Auxiliary device for nursing operation
By designing nursing work assistive devices, using voice analysis and medical scenario database technology, the complex and inconvenient nursing work are solved, and more efficient and accurate nursing operations and recording are achieved.
Patent Information
- Application Number
- CN202510274475.8
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-03-10
- Publication Date
- 2025-06-13
AI Technical Summary
Nursing work is complicated and inconvenient to record, resulting in missed operational items or delayed recording, especially in critically illegitimate departments. The risk of information loss is high.
A nursing job assist device is designed, including a first processing module for analyzing the original voice information, a second processing module for analyzing external voice data based on a medical scene database, and a third processing module for processing nursing job prompt information based on nursing job object data and nursing behavior data.
By automatically obtaining and analyzing voice data, it reduces the difficulty of data acquisition, improves the efficiency of nursing operations, ensures operational integrity and record accuracy, and reduces the risk of information loss.
Smart Images
Figure CN120148516A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of nursing appliances, and particularly to a nursing operation assistance device. Background Art
[0002] Nursing work is highly dynamic and complex. Especially in critical care departments, nurses not only need to complete basic nursing operations, but also quickly respond to emergencies, perform multi-step first aid operations, and record the changes in the patient's condition in real time. In this scenario, there are significant pain points in the traditional work mode and manual recording mode: the multi-item and high-load work is prone to omission of operation items (such as drug dosage, treatment time node) or delayed and inaccurate recording, and the time-sensitive recording during rescue magnifies the risk of information loss. Summary of the Invention
[0003] The present invention provides a nursing operation assistance device to solve the problems of complex operations and inconvenient recording in existing nursing work.
[0004] To achieve the above object, the technical solution adopted by the present invention is:
[0005] A nursing operation assistance device, comprising:
[0006] A first processing module, configured to parse the original voice information to obtain external voice data;
[0007] A second processing module, configured to parse the external voice data based on a medical scenario database to obtain nursing behavior data;
[0008] A third processing module, configured to process and obtain nursing operation prompt information for output according to nursing operation object data and the nursing behavior data.
[0009] Further, the nursing operation object data includes the nursing task type and task arrangement information of the cared object; correspondingly, the third processing module is configured to determine whether the nursing behavior data matches according to the nursing task type and the task arrangement information, and obtain the corresponding nursing operation prompt information according to the matching result.
[0010] Further, obtaining the corresponding nursing operation prompt information according to the matching result includes: when the matching result is a match, outputting a corresponding prompt voice according to the nursing task type; or recording the executed nursing behavior data of the nursing executor through the first processing module and the second processing module; when the matching result corresponding to the nursing task type is a mismatch, outputting nursing task information related to the nursing task type; when the matching result corresponding to the task arrangement information is a mismatch, outputting nursing task information with a higher priority based on the corresponding clinical guidelines.
[0011] Further, the medical scenario database includes a special vocabulary library and a basic behavior vocabulary library. Correspondingly, parsing the external voice data based on the medical scenario database to obtain nursing behavior data includes determining relevant voice characters and corresponding character sorting in the external voice data by character matching to obtain a character combination; and based on the character combination, matching and obtaining the corresponding nursing behavior data from the special vocabulary library and the basic behavior vocabulary library.
[0012] Further, the medical scenario database further includes an environmental sound library. Correspondingly, the first processing module is further configured to exclude environmental noise in the original voice information according to the noise type and noise data characteristics in the environmental sound library.
[0013] Further, the medical scenario database further includes a dialect library. Correspondingly, the first processing module is configured to parse the original voice information according to the dialect library and convert dialect characters and / or dialect character voices into the external voice data that conforms to the specified language rules.
[0014] Further, recording the executed nursing behavior data of the nursing executor through the first processing module and the second processing module includes: the first processing module obtains the original voice information and identifies the nursing executor's behavior voice according to the voice data characteristics of the nursing executor; the second processing module parses the nursing executor's behavior voice based on the medical scenario database to obtain the executed nursing behavior data.
[0015] Further, it further includes a fourth processing module for customizing call keywords, instruction keywords, and custom exclusive voices.
[0016] Further, when the second processing module recognizes the call keyword, it converts continuous speech into event node data based on the time axis compression algorithm; the third processing module is further configured to output the nursing operation prompt information according to the instruction keyword and output a record log according to a preset time period.
[0017] Further, based on the AI model, the functions of at least one of the first processing module, the second processing module, the third processing module, and the fourth processing module are realized.
[0018] Due to the adoption of the above technical solutions, the present invention has the following beneficial effects:
[0019] 1. By analyzing the original voice information, the present invention obtains external voice data, which can automatically acquire external data for subsequent analysis and processing, reducing the difficulty of data acquisition to improve the efficiency of subsequent implementation of nursing-related auxiliary functions; by analyzing the external voice data based on the medical scenario database to obtain nursing behavior data, the data of nursing behavior can be obtained by means of voice data, which can liberate the hands and feet of operators and improve the efficiency of nursing operations; by processing the nursing operation object data and the nursing behavior data to obtain nursing operation prompt information for output, it can provide a data basis for subsequent execution of nursing-related auxiliary functions.
[0020] 2. Based on the matching results of the nursing operation object data and the nursing behavior data, the present invention executes different nursing auxiliary functions / outputs corresponding nursing operation prompt information, which can adapt to different nursing auxiliary requirements.
[0021] 3. Based on the special vocabulary, basic behavior vocabulary, environmental sound library, and dialect library, the present invention can adapt to and overcome the problems of large differences in the voice characteristics of relevant personnel and complex working environments in practice. BRIEF DESCRIPTION OF THE DRAWINGS
[0022] Figure 1 It is a schematic diagram of the nursing operation auxiliary device proposed by the present invention;
[0023] Figure 2 It is the process of a nurse using AI-assisted nursing proposed by the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS
[0024] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.
[0025] The nursing operation auxiliary device includes:
[0026] The first processing module 1 is used to analyze the original voice information to obtain external voice data;
[0027] The second processing module 2 is used to analyze the external voice data based on the medical scenario database to obtain nursing behavior data;
[0028] The third processing module 3 is used to process the nursing operation object data and the nursing behavior data to obtain nursing operation prompt information for output.
[0029] The nursing operation assistance device includes a housing, internal software and hardware, and other components / structures. This device or equipment can be integrated into a small electronic instrument for easy portability by nursing staff. For example, it can be placed in the upper pocket of a white coat, clipped on the arm, or hung on the waist. Among them, the internal software and hardware include a first processing module for parsing the original voice information to obtain external voice data. Its purpose is to meet the needs of nursing work / operations. Nursing operations are related to the health of the cared-for person. Therefore, in actual operations, there are many institutional regulations and requirements. Implementing these regulations and requirements will consume a large amount of energy (manual and mental) of nursing staff, leaving them no time or mental capacity to record or do other work. And a person's voice ability is relatively idle during nursing activities, and nursing staff will develop certain conditioned reflexes during long-term work and training, that is, when performing certain work processes, they will simultaneously read out the corresponding content to convey information to the cared-for person or other colleagues working together. The first processing module obtains the voice around the device, that is, the original voice information, through devices such as microphones, and converts the original voice information into data that can be stored and processed, that is, external voice data.
[0030] Nursing is part of medical behavior and there are many specialized terms. By identifying specialized terms, it can be distinguished from general conversations, and the useful part of the external voice data, that is, the part related to nursing operations, can be separately extracted. Reducing the amount of data to be processed can reduce the subsequent data processing pressure and improve data processing efficiency. After all, the device cannot be made too large and there are certain limitations on its functional specifications. The data representing the part of the behavior related to nursing operations extracted is the nursing behavior data. The internal software and hardware include a second processing module for obtaining nursing behavior data.
[0031] Since the objects of nursing are different, the items / operations that need to be carried out for nursing are also different. If incorrect nursing is carried out, it may even lead to the aggravation of the condition. Therefore, it is necessary to strictly ensure that there are no errors between nursing operations and the cared-for objects. Specifically, the nursing operation object data can include the identity information (name, age) of the cared-for object, physical data (height, weight), health data (measurement data within a certain range), condition data (suffering from what disease, showing what data), items that need to be nursed, and nursing parameters, etc. By combining the nursing operation object data and the nursing behavior data, nursing operation prompt information is processed. Its principle is to select or generate relevant nursing operation prompt information by detecting the matching degree between the nursing operation object data and the nursing behavior data. The nursing operation prompt information can include data records and voice prompt information. Among them, the data record can play roles such as condition analysis and file storage, and the voice prompt information can prompt what the nursing staff has done (main function) or what should be done (secondary function).
[0032] The present invention can record the content instructed by the nursing staff, remind the nursing staff to perform special nursing care on time, significantly improve the efficiency and safety of clinical nursing work event records, reduce omissions of the nursing staff in their busy daily work, enable patients to be evaluated in a timely manner, and reduce the writing omissions of electronic nursing records.
[0033] The nursing operation object data includes the nursing task type and task arrangement information of the object to be nursed; correspondingly, the third processing module is used to determine whether the nursing behavior data matches according to the nursing task type and the task arrangement information, and obtain the corresponding nursing operation prompt information according to the matching result.
[0034] There are many types of nursing tasks, such as: patient condition observation, intravenous infusion, aseptic technique operation, donning and doffing isolation gowns, suctioning, preparing skin test solutions, performing cardiopulmonary resuscitation, managing drainage tubes, nasogastric intubation, catheterization, single-person and double-person handling techniques, gastrointestinal decompression, intramuscular and intradermal injections, subcutaneous injections, oral care, oxygen inhalation, and nebulization inhalation, etc. For different objects to be nursed, the suitable types of nursing tasks are different, and the time for performing nursing, that is, the task arrangement information, is also different. It is necessary to strictly and accurately determine whether the nursing behavior data matches, and obtain the corresponding nursing operation prompt information according to the matching result, which can effectively ensure the smooth completion of nursing work and reduce the probability of medical accidents.
[0035] Obtaining the corresponding nursing operation prompt information according to the matching result includes: when the matching result is a match, outputting the corresponding prompt voice according to the nursing task type; or, recording the executed nursing behavior data of the nursing executor through the first processing module and the second processing module; when the matching result corresponding to the nursing task type is a mismatch, outputting the nursing task information related to the nursing task type; when the matching result corresponding to the task arrangement information is a mismatch, outputting the nursing task information with a higher priority based on the corresponding clinical guidelines.
[0036] Specifically, the matching principle includes that the nursing operation assistance device can identify the current object being cared for. The specific method can be to set up a camera for appearance recognition, or to identify the tag of the object being cared for through radio frequency signals, or for the caregiver to directly input the ID. After identifying the object being cared for, the data of the object being cared for is loaded from the hospital's database, and the nursing task type and task arrangement information can be determined. Among them, if the matching result is a match, then according to the nursing task type, the corresponding prompt voice is output. For example, patient A needs to prepare skin test solution at 11:30. Please pay attention to the reagent type and dosage. Or, if the match is okay and the caregiver does not need to be prompted to do anything but needs to record the nursing process, then the caregiver can speak out the nursing operations being carried out while performing the nursing, and then this device can collect these oral voices and analyze and convert their contents into electronic data for subsequent viewing. If there is no match, it means that there is a certain conflict in the current nursing operation of the caregiver. For example, originally scheduled to measure the blood sugar of patient B at 12:00, on the way there, the nurse reported that patient C suddenly said that he felt a little uncomfortable and needed to be evaluated at 12:00. At this time, there are two task arrangements with a time conflict. Based on the corresponding clinical guidelines, the nursing task information with a higher priority is output. Specifically, please first evaluate the physical condition of patient C, and then perform the blood sugar measurement for patient B.
[0037] Priority sorting rules:
[0038] 1. Direct threat to life > Prevention of potential complications > Routine care;
[0039] 2. Invasive procedures > Non-invasive procedures;
[0040] 3. Time-sensitive tasks > Tasks with time flexibility;
[0041] 4. Operations with high regulatory requirements > General nursing operations.
[0042] The medical scenario database includes a special vocabulary library and a basic behavior vocabulary library. Correspondingly, parsing the external voice data based on the medical scenario database to obtain nursing behavior data includes determining the relevant voice characters and the corresponding character sorting in the external voice data through character matching to obtain a character combination; based on the character combination, matching and obtaining the corresponding nursing behavior data from the special vocabulary library and the basic behavior vocabulary library.
[0043] Speech recognition is a relatively commonly used technology. In this solution, a medical scenario database is mainly set up additionally to improve the accuracy and professionalism of recognition. The specific recognition principle includes recognizing speech to obtain speech characters, and the speech characters and combinations will conform to certain rules. There are a special vocabulary library (i.e., a thesaurus of specialized terms related to medical care, including specialized terms and corresponding speech features, such as intradermal test - intradermal injection, intravenous push - intravenous injection, intramuscular injection - intramuscular injection, peritoneal dialysis - peritoneal dialysis, hemodialysis - hemodialysis, etc.) and a basic behavior vocabulary library (i.e., a general behavior thesaurus, such as nasal feeding, bolus injection, assessment, verification, monitoring, walk, take, do, use how many milliliters, etc.). By presetting some nursing behavior data and binding them to the corresponding character combinations, the corresponding nursing behavior data can be output.
[0044] The medical scenario database further includes an environmental sound library. Correspondingly, the first processing module is further configured to exclude environmental noise in the original speech information according to the noise type and noise data characteristics in the environmental sound library.
[0045] Hospitals are equipped with different devices that emit various sounds. People coming and going in the hospital also generate different sounds, and hospitals often conduct broadcasts. These sounds will interfere with the processing of the original speech information. Fortunately, the hospital environment is relatively stable, and the types of possible noises are fixed. By setting up an environmental sound library, the environmental noise in the original speech information can be excluded.
[0046] The medical scenario database further includes a dialect library. Correspondingly, the first processing module is configured to parse the original speech information according to the dialect library and convert dialect characters and / or dialect character voices into the external speech data that conforms to the specified language rules.
[0047] Although Mandarin is the common language, in practice, due to regional and educational level differences, dialects will be generated. Different dialects will affect nursing work and increase the communication difficulty. However, by widely collecting dialect data and classifying them, dialects can still be converted into Mandarin, which can facilitate the progress of nursing work. Foreign languages can be translated in real time, and it is also technically possible to convert dialects into Mandarin, and even convert Mandarin into medical phrases to improve the information transmission efficiency.
[0048] Through the first processing module and the second processing module, the executed nursing behavior data of the nursing executor is recorded, including: the first processing module obtains the original speech information and recognizes the nursing executor's behavior speech according to the speech data characteristics of the nursing executor; the second processing module analyzes the nursing executor's behavior speech based on the medical scenario database to obtain the executed nursing behavior data.
[0049] Medical and nursing behaviors fall within the scope of modern science. Therefore, in order to prevent accidents caused by misoperations or as a basis for subsequent observations, it is necessary to record the actions of nursing staff. At this time, by identifying the voice data characteristics of the nursing executor (nursing staff), personal information and corresponding behavior information can be identified, and the executed nursing behavior data can be obtained. Specifically, the voice recognition of a smart phone can be referred to.
[0050] Based on the AI model, the functions of at least one of the first processing module, the second processing module, the third processing module, and the fourth processing module are implemented.
[0051] The core innovation of the present invention lies in constructing an intelligent voice interaction system based on artificial intelligence technology, which significantly optimizes the clinical nursing work process. Its innovation is mainly reflected in the following dimensions: First, the present invention works completely through voice control, establishing a non-contact interaction mechanism, effectively avoiding the risk of iatrogenic cross-infection caused by traditional manual operations, and strictly conforming to the JCI international medical nursing standards; Second, exclusive human voice wake-up is adopted. While ensuring accurate triggering, dynamic storage allocation is innovatively adopted, and structured work logs are only generated in real time when the working state is activated. Compared with the existing technical solutions, most of the invalid data storage can be reduced, and the utilization rate of storage resources is significantly improved.
[0052] Specifically, the nursing operation assistance device is a 2.0-inch touch-screen electronic product, which is convenient for nursing staff to carry (can be placed in the upper pocket of the white coat).
[0053] AI model setting
[0054] 1.1 Wake up exclusive human voice recognition (can communicate with the user by voice)
[0055] 1.2 Nickname setting (Little Nurse Assistant)
[0056] 2. Medical scenario-based voice engine
[0057] 2.1 Noise suppression algorithm
[0058] 2.2 Term dynamic enhancement: Import the CMN medical term library (covering drug / operation / symptom terms)
[0059] 2.3 Dialect compatibility solution
[0060] 3. Intelligent task logic engine
[0061] 3.1 Conflict resolution mechanism: When the reminder times of multiple instructions overlap, based on clinical guidelines, the more important nursing tasks are reminded first.
[0062] 4. Rescue voice recording
[0063] 4.1 Voice recording of the whole process of rescue
[0064] 4.2 Timeline Compression Algorithm: Converting Continuous Voice Records into Event Nodes (Example):
[0065] 14:05:22 [Voice Keyword] Chest Compression
[0066] 14:07:15 [Voice Keyword] Defibrillator Discharge at 200J
[0067] 14:09:40 [Voice Keyword] Intravenous Injection of 1mg Epinephrine
[0068] 5. Daily Log Generation: It can generate all the voice content recorded today with the function of converting voice into text. The daily log can be saved for at least 3 days and will be automatically deleted after 3 days. Users can also set the saving duration according to their needs.
[0069] (Example):
[0070] 2025 / 2 / 27
[0071] 10:00 [Voice Content (Click to Play)]: Xiaohu (AI model nickname), record the intramuscular injection of tramadol for patient XXX in bed 1, and please remind me to evaluate the patient's condition after 30 minutes.
[0072] 10:30 Remind the nursing staff to evaluate the patient.
[0073] Nursing Staff: "Xiaohu (AI model nickname), I have received the reminder."
[0074] Reminder completed.
[0075] 11:00 [Voice Content (Click to Play)]: Xiaohu (AI model nickname), record the blood glucose value of 7.2 mmol / L for patient in bed 1 2 hours after breakfast.
[0076] ……
[0077] The nursing operation assistance device also includes a fourth processing module, which is used to customize call keywords, instruction keywords, and a customized exclusive voice, and can also link to the hospital management system to extract newly issued doctor's orders and remind the nursing staff to execute the orders (such as emergency blood tests, new treatments, postoperative medications, special medications, etc.).
[0078] Call keywords can be medical jargon or agreed-upon terms. When a call keyword appears, the device can execute the corresponding function. Instruction keywords can be function-specific terms or agreed-upon terms. When an instruction keyword appears, the device can execute the corresponding function (such as turning on or off). Customizing an exclusive voice means that the nursing operation assistance device outputs the voice that the nursing staff likes, which can improve the mood of the nursing staff.
[0079] The second processing module is further configured to, when the call keyword is recognized, convert continuous speech into event node data based on a timeline compression algorithm; the third processing module is further configured to output the nursing operation prompt information according to the instruction keyword, and is further configured to output a record log according to a preset time period.
[0080] After a nurse injects tramadol into a certain pain patient's muscle, the nurse issues an instruction: "Xiaohu (nickname of the AI model), record that patient XXX in bed 1 has received intramuscular injection of tramadol, and please remind me to re-evaluate the patient's pain situation in 30 minutes." AI model: "Okay, I have recorded it for you. I will remind you to evaluate the pain of patient XXX in bed 1 in 30 minutes." 30 minutes later, the AI model reminds the nurse to perform the work task, and the nurse replies that they have received the reminder. The AI model generates this voice record in the daily log.
[0081] When the on-duty nurse hands over the shift, they have already evaluated the basic conditions of the patients under their care, and then give several prompts (for example, after patient in bed E has completed project C, pay attention to whether situation F occurs, etc.). These prompts were previously set according to the nurse's judgment of the severity of the patient's condition and issued instructions. As an auxiliary reminder, the AI model can be like this: 15 minutes before blood transfusion, it is necessary to closely observe whether the patient has a blood transfusion reaction. At this time, there is also a prompt to monitor blood glucose (the blood transfusion observation prompt and the blood glucose monitoring prompt overlap in time). At this time, according to the conflict resolution mechanism and based on clinical guidelines, the more important nursing task is prioritized for reminder. Then, the priority is to remind to observe the blood transfusion patient.
[0082] In addition to completing the daily basic nursing, a certain nurse also needs to pay attention to the care of special patients, such as patients who have their blood glucose and blood pressure measured regularly. At this time, after learning the patient's breakfast time, the nurse can issue an instruction in advance: "Xiaohu (nickname of the AI model - call keyword), remind (instruction keyword) me to measure the blood pressure (instruction keyword) of patient in bed 5 and measure the blood glucose (instruction keyword) of patient in bed 6 at 10:30." AI model: "Okay, I have set the reminder for you." The AI model generates this reminder record in the daily log.
[0083] When the nurse discovers that the patient's condition has suddenly changed and starts the rescue together with the doctor, the nurse issues an instruction: "Xiaohu (nickname of the AI model), start the rescue of patient XXX in bed 1, and please open the rescue voice record." AI model: "Okay, I have opened the rescue voice record for you." The AI model generates this rescue voice record in the daily log.
[0084] For example, when the patient in bed 5 returns to the ward safely after surgery and the doctor has prescribed the postoperative medication for this patient, the AI model will give the first reminder to the responsible nurse after extracting this medical order: The postoperative medication for patient in bed 5 has been prescribed, please execute it in a timely manner. After the office nurse has passed this medical order, the AI model will give the second reminder: The office nurse has passed the medical order for the postoperative medication of patient in bed 5, please execute it in a timely manner.
[0085] As Figure 2 shown, the process of nursing staff using AI-assisted nursing includes:
[0086] The nursing staff wakes up the A model and issues an instruction;
[0087] The AI model receives the instruction and can then generate instruction reminder content or generate a voice record;
[0088] After generating the instruction reminder content, it can remind the nursing staff to perform the work task, and then the nursing staff replies to the AI model to complete the reminder;
[0089] After generating the instruction reminder content, it can also directly generate a daily log;
[0090] After generating the voice record, it can directly generate a daily log.
[0091] The above description is a detailed description of the preferred and feasible embodiments of the present invention, but the embodiments are not intended to limit the scope of the patent application of the present invention. Any equivalent changes or modifications made under the technical spirit disclosed by the present invention shall fall within the scope of the patent covered by the present invention.
Claims
1. A nursing work assisting device, characterized in that: include: The first processing module is used to analyze the original voice information and obtain external voice data; A second processing module is used to parse the external voice data based on the medical scenario database to obtain nursing behavior data; The third processing module is used to process the nursing task object data and the nursing behavior data to obtain nursing task prompt information for output.
2. The nursing work assisting device according to claim 1, characterized in that: The nursing task object data includes the nursing task type and task arrangement information of the nursing object; correspondingly, The third processing module is used to determine whether the nursing behavior data matches according to the nursing task type and the task arrangement information, and obtain the corresponding nursing task prompt information according to the matching result.
3. The nursing work assisting device according to claim 2, characterized in that: The obtaining of the corresponding nursing task prompt information according to the matching result includes: When the matching result is a match, a corresponding prompt voice is output according to the nursing task type; Alternatively, the first processing module and the second processing module are used to record the nursing behavior data performed by the nursing executor; When the matching result corresponding to the nursing task type is not a match, outputting nursing task information related to the nursing task type; When the matching result corresponding to the task arrangement information is not a match, the nursing task information with a higher priority is output based on the corresponding clinical guideline.
4. The nursing work assisting device according to claim 1, characterized in that: The medical scenario database includes a special term library and a basic behavior term library, correspondingly, The step of parsing the external voice data based on the medical scenario database to obtain nursing behavior data includes determining relevant voice characters and corresponding character sequences in the external voice data by character matching to obtain character combinations; Based on the character combination, the corresponding nursing behavior data is matched and obtained from the special term library and the basic behavior term library.
5. The nursing work assisting device according to claim 1, characterized in that: The medical scenario database also includes an environmental sound library, correspondingly, The first processing module is further used to eliminate the environmental noise in the original voice information according to the noise type and noise data characteristics in the environmental sound library.
6. The nursing work assisting device according to claim 1, characterized in that: The medical scenario database also includes a dialect database, correspondingly, The first processing module is used to parse the original voice information according to the dialect library, and convert the dialect characters and / or dialect character voices into the external voice data that conforms to the specified language rules.
7. The nursing work assisting device according to claim 3, characterized in that: The step of recording the nursing behavior data performed by the nursing executor through the first processing module and the second processing module includes: The first processing module acquires the original voice information, and obtains the behavior voice of the nursing executor according to the voice data feature recognition of the nursing executor; The second processing module analyzes the nursing practitioner's behavior voice based on a medical scenario database to obtain the executed nursing behavior data.
8. The nursing work assisting device according to claim 7, characterized in that: It also includes a fourth processing module for customizing call keywords, instruction keywords and customizing exclusive human voices.
9. The nursing work assisting device according to claim 8, characterized in that: The second processing module is further configured to convert the continuous speech into event node data based on a time axis compression algorithm when the call keyword is identified; The third processing module is further used to output the nursing operation prompt information according to the instruction keyword, and is also used to output the record log according to a preset time period.
10. The nursing work assisting device according to claim 9, characterized in that: Based on the AI model, the function of at least one of the first processing module, the second processing module, the third processing module and the fourth processing module is implemented.