterized in that, The text structuring module specifically includes: The speech recognition unit is used to input the speech recording data from the supplementary recording data into the speech recognition model and convert it into initial text. An information extraction unit is used to extract key medical entities and numerical information from the initial text using a medical named entity recognition model. The template filling unit is used to fill the extracted information into the corresponding fields of a predefined nursing record structure template to form structured nursing text.

9. The multifunctional electronic medical record generation system for clinical nursing according to claim 6, characterized in that, The comparative analysis module specifically includes: The image segmentation unit is used to segment the standard scale reference area and the target nursing area in each nursing image; The image calculation unit is used to calculate the actual physical area change rate, color feature distance, and texture feature distance of nursing images at two different time points.

10. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the multifunctional electronic medical record generation method for clinical nursing as described in any one of claims 1-5.

Images