A method of scheduling care tasks and related devices

Abstract

Embodiments of the present application disclose a nursing task scheduling method and related devices, which can include a nursing task scheduling apparatus, an electronic device, a computer program product, and a computer-readable storage medium. After receiving a nursing request for a to-be-nursed object, embodiments of the present application perform intent recognition on the to-be-nursed object to obtain nursing intent information and determine a current object state. According to the nursing intent information and the current object state, a current nursing task of the to-be-nursed object is generated, and a task priority of the current nursing task is determined. Object attribute information of at least one nursing object is obtained, and according to the task priority and the object attribute information, a target nursing object is selected from the at least one nursing object. The current nursing task and an object identifier of the target nursing object are sent to at least one nursing terminal, so that the nursing terminal prompts the target nursing object to perform the current nursing task. The scheme can improve the accuracy of nursing task scheduling.

Description

Technical Field

[0001] This invention relates to the field of artificial intelligence technology, specifically to a nursing task scheduling method and related equipment, which may include a nursing task scheduling device, electronic equipment, computer program products, and computer-readable storage media. Background Technology

[0002] In clinical nursing practice, nurses' perception of patient conditions and their decision-making in nursing care heavily rely on traditional shift handover reports, alarm information from vital sign monitoring instruments, and bedside observation during scheduled ward rounds. This traditional nursing model has significant limitations. Ward rounds are spaced out, and it's difficult to capture patients' immediate physiological changes and subjective discomfort outside of these rounds, leading to a lag in nursing response. Therefore, how to perceive patient conditions in real time and generate targeted nursing tasks accordingly to improve the timeliness and accuracy of nursing responses has become a pressing technical problem in this field. Summary of the Invention

[0003] This application provides a nursing task scheduling method and related equipment. The related equipment includes a nursing task scheduling device, an electronic device, and a computer-readable storage medium, which can improve the accuracy of nursing task scheduling.

[0004] A nursing task scheduling method, comprising: Receive a nursing request for a person to be cared for, the nursing request carrying object data of at least one modality of the person to be cared for; Based on object data, the intent of the object to be cared for is identified to obtain the nursing intent information of the object to be cared for and to determine the current object status of the object to be cared for. Based on the nursing intent information and the current state of the object, generate the current nursing task for the object to be cared for, and determine the task priority of the current nursing task; Obtain object attribute information for at least one nursing object, and filter out the target nursing object from at least one nursing object based on task priority and object attribute information; The current nursing task and the object identifier of the target nursing object are sent to at least one nursing terminal so that the nursing terminal can prompt the target nursing object to perform the current nursing task.

[0005] Accordingly, embodiments of this application provide a nursing task scheduling device, including: The first receiving unit is used to receive a nursing request for a subject to be cared for, the nursing request carrying object data of at least one modality of the subject to be cared for. The identification unit is used to identify the intention of the object to be cared for based on object data, obtain the nursing intention information of the object to be cared for, and determine the current object status of the object to be cared for. The determination unit is used to generate the current nursing task for the object to be cared for based on the nursing intent information and the current object status, and to determine the task priority of the current nursing task; The filtering unit is used to obtain object attribute information of at least one nursing object, and filter out the target nursing object from at least one nursing object according to task priority and object attribute information. The sending unit is used to send the current nursing task and the object identifier of the target nursing object to at least one nursing terminal, so that the nursing terminal prompts the target nursing object to perform the current nursing task.

[0006] Optionally, in some embodiments, this application may also provide another method for scheduling nursing tasks, including: When object data of at least one modality of the object to be cared for is collected, a care request for the object to be cared for is generated based on the object data; Send the nursing request to the server and receive the task information of the current nursing task corresponding to the nursing request returned by the server; When the target care object is detected to have entered the object area where the care object is located, the task information of the current care task is displayed; When a target care object is detected to have left the object area, a status update request for the current care task is sent to the server so that the server can update the current care task based on the status update request.

[0007] Accordingly, embodiments of this application provide another nursing task scheduling device, including: The generation unit is used to generate a nursing request for the object to be cared for based on the object data when at least one modality of the object to be cared for is collected. The second receiving unit is used to send the nursing request to the server and receive the task information of the current nursing task corresponding to the nursing request returned by the server. The display unit is used to display the task information of the current nursing task when the target nursing object is detected to have entered the object area where the object to be cared for is located; The update unit is used to send a status update request for the current care task to the server when it is detected that the target care object has left the object area, so that the server can update the current care task based on the status update request.

[0008] In some embodiments, the identification unit may be specifically used to extract audio data of the object to be cared for from the object data, and convert the audio data into text to obtain converted text; identify at least one keyword in the converted text, and determine at least one object intent of the object to be cared for based on the keyword to obtain object intent information.

[0009] In some embodiments, the identification unit may be specifically used to acquire reference audio data of the object to be cared for, and to extract features from the reference audio data to obtain reference acoustic features of the object to be cared for; to filter out audio data in audio modalities from the object data to obtain original audio data; and to extract audio data associated with the object to be cared for from the original audio data based on the reference acoustic features.

[0010] In some embodiments, the identification unit may be specifically used to filter out the object physiological data of the object in the object data, and determine the initial object state of the object to be cared for based on the object physiological data; extract at least one dimension of emotional features in the audio data, and determine the current emotional type of the object to be cared for based on the emotional features; and adjust the initial object state according to the emotional type to obtain the current object state of the object to be cared for.

[0011] In some embodiments, the identification unit may be specifically used to extract object videos in video modal from object data, and extract facial image sequences of the object to be cared for from the object videos; perform feature extraction on the facial image sequences to obtain facial expression features of the object to be cared for, the facial expression features indicating the changes in the facial expression of the object to be cared for; identify the body movement information of the object to be cared for in the object videos, and determine the current emotion type of the object to be cared for based on emotion features, facial expression features and body movement information.

[0012] In some embodiments, the determining unit may be specifically used to determine the nursing task type of the object to be cared for based on nursing intent information and the current object status; obtain at least one historical nursing task of the object to be cared for, and query the target historical nursing task corresponding to the nursing task type in the historical nursing tasks; and generate the current nursing task of the object to be cared for based on the query result of the target historical nursing task.

[0013] In some embodiments, the determining unit may be specifically used to extract at least one object task data associated with the nursing task from the object data and determine the trigger time of the nursing request; when a target historical nursing task exists, the target historical nursing task is updated based on at least one object task data and the trigger time to obtain the current nursing task; when no target historical nursing task exists, at least one nursing task for the object to be cared for is constructed according to the nursing task type, at least one object task data and the trigger time to obtain the current nursing task.

[0014] In some embodiments, the determining unit may be specifically used to determine the waiting time of a nursing request based on the triggering time; when the waiting time exceeds a preset time threshold, the preset task priority of the current nursing task is adjusted based on the waiting time to obtain the task priority of the current nursing task; when the waiting time does not exceed the preset time threshold, the preset task priority of the current nursing task is used as the task priority of the current nursing task.

[0015] In some embodiments, the determining unit may be specifically used to acquire at least one historical nursing request for the object to be cared for within a preset time range, and calculate the request similarity between the nursing request and at least one historical nursing request; count the number of historical nursing requests whose request similarity exceeds a preset similarity threshold in at least one historical nursing request; and adjust the preset task priority of the current nursing task based on the number of requests to obtain the task priority of the current nursing task.

[0016] In some embodiments, the determining unit may be specifically used to perform anomaly detection on object data of at least one modality to obtain anomaly detection results, wherein the anomaly detection results indicate the degree of anomaly in the object to be cared for; when the anomaly detection results indicate that anomaly exists, the preset task priority of the current nursing task is adjusted based on the anomaly detection results to obtain the task priority of the current nursing task; when the anomaly detection results indicate that no anomaly exists, the preset task priority of the current nursing task is used as the task priority of the current nursing task.

[0017] In some embodiments, the filtering unit may be specifically used to determine at least one filtering condition corresponding to the current nursing task from a preset set of filtering conditions based on task priority; and to filter out the target nursing object from at least one nursing object based on object attribute information and at least one filtering condition.

[0018] In some embodiments, the nursing task scheduling device can also be used to update the status of the current nursing task based on the status update request when receiving a status update request for the current nursing task returned by a nursing terminal; filter out the remaining nursing terminals other than those that returned the status update request from at least one nursing terminal; and send the status update request to the remaining nursing terminals so that the remaining nursing terminals can update the status of the current nursing task based on the status update request.

[0019] In some embodiments, the nursing task scheduling device can also be used to display a nursing page corresponding to the object area. The nursing page includes a nursing task display area, which includes historical task information of at least one historical nursing task. When the target nursing object is detected to enter the object area where the object to be cared for is located, the task information of the current nursing task is displayed, including: displaying the task information of the current nursing task in the nursing task display area.

[0020] In some embodiments, the nursing task scheduling device can also be used to update the task information of the current nursing task based on the status update request; and display the updated task information of the current nursing task in the nursing task display area.

[0021] Furthermore, embodiments of this application also provide an electronic device, including a processor and a memory, wherein the memory stores an application program, and the processor is used to run the application program in the memory to execute the nursing task scheduling method provided in embodiments of this application.

[0022] Furthermore, embodiments of this application also provide a computer-readable storage medium storing a plurality of instructions adapted for loading by a processor to execute steps in any of the text care task scheduling methods provided in embodiments of this application.

[0023] Furthermore, this application also provides a computer program product, including a computer program or instructions, which, when executed by a processor, implement the steps in the nursing task scheduling method provided in this application.

[0024] This embodiment of the application, upon receiving a nursing request for a patient, performs intent recognition on the patient based on object data to obtain the patient's nursing intent information and determine the patient's current object state. Then, based on the nursing intent information and the current object state, it generates the current nursing task for the patient and determines the task priority. Next, it obtains object attribute information for at least one patient and, based on the task priority and object attribute information, selects the target patient from among the at least one patient. Finally, it sends the current nursing task and the object identifier of the target patient to at least one nursing terminal so that the nursing terminal can prompt the target patient to perform the current nursing task. Because this scheme can obtain the patient's nursing request, identify the patient's nursing intent information and current object state, and then automatically generate the current nursing task based on the nursing intent information and current object state, it improves the accuracy of nursing task scheduling. Attached Figure Description

[0025] To more clearly illustrate the technical solutions in the embodiments of this application, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying 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.

[0026] Figure 1 This is a schematic diagram of a scenario for the nursing task scheduling method provided in the embodiments of this application; Figure 2 This is a flowchart illustrating the nursing task scheduling method provided in the embodiments of this application; Figure 3 This is another flowchart illustrating the nursing task scheduling method provided in the embodiments of this application; Figure 4 This is a schematic diagram of a patient session provided in an embodiment of this application; Figure 5 This is a schematic diagram of the medical host interface provided in an embodiment of this application; Figure 6 This is a schematic diagram of the nursing dashboard interface provided in an embodiment of this application; Figure 7 This is a schematic diagram of the door extension interface provided in an embodiment of this application; Figure 8 This is a schematic diagram of another interface of the door extension unit provided in an embodiment of this application; Figure 9 This is a schematic diagram of the structure of the nursing task scheduling device provided in the embodiments of this application; Figure 10 This is another structural schematic diagram of the nursing task scheduling device provided in the embodiments of this application; Figure 11 This is a schematic diagram of the structure of the electronic device provided in the embodiments of this application. Detailed Implementation

[0027] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0028] This application provides a nursing task scheduling method and related equipment. The related equipment may include a nursing task scheduling device, an electronic device, a computer program product, and a computer-readable storage medium. The nursing task scheduling device may be integrated into an electronic device, which may be a server or a terminal, etc.

[0029] The server can be a standalone physical server, a server cluster or distributed system composed of multiple physical servers, or a cloud server providing basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, content delivery network (CDN), and big data and artificial intelligence platforms. The terminal can be a smartphone, tablet, laptop, desktop computer, smart speaker, smartwatch, etc., but is not limited to these. The terminal and server can be directly or indirectly connected via wired or wireless communication, which is not limited herein.

[0030] For example, see Figure 1 Taking the integration of a nursing task scheduling device into an electronic device as an example, after receiving a nursing request for a patient, the electronic device performs intent recognition on the patient based on the patient's data to obtain the patient's nursing intent information and determine the patient's current state. Then, based on the nursing intent information and the current state, it generates the current nursing task for the patient and determines the task priority. Next, it acquires the object attribute information of at least one patient and, based on the task priority and object attribute information, selects the target patient from among the at least one patient. Finally, it sends the current nursing task and the object identifier of the target patient to at least one nursing terminal so that the nursing terminal can prompt the target patient to execute the current nursing task, thereby improving the accuracy of nursing task scheduling.

[0031] It is understood that, in the specific embodiments of this application, object data and other related data involving at least one modality of the object to be cared for are required to obtain permission or consent when the following embodiments of this application are applied to specific products or technologies, and the collection, use and processing of related data must comply with the relevant laws, regulations and standards of the relevant countries and regions.

[0032] The following sections provide detailed descriptions of each example. It should be noted that the order in which the embodiments are described is not intended to limit the preferred order of the embodiments.

[0033] This embodiment will be described from the perspective of a nursing task scheduling device, which can be integrated into an electronic device, such as a server or a terminal. The terminal can include tablet computers, laptops, personal computers (PCs), wearable devices, virtual reality devices, or other smart devices capable of scheduling nursing tasks.

[0034] A nursing task scheduling method, comprising: The system receives a nursing request for a patient, the request carrying object data of at least one modality of the patient; based on the object data, it performs intent recognition on the patient to obtain nursing intent information and determines the current object state; based on the nursing intent information and the current object state, it generates a current nursing task for the patient and determines the task priority of the current nursing task; it obtains object attribute information of at least one patient and, based on the task priority and object attribute information, selects a target patient from at least one patient; and it sends the current nursing task and the object identifier of the target patient to at least one nursing terminal so that the nursing terminal prompts the target patient to perform the current nursing task.

[0035] like Figure 2 As shown, Figure 2 This is a schematic diagram of an optional nursing task scheduling method provided in an embodiment of the present disclosure. The nursing task scheduling method can be executed by the server alone, and the specific flow of the nursing task scheduling method can be as follows: 101. Receive nursing requests for patients requiring care.

[0036] The nursing request carries object data in at least one modality of the patient requiring care. For example, object data may include audio data, video data, physiological data, etc., of the patient. Physiological data may include data representing the patient's basic vital signs such as heart rate, blood pressure, and body temperature; data representing the patient's laboratory indicators such as complete blood count, electrolytes, and blood glucose; and physiological signals such as the number of times the patient gets out of bed and turns over. Additionally, the nursing request may also carry basic physical data of the patient requiring care, such as age, gender, height, and weight.

[0037] 102. Based on object data, perform intent recognition on the object to be cared for, obtain the nursing intent information of the object to be cared for, and determine the current object status of the object to be cared for.

[0038] For example, object data can be used to identify the intentions of the person being cared for, such as whether they want water or need their medication changed. Furthermore, object data can also be used to determine the current state of the person being cared for, such as their current physiological or emotional state.

[0039] One method for obtaining nursing intent information of the object to be cared for based on object data is as follows: For example, audio data of the object to be cared for is extracted from the object data, and the audio data is converted into text to obtain the converted text; at least one keyword is identified in the converted text, and based on the keyword, at least one object intent of the object to be cared for is determined to obtain object intent information.

[0040] For example, audio data, such as phrases like "I have a slight fever," "I'm not feeling well," or "Call the nurse," can be extracted from object data and then converted into text. Keywords, such as "fever," "drink water," or "change dressing," can then be identified from the converted text. Based on these keywords, the intent of the person requiring care can be determined; for example, the intent might be to want water or need a dressing change. Alternatively, this application can also directly identify the object's intent information from the audio data using an intent recognition model.

[0041] Since the raw audio data in the object data may include noise from medical equipment, other people, environmental noise, etc., it is necessary to process the raw audio data to extract the audio data of the object to be cared for. The method for extracting the audio data of the object to be cared for from the object data can be as follows: For example, obtain reference audio data of the object to be cared for, and extract features from the reference audio data to obtain reference acoustic features of the object to be cared for; filter out audio data of audio modes in the object data to obtain raw audio data; and extract audio data associated with the object to be cared for from the raw audio data based on the reference acoustic features.

[0042] For example, reference audio data of the person to be cared for can be collected in advance, that is, audio of the person to be cared for can be collected beforehand to understand the person's voice characteristics. The reference acoustic characteristics of the person to be cared for can be determined based on the reference audio data. Raw audio data can be filtered from the object data, and audio data can be extracted from the raw audio data based on the reference acoustic characteristics. Because this audio data is extracted based on the person to be cared for's voice characteristics, it is audio data associated with the person to be cared for, thus avoiding misidentification of other people's audio as the person to be cared for, thereby improving the accuracy of audio data extraction.

[0043] In some embodiments, adaptive noise suppression is required on the raw audio data to suppress interference factors in the acquisition environment. This step aims to filter out or reduce persistent low-frequency humming from medical equipment, noise from ambient air conditioning and ventilation systems, and non-stationary background noise generated by personnel activities, thereby improving the signal-to-noise ratio of the target speech signal. Based on this, dynamic range compression and spectral enhancement processing can be applied to the denoised audio signal. Gain adjustment and frequency equalization techniques are used to strengthen the formant structure of the speech, allowing subtle speech details (such as breathiness and codas) to be clearly presented, ensuring the accuracy of subsequent feature extraction.

[0044] In some embodiments, since the original audio data may also include audio segments that do not contain substantial language content, a speech activity detection (VAD) method can be used to establish an effective discrimination model based on the time-domain energy envelope, zero-crossing rate, and frequency-domain characteristic parameters of the audio signal, in order to distinguish between the patient's speech segments and non-speech segments. VAD can automatically identify and eliminate non-speech vocalizations, silence, or ambient noise.

[0045] Nonverbal vocalizations refer to subconscious groans produced by the patient due to pain, such as "hmm," "ah," coughing, and sighing sounds that do not possess a linguistic structure. Silence and ambient noise refer to periods when the patient is silent or when only background noise is present.

[0046] By processing the raw audio data, we can obtain effective audio data that is clean, has complete language content, and has clinical analysis value, laying a reliable data foundation for subsequent acoustic feature analysis and auxiliary assessment of the patient's condition.

[0047] The methods for determining the current state of the object to be cared for can be as follows: For example, physiological data of objects are filtered out from object data, and the initial object state of the object to be cared for is determined based on the physiological data; at least one dimension of emotional features is extracted from audio data, and the current emotional type of the object to be cared for is determined based on the emotional features; the initial object state is adjusted according to the emotional type to obtain the current object state of the object to be cared for.

[0048] For example, physiological data can be filtered from the patient's data, and based on a comprehensive assessment of multi-dimensional physiological data, the patient's immediate clinical status, i.e., the initial patient status, can be accurately determined. This process requires the systematic integration and joint interpretation of vital signs such as heart rate, blood pressure, respiratory rate, blood oxygen saturation, and body temperature, along with fluid analysis indicators such as complete blood count, electrolytes, blood gas analysis, and myocardial enzyme profiles. Abnormalities in a single physiological indicator often lack specificity. Only by identifying patterns of indicator combinations with pathophysiological logic, such as decreased blood pressure accompanied by increased heart rate and elevated lactate indicating circulatory dysfunction, or decreased blood oxygen accompanied by increased respiratory rate and abnormal carbon dioxide partial pressure indicating respiratory insufficiency, can an objective understanding of the patient's overall condition be formed.

[0049] Furthermore, in addition to cross-sectional comprehensive interpretation, time-series dynamic trend analysis should be introduced to compare current indicators with the patient's previous baseline values ​​and historical evolution trajectory, distinguishing between transient physiological fluctuations and progressive pathological changes, thereby classifying the patient's status into different levels such as stable, compensated, decompensated, and crisis. This assessment framework, with multi-indicator joint interpretation as its core and dynamic trend tracking as a supplement, can provide continuous, quantitative, and early warning-capable situational awareness evidence for clinical decision-making.

[0050] In this process, after determining the initial state of the object to be cared for based on the object's physiological data, at least one dimension of emotional features can be extracted from the audio data. Based on these emotional features, the current emotional type of the object to be cared for can be determined. The initial state of the object is then adjusted according to the emotional type to obtain the current state of the object to be cared for.

[0051] For example, emotional features such as pitch, speech rate, volume, formants, and pauses can be identified from audio data, and these features can be used to determine the patient's psychological and emotional state. High-pitched, fast-paced, loud speech with rich pitch variations tends to indicate excitement or anger, while low-pitched, slow-paced, low-volume speech with a steady rhythm is often associated with sadness or fatigue. Furthermore, paralinguistic information such as pause patterns, breathing intervals, and speech quality (e.g., breath sounds, throaty sounds) can also be used for emotion assessment. For instance, a significantly shortened breathing interval accompanied by an increased harmonic noise ratio (throaty sounds) often indicates high-arousal negative emotions such as anxiety, pain, or tension; while slower breathing and an increase in high-frequency breath components tend to indicate fatigue or depression. Specifically, prosodic and spectral features, i.e., emotional features, can be extracted from audio data first, and then temporal modeling and classification can be performed using deep learning models (such as LSTM and Transformer) to capture dynamic changes in emotion. To improve accuracy, the converted text after speech-to-text transcription can be integrated to form a multimodal comprehensive judgment.

[0052] One method for determining the current emotional type of a person requiring care based on emotional characteristics is as follows: For example, the video modality of the object is extracted from the object data, and the facial image sequence of the object to be cared for is extracted from the object video; the facial image sequence is used to extract features to obtain the facial expression features of the object to be cared for, and the facial expression features indicate the changes in the facial expression of the object to be cared for; the body movement information of the object to be cared for is identified in the object video, and the current emotional type of the object to be cared for is determined based on the emotional features, facial expression features and body movement information.

[0053] The object video can be acquired by video acquisition equipment installed in the patient's environment, and the object video includes video images of the patient over a period of time.

[0054] For example, specific emotions often correspond to specific combinations of action units and dynamic sequences. Happiness is typically accompanied by a pulled-back mouth and contraction of the orbicularis oculi muscle, while anger manifests as lowered eyebrows, taut eyelids, and tightly closed lips. Therefore, after acquiring facial image sequences related to the patient's face in a video, facial muscle contractions can be decomposed into several basic action units, such as raised inner eyebrows, tightened eyelids, and pulled-back mouth. Specifically, neural networks can be used for facial keypoint detection and feature extraction, followed by modeling and classifying the time series of action units. Simultaneously, auxiliary cues such as head posture and gaze direction can be combined to comprehensively determine the patient's emotions.

[0055] Furthermore, rapid, large-amplitude, and irregular limb movements are usually associated with high arousal emotions (such as anger, anxiety, or excitement), while slow, gentle, and fluid movements tend to indicate low arousal states (such as calmness, sadness, or fatigue). Therefore, it is also possible to identify the body movement information of the person being cared for in the video, and to identify the patient's emotions through body movement information.

[0056] One approach is to use network models that simultaneously consider emotional characteristics, facial expressions, and bodily movements to determine the current emotional state of the person receiving care. This method allows for a more accurate assessment of the patient's current emotional state, enabling the determination of the appropriate tasks to be provided and their urgency.

[0057] 103. Based on the nursing intent information and the current status of the object, generate the current nursing task for the object to be cared for, and determine the task priority of the current nursing task.

[0058] The current nursing task can include fields such as task type, patient's bed number, patient's name, task content, task priority, trigger time, and task status. For example, based on the nursing intent information, it can be determined that patient A currently wants to turn over and drink water, and based on the current status, it can be determined that the patient needs to be turned over as soon as possible, and then helped to drink water. Then, current nursing task A can be generated, which includes two sub-tasks: turning over and drinking water, and the task priority of turning over can be determined to be higher than that of drinking water. As another example, if it is determined that patient B currently wants to remove an IV and change medication, then current nursing task B can be generated. Since removing the IV and changing medication is more urgent, the task priority of current nursing task B is determined to be higher than the overall task priority of current nursing task A.

[0059] The method for generating the current nursing task for the object to be cared for, based on nursing intent information and the current object status, can be as follows: For example, based on nursing intent information and the current object status, determine the nursing task type of the object to be cared for; obtain at least one historical nursing task of the object to be cared for, and query the target historical nursing task corresponding to the nursing task type in the historical nursing tasks; generate the current nursing task of the object to be cared for based on the query results of the target historical nursing task.

[0060] For example, the patient's task type needs can be clarified first based on nursing intent information and current status. Then, a matching target historical nursing task can be retrieved from the patient's historical nursing records. Next, the execution parameters, completion quality, and feedback results of the target historical tasks are analyzed and statistically summarized, serving as the basis for generating the current nursing task plan. This process, by using historical execution results as an important reference for task decision-making, achieves data-driven and personalized adaptation in nursing task formulation. It ensures that the recommended tasks not only align with the patient's long-term recovery trajectory but also fully respect their current physiological and psychological limits, improving the rationality and predictability of task arrangement.

[0061] Based on the query results of the target historical nursing tasks, the current nursing task for the patient to be cared for is generated, including: For example, extract at least one object task data associated with the nursing task from the object data and determine the trigger time of the nursing request; when a target historical nursing task exists, update the target historical nursing task based on at least one object task data and the trigger time to obtain the current nursing task; when no target historical nursing task exists, construct at least one nursing task for the object to be cared for according to the nursing task type, the at least one object task data and the trigger time to obtain the current nursing task.

[0062] For example, if a target historical nursing task exists corresponding to the nursing task type, it can be updated based on the object task data extracted from the object data and the trigger time of the patient's request, so that the generated current nursing task can be directly associated with the patient's previous nursing history. If no target historical nursing task exists corresponding to the nursing task type, a new current nursing task can be created.

[0063] Specifically, refer to Figure 4 After the intelligent agent generates the current nursing task based on the patient's audio data, it can display the information corresponding to the current nursing task on the screen and prompt "A nursing task has been generated for you", so that the patient knows that the request has been responded to, thereby reducing the patient's anxiety.

[0064] The methods for determining the task priority of the current nursing task can be as follows: For example, the waiting time for a nursing request is determined based on the trigger time; when the waiting time exceeds a preset time threshold, the preset task priority of the current nursing task is adjusted based on the waiting time to obtain the task priority of the current nursing task; when the waiting time does not exceed the preset time threshold, the preset task priority of the current nursing task is used as the task priority of the current nursing task.

[0065] For example, each task is assigned a preset task priority upon generation, which can be determined based on the task type. Then, the waiting time from when the patient triggers the request to the current time can be calculated. The longer the waiting time, the more impatient the patient is, and the more quickly the request needs to be processed. Therefore, when the waiting time exceeds a preset threshold, the preset task priority of the current nursing task can be increased based on the waiting time, resulting in the current nursing task's task priority. Furthermore, the number of priority increases can be determined based on the waiting time; tasks with longer waiting times can receive a higher priority increase.

[0066] The methods for determining the task priority of the current nursing task can be as follows: For example, acquire at least one historical nursing request for the subject to be cared for within a preset time range, and calculate the request similarity between the nursing request and at least one historical nursing request; count the number of historical nursing requests whose request similarity exceeds a preset similarity threshold in at least one historical nursing request; based on the number of requests, adjust the preset task priority of the current nursing task to obtain the task priority of the current nursing task.

[0067] For example, if a patient is impatient, they may send the same request multiple times in a short period of time. If each request generates a corresponding task, there will be too many meaningless tasks. These tasks not only consume resources and affect efficiency, but also affect the judgment of medical staff. Therefore, when the same patient repeatedly expresses the same or similar requests in a short period of time, multiple tasks are not generated repeatedly. Instead, existing tasks are merged, the time is refreshed, the number of calls is increased, or the priority is increased.

[0068] For example, it can retrieve at least one historical nursing request for a patient within a preset time range and calculate the request similarity between the nursing request and at least one historical nursing request. If it is found that the same or similar requests are sent repeatedly within the preset time range, the task priority can be adjusted according to the number of these same or similar requests. In other words, the more times the patient sends the same or similar requests, the more quickly the task needs to be executed, and thus the task priority needs to be increased.

[0069] The methods for determining the task priority of the current nursing task can be as follows: For example, anomaly detection is performed on object data of at least one modality to obtain anomaly detection results, which indicate the degree of abnormality in the object to be cared for; when the anomaly detection results indicate the presence of anomalies, the preset task priority of the current nursing task is adjusted based on the anomaly detection results to obtain the task priority of the current nursing task; when the anomaly detection results indicate the absence of anomalies, the preset task priority of the current nursing task is used as the task priority of the current nursing task.

[0070] For example, anomaly detection can be performed on the data. If abnormal behavior, unusual sounds, or abnormal physiological signals are detected from a patient, it indicates that the current situation requires special attention. Therefore, when anomaly detection results indicate an anomaly, the preset task priority of the current nursing task can be adjusted based on the anomaly detection results. For example, if a patient makes abnormal cries of pain, medical staff need to care for the patient as soon as possible, so the corresponding task priority can be increased.

[0071] 104. Obtain object attribute information for at least one nursing object, and filter out the target nursing object from at least one nursing object based on task priority and object attribute information.

[0072] In this embodiment of the application, the object of care may refer to medical personnel, such as doctors, nurses, or other personnel related to care.

[0073] Given the large number of medical and nursing staff in hospitals, and the fact that these staff may have other tasks to perform, it is necessary to select the most suitable medical and nursing personnel based on various criteria and assign them corresponding nursing tasks. A method for selecting target nursing subjects from at least one nursing subject based on task priority and object attribute information can be as follows: For example, based on task priority, at least one filtering condition corresponding to the current nursing task is determined from a set of preset filtering conditions; and the target nursing object is selected from at least one nursing object based on object attribute information and at least one filtering condition.

[0074] For example, screening criteria can be determined based on task priority. These criteria might include the bed assigned to the healthcare worker, their current location, and their workload. For instance, the most urgent tasks should be handled by the healthcare worker closest to the patient; moderately urgent tasks require consideration of the bed assigned to the healthcare worker, their current location, and their workload; less urgent tasks only require consideration of the bed assigned to the healthcare worker. Furthermore, screening criteria can be adjusted based on actual circumstances and are not limited to the three criteria mentioned above.

[0075] 105. Send the current nursing task and the object identifier of the target nursing object to at least one nursing terminal so that the nursing terminal can prompt the target nursing object to perform the current nursing task.

[0076] For example, after determining the current nursing task, the current nursing task and the patient's name, bed, age, and other object identifiers can be sent to multiple nursing terminals. These multiple nursing terminals correspond to multiple medical staff. This method can not only notify the medical staff who need to perform the task, but also avoid information silos caused by only the medical staff who need to perform the task knowing about it.

[0077] In addition, some nursing tasks require different medical staff to handle. For example, some nursing tasks require a specific doctor, while others require a nurse. Therefore, different types of medical staff can be assigned based on the specific nursing task. Some nursing tasks also include multiple sub-tasks, so appropriate medical staff need to be assigned based on the urgency of each sub-task. If a nursing task includes two sub-tasks that need to be handled simultaneously, two medical staff can be assigned to perform the task together; if a nursing task includes two sub-tasks that need to be handled sequentially, one medical staff member is sufficient.

[0078] This application's embodiments can be applied to smart ward nursing task processing scenarios, involving terminals such as bedside units, door units, medical staff mainframes, and nursing dashboards. Patients can interact via voice or text through the intelligent agent on the bedside unit. The system automatically determines nursing requests and generates nursing tasks based on information such as the patient's expression content, speech rate, tone, and abnormal sounds, and simultaneously displays them to each nurse's terminal device.

[0079] The bedside unit allows patients to initiate nursing requests. The door unit displays patient tasks for the room, enabling nurses to confirm tasks upon arrival and review them. The main control panel allows nurses to view, receive, and process nursing tasks across the entire ward. The nursing dashboard displays the number of tasks, beds, urgency level, and processing status for the entire ward.

[0080] For example, after generating the current nursing task, it can be simultaneously pushed to the medical staff main unit, nursing dashboard, and door-side extension. The bedside extension is equipped with a microphone, speaker, and touchscreen to collect patient voice, text input, and abnormal sounds. The medical staff main unit, nursing dashboard, and door-side extension are all networked smart terminals that support task display, task processing, and status synchronization. The door-side extension also supports card swiping and NFC, and can be optionally equipped with Bluetooth AOA positioning and other nurse identity and spatial behavior recognition capabilities.

[0081] The healthcare server displays bed availability, patients, nursing requests, urgency levels, and wait times in a task list format. (Reference) Figure 5The medical care main unit includes multiple areas, each corresponding to a patient. The area displays the patient's ID, name, gender, and age. Different colors or symbols indicate different levels of care, such as special care, level one, level two, level three, and unattended calls, clearly displaying the required level of care for each patient within the designated area. (Reference) Figure 6 The nursing dashboard includes patients in different beds and displays pending tasks using red markers or numerical prompts. The dashboard allows users to track the current status of each task. (Reference) Figure 7 The door extension displays task reminders in the corresponding bed area, allowing you to know which medical staff and patients are performing tasks in that room.

[0082] The method for updating the status of a task after medical staff have performed it can be as follows: For example, when receiving a status update request for the current nursing task from a nursing terminal, the status of the current nursing task is updated based on the status update request; the remaining nursing terminals other than those that returned the status update request are selected from at least one nursing terminal; the status update request is sent to the remaining nursing terminals so that the remaining nursing terminals can update the status of the current nursing task based on the status update request.

[0083] For example, after medical staff complete a task, the task status can be updated from "incomplete" to "completed". Furthermore, the task status of all medical staff's terminals can be updated simultaneously via MQTT or Web Socket protocols to avoid issues such as repeated execution of a task due to information discrepancies.

[0084] The technical solution of this application enables patients to freely express their needs via voice. The system can automatically complete voice recognition, intent judgment, and task generation, eliminating the need for patients to manually select menus or for nurses to relay and input the information, significantly reducing manual operation costs. Furthermore, the system can not only recognize keywords in the patient's voice but also combine information such as tone, speed, volume, pauses, and abnormal sounds to determine the patient's needs and urgency, making nursing task generation more accurate. In addition, the system can dynamically adjust task priorities based on the patient's status and merge or escalate repeated calls from the same patient within a short period, reducing duplicate reminders and preventing task omissions or multiple processing. Nursing task status includes "added," "processing," "processed," and "cleared," which can be pushed in real-time via the backend server to bedside extensions, medical staff mainframes, nursing dashboards, and door extensions, ensuring consistent display across all terminals and eliminating information silos. The system also senses nurses' entry and exit from the ward through card swiping, NFC, or Bluetooth AOA positioning. After the nurse completes the task and leaves the ward, the system automatically clears the relevant task and synchronously updates all terminals, reducing unnecessary reminders and the risk of misoperation, ensuring a complete nursing loop.

[0085] As described above, after receiving a nursing request for a patient, this embodiment of the application identifies the patient's intent based on object data, obtains the patient's nursing intent information, and determines the patient's current object state. Then, based on the nursing intent information and the current object state, it generates the current nursing task for the patient and determines the task priority. Next, it obtains the object attribute information of at least one patient and, based on the task priority and object attribute information, selects the target patient from among the at least one patient. Finally, it sends the current nursing task and the object identifier of the target patient to at least one nursing terminal, so that the nursing terminal can prompt the target patient to perform the current nursing task. Because this solution can obtain the patient's nursing request, identify the patient's nursing intent information and current object state, and then automatically generate the current nursing task based on the nursing intent information and current object state, it improves the accuracy of nursing task scheduling.

[0086] Based on the method described in the above embodiments, the following examples will provide further detailed explanations.

[0087] In this embodiment, the nursing task scheduling device will be specifically integrated into a bedside terminal as an example for explanation. Figure 3 As shown, Figure 3 This is an optional flowchart illustrating a nursing task scheduling method provided in an embodiment of the present disclosure. The specific flow of the nursing task scheduling method can be as follows: 201. When object data of at least one modality of the object to be cared for is collected, a nursing request for the object to be cared for is generated based on the object data.

[0088] For example, when a terminal device placed at the bedside collects object data of at least one modality of the patient, a care request can be generated.

[0089] 202. Send the nursing request to the server and receive the task information of the current nursing task corresponding to the nursing request returned by the server.

[0090] For example, the server can generate task information for the current nursing task based on the nursing request.

[0091] 203. When the target nursing object is detected to have entered the object area where the object to be cared for is located, the task information of the current nursing task is displayed.

[0092] For example, when a healthcare worker assigned a task enters a patient's room via card swiping, NFC, or Bluetooth AOA positioning, the task information for the current nursing assignment can be displayed on the door intercom. This task information can be shown in a pop-up window. (Reference) Figure 8 When a nurse is detected entering a patient's room, a pop-up window can be used to display the tasks to be performed in that room.

[0093] This includes displaying the nursing page corresponding to the object area, which includes a nursing task display area and historical task information for at least one historical nursing task. When a target nursing object is detected to have entered the object area where the object to be cared for is located, the task information of the current nursing task is displayed, including displaying the task information of the current nursing task in the nursing task display area.

[0094] For example, the nursing page can display historical task information so that medical staff can understand what tasks have been performed on the patient and make more appropriate decisions.

[0095] 204. When the target nursing object is detected to have left the object area, a status update request for the current nursing task is sent to the server so that the server can update the current nursing task based on the status update request.

[0096] For example, when it is confirmed by swiping a card or using a location signal that medical staff have completed their tasks and left the room, the door extension can send a "room task clearing" command to the server. Upon receiving the command, the server will broadcast it to all terminals and update the task status.

[0097] This includes updating task information for the current nursing task based on status update requests; and displaying the updated task information in the nursing task display area. For example, the medical staff main unit, nursing dashboard, door extension, and bedside extension will all update the task list, red icon, numerical badge, and task status in real time after receiving instructions to add, upgrade, process, or clear a task.

[0098] As described above, in this embodiment, when object data of at least one modality of the object to be cared for is collected, a nursing request for the object to be cared for is generated based on the object data. The nursing request is then sent to the server, and the server returns task information for the current nursing task corresponding to the nursing request. When the target object is detected entering the object area where the object to be cared for is located, the task information for the current nursing task is displayed. When the target object is detected leaving the object area, a status update request for the current nursing task is sent to the server, so that the server updates the current nursing task based on the status update request. Because this scheme can obtain the nursing request of the object to be cared for, identify the nursing intent information and current object status of the object to be cared for, and then automatically generate the current nursing task based on the nursing intent information and current object status, the accuracy of nursing task scheduling is improved.

[0099] To better implement the above methods, this application also provides a nursing task scheduling device, which can be integrated into a server. The terminal can include a tablet computer, a laptop computer, and / or a personal computer, etc.

[0100] For example, such as Figure 9 As shown, the nursing task scheduling device may include a first receiving unit 301, an identification unit 302, a determination unit 303, a filtering unit 304, and a sending unit 305, as follows: (1) First receiving unit 301; The first receiving unit 301 is used to receive a nursing request for a person to be cared for, the nursing request carrying object data of at least one modality of the person to be cared for.

[0101] (2) Identification unit 302; The identification unit 302 is used to identify the intention of the object to be cared for based on object data, obtain the nursing intention information of the object to be cared for, and determine the current object status of the object to be cared for.

[0102] (3) Determine unit 303; The determining unit 303 is used to generate the current nursing task for the object to be cared for based on the nursing intention information and the current object status, and to determine the task priority of the current nursing task.

[0103] (4) Filtering unit 304; The filtering unit 304 is used to obtain object attribute information of at least one nursing object, and filter out the target nursing object from at least one nursing object according to task priority and object attribute information.

[0104] (5) Transmitting unit 305; The sending unit 305 is used to send the current nursing task and the object identifier of the target nursing object to at least one nursing terminal, so that the nursing terminal prompts the target nursing object to perform the current nursing task.

[0105] In some embodiments, the identification unit 302 may be specifically used to extract audio data of the object to be cared for from the object data, and convert the audio data into text to obtain converted text; identify at least one keyword in the converted text, and determine at least one object intent of the object to be cared for based on the keyword to obtain object intent information.

[0106] In some embodiments, the identification unit 302 may be specifically used to acquire reference audio data of the object to be cared for, and to extract features from the reference audio data to obtain reference acoustic features of the object to be cared for; to filter out audio data in audio modalities from the object data to obtain original audio data; and to extract audio data associated with the object to be cared for from the original audio data based on the reference acoustic features.

[0107] In some embodiments, the identification unit 302 may be specifically used to filter out the object physiological data of the object in the object data, and determine the initial object state of the object to be cared for based on the object physiological data; extract at least one dimension of emotional features in the audio data, and determine the current emotional type of the object to be cared for based on the emotional features; and adjust the initial object state according to the emotional type to obtain the current object state of the object to be cared for.

[0108] In some embodiments, the identification unit 302 may be specifically used to extract object videos in video mode from object data, and extract facial image sequences of the object to be cared for from the object videos; perform feature extraction on the facial image sequences to obtain facial expression features of the object to be cared for, the facial expression features indicating the changes in the facial expression of the object to be cared for; identify the body movement information of the object to be cared for in the object videos, and determine the current emotion type of the object to be cared for based on emotion features, facial expression features and body movement information.

[0109] In some embodiments, the determining unit 303 may be specifically used to determine the nursing task type of the object to be cared for based on nursing intention information and the current object status; obtain at least one historical nursing task of the object to be cared for, and query the target historical nursing task corresponding to the nursing task type in the historical nursing tasks; and generate the current nursing task of the object to be cared for based on the query result of the target historical nursing task.

[0110] In some embodiments, the determining unit 303 may be specifically used to extract at least one object task data associated with the nursing task from the object data and determine the trigger time of the nursing request; when a target historical nursing task exists, the target historical nursing task is updated based on at least one object task data and the trigger time to obtain the current nursing task; when no target historical nursing task exists, at least one nursing task for the object to be cared for is constructed according to the nursing task type, at least one object task data and the trigger time to obtain the current nursing task.

[0111] In some embodiments, the determining unit 303 may be specifically used to determine the waiting time of a nursing request based on the trigger time; when the waiting time exceeds a preset time threshold, the preset task priority of the current nursing task is adjusted based on the waiting time to obtain the task priority of the current nursing task; when the waiting time does not exceed the preset time threshold, the preset task priority of the current nursing task is used as the task priority of the current nursing task.

[0112] In some embodiments, the determining unit 303 may be specifically used to acquire at least one historical nursing request for the object to be cared for within a preset time range, and calculate the request similarity between the nursing request and at least one historical nursing request; count the number of historical nursing requests whose request similarity exceeds a preset similarity threshold in at least one historical nursing request; and adjust the preset task priority of the current nursing task based on the number of requests to obtain the task priority of the current nursing task.

[0113] In some embodiments, the determining unit 303 may be specifically used to perform anomaly detection on object data of at least one modality to obtain anomaly detection results, wherein the anomaly detection results indicate the degree of anomaly in the object to be cared for; when the anomaly detection results indicate that anomaly exists, the preset task priority of the current nursing task is adjusted based on the anomaly detection results to obtain the task priority of the current nursing task; when the anomaly detection results indicate that no anomaly exists, the preset task priority of the current nursing task is used as the task priority of the current nursing task.

[0114] In some embodiments, the filtering unit 304 may be specifically used to determine at least one filtering condition corresponding to the current nursing task from a preset set of filtering conditions based on task priority; and to filter out the target nursing object from at least one nursing object based on object attribute information and at least one filtering condition.

[0115] In practice, each of the above units can be implemented as an independent entity or can be arbitrarily combined to be implemented as the same or several entities. For the specific implementation of each of the above units, please refer to the previous method embodiments, which will not be repeated here.

[0116] As described above, after receiving a nursing request for a patient, this embodiment of the application identifies the patient's intent based on object data, obtains the patient's nursing intent information, and determines the patient's current object state. Then, based on the nursing intent information and the current object state, it generates the current nursing task for the patient and determines the task priority. Next, it obtains the object attribute information of at least one patient and, based on the task priority and object attribute information, selects the target patient from among the at least one patient. Finally, it sends the current nursing task and the object identifier of the target patient to at least one nursing terminal, so that the nursing terminal can prompt the target patient to perform the current nursing task. Because this solution can obtain the patient's nursing request, identify the patient's nursing intent information and current object state, and then automatically generate the current nursing task based on the nursing intent information and current object state, it improves the accuracy of nursing task scheduling.

[0117] To better implement the above methods, this application also provides a nursing task scheduling device, which can be integrated into a terminal, such as a tablet computer, a laptop computer, or a personal computer.

[0118] For example, such as Figure 10 As shown, the nursing task scheduling device may include a generation unit 306, a second receiving unit 307, a display unit 308, and an update unit 309, as follows: (1) Generating unit 306; The generation unit 306 is used to generate a nursing request for the object to be cared for based on the object data when at least one modality of the object to be cared for is collected.

[0119] (2) Second receiving unit 307; The second receiving unit 307 is used to send a nursing request to the server and receive the task information of the current nursing task corresponding to the nursing request returned by the server.

[0120] (3) Display Unit 308; The display unit 308 is used to display the task information of the current nursing task when the target nursing object is detected to have entered the object area where the object to be cared for is located.

[0121] (4) Update unit 309; The update unit 309 is used to send a status update request for the current nursing task to the server when the target nursing object is detected to have left the object area, so that the server can update the current nursing task based on the status update request.

[0122] In some embodiments, the nursing task scheduling device can also be used to update the status of the current nursing task based on the status update request when receiving a status update request for the current nursing task returned by a nursing terminal; filter out the remaining nursing terminals other than those that returned the status update request from at least one nursing terminal; and send the status update request to the remaining nursing terminals so that the remaining nursing terminals can update the status of the current nursing task based on the status update request.

[0123] In some embodiments, the nursing task scheduling device can also be used to display a nursing page corresponding to the object area. The nursing page includes a nursing task display area, which includes historical task information of at least one historical nursing task. When the target nursing object is detected to enter the object area where the object to be cared for is located, the task information of the current nursing task is displayed, including: displaying the task information of the current nursing task in the nursing task display area.

[0124] As described above, in this embodiment, when object data of at least one modality of the object to be cared for is collected, a nursing request for the object to be cared for is generated based on the object data. The nursing request is then sent to the server, and the server returns task information for the current nursing task corresponding to the nursing request. When the target object is detected entering the object area where the object to be cared for is located, the task information for the current nursing task is displayed. When the target object is detected leaving the object area, a status update request for the current nursing task is sent to the server, so that the server updates the current nursing task based on the status update request. Because this scheme can obtain the nursing request of the object to be cared for, identify the nursing intent information and current object status of the object to be cared for, and then automatically generate the current nursing task based on the nursing intent information and current object status, the accuracy of nursing task scheduling is improved.

[0125] This application also provides an electronic device, such as... Figure 11 As shown, it illustrates a structural schematic diagram of the electronic device involved in the embodiments of this application, specifically: The electronic device may include components such as a processor 401 with one or more processing cores, a memory 402 with one or more computer-readable storage media, a power supply 403, and an input unit 404. Those skilled in the art will understand that... Figure 11 The electronic device structure shown does not constitute a limitation on the electronic device and may include more or fewer components than shown, or combine certain components, or have different component arrangements. Wherein: The processor 401 is the control center of the electronic device, connecting various parts of the device via various interfaces and lines. It executes software programs and / or modules stored in the memory 402, and calls data stored in the memory 402, to perform various functions and process data. Optionally, the processor 401 may include one or more processing cores; preferably, the processor 401 may integrate an application processor and a modem processor, wherein the application processor mainly handles the operating system, user interface, and applications, and the modem processor mainly handles wireless communication. It is understood that the modem processor may not be integrated into the processor 401.

[0126] The memory 402 can be used to store software programs and modules. The processor 401 executes various functional applications and data processing by running the software programs and modules stored in the memory 402. The memory 402 may mainly include a program storage area and a data storage area. The program storage area may store the operating system, application programs required for at least one function (such as sound playback function, image playback function, etc.), etc.; the data storage area may store data created according to the use of the electronic device, etc. In addition, the memory 402 may include high-speed random access memory, and may also include non-volatile memory, such as at least one disk storage device, flash memory device, or other volatile solid-state storage device. Accordingly, the memory 402 may also include a memory controller to provide the processor 401 with access to the memory 402.

[0127] The electronic device also includes a power supply 403 that supplies power to the various components. Preferably, the power supply 403 can be logically connected to the processor 401 through a power management system, thereby enabling functions such as charging, discharging, and power consumption management through the power management system. The power supply 403 may also include one or more DC or AC power supplies, recharging systems, power fault detection circuits, power converters or inverters, power status indicators, and other arbitrary components.

[0128] The electronic device may also include an input unit 404, which can be used to receive input digital or character information, and generate keyboard, mouse, joystick, optical or trackball signal inputs related to user settings and function control.

[0129] Although not shown, the electronic device may also include a display unit, etc., which will not be described in detail here. Specifically, in this embodiment, the processor 401 in the electronic device loads the executable files corresponding to the processes of one or more applications into the memory 402 according to the following instructions, and the processor 401 runs the applications stored in the memory 402 to realize various functions, as follows: The system receives a nursing request for a patient, the request carrying object data of at least one modality of the patient; based on the object data, it performs intent recognition on the patient to obtain nursing intent information and determines the current object state; based on the nursing intent information and the current object state, it generates a current nursing task for the patient and determines the task priority of the current nursing task; it obtains object attribute information of at least one patient and, based on the task priority and object attribute information, selects a target patient from at least one patient; and it sends the current nursing task and the object identifier of the target patient to at least one nursing terminal so that the nursing terminal prompts the target patient to perform the current nursing task.

[0130] or, When object data for at least one modality of the object to be cared for is collected, a nursing request for the object is generated based on the object data. The nursing request is sent to the server, and the server returns task information for the current nursing task corresponding to the nursing request. When the target object is detected entering the object area where the object to be cared for is located, the task information for the current nursing task is displayed. When the target object is detected leaving the object area, a status update request for the current nursing task is sent to the server so that the server can update the current nursing task based on the status update request. Specific implementation details of each of the above operations can be found in the preceding embodiments and will not be repeated here.

[0131] As described above, after receiving a nursing request for a patient, this embodiment of the application identifies the patient's intent based on object data, obtains the patient's nursing intent information, and determines the patient's current object state. Then, based on the nursing intent information and the current object state, it generates the current nursing task for the patient and determines the task priority. Next, it obtains the object attribute information of at least one patient and, based on the task priority and object attribute information, selects the target patient from among the at least one patient. Finally, it sends the current nursing task and the object identifier of the target patient to at least one nursing terminal, so that the nursing terminal can prompt the target patient to perform the current nursing task. Because this solution can obtain the patient's nursing request, identify the patient's nursing intent information and current object state, and then automatically generate the current nursing task based on the nursing intent information and current object state, it improves the accuracy of nursing task scheduling.

[0132] Those skilled in the art will understand that all or part of the steps in the various methods of the above embodiments can be performed by instructions, or by instructions controlling related hardware. These instructions can be stored in a computer-readable storage medium and loaded and executed by a processor.

[0133] Therefore, embodiments of this application provide a computer-readable storage medium storing a plurality of instructions that can be loaded by a processor to execute steps in any of the nursing task scheduling methods provided in embodiments of this application.

[0134] For details on the implementation of each of the above operations, please refer to the previous examples, which will not be repeated here.

[0135] The computer-readable storage medium may include: read-only memory (ROM), random access memory (RAM), disk or optical disk, etc.

[0136] Since the instructions stored in the computer-readable storage medium can execute the steps in any of the nursing task scheduling methods provided in the embodiments of this application, the beneficial effects that any of the nursing task scheduling methods provided in the embodiments of this application can achieve can be realized, as detailed in the preceding embodiments, and will not be repeated here.

[0137] According to one aspect of this application, a computer program product or computer program is provided, comprising computer instructions stored in a computer-readable storage medium. A processor of an electronic device reads the computer instructions from the computer-readable storage medium and executes the computer instructions, causing the electronic device to perform the methods provided in the various alternative implementations of the nursing task scheduling or image classification described above.

[0138] The foregoing has provided a detailed description of a nursing task scheduling method and related equipment provided in the embodiments of this application. The related equipment may include a nursing task scheduling device, electronic equipment, computer program products, and computer-readable storage media. Specific examples have been used to illustrate the principles and implementation methods of the present invention. The description of the above embodiments is only for the purpose of helping to understand the method and core ideas of the present invention. At the same time, for those skilled in the art, there will be changes in the specific implementation methods and application scope based on the ideas of the present invention. Therefore, the content of this specification should not be construed as a limitation of the present invention.

Claims

1. A method for scheduling nursing tasks, characterized in that, include: Receive a nursing request for a person to be cared for, the nursing request carrying object data of at least one modality of the person to be cared for; Based on the object data, the intention of the object to be cared for is identified to obtain the nursing intention information of the object to be cared for, and the current object status of the object to be cared for is determined. Based on the nursing intent information and the current object status, generate the current nursing task for the object to be cared for, and determine the task priority of the current nursing task; Obtain object attribute information of at least one nursing object, and filter out the target nursing object from the at least one nursing object according to the task priority and the object attribute information; The current nursing task and the object identifier of the target nursing object are sent to at least one nursing terminal so that the nursing terminal prompts the target nursing object to perform the current nursing task.

2. The nursing task scheduling method according to claim 1, characterized in that, The step of identifying the intent of the object to be cared for based on the object data to obtain the nursing intent information of the object to be cared for includes: The audio data of the object to be cared for is extracted from the object data, and the audio data is converted into text to obtain the converted text; At least one keyword is identified in the converted text, and based on the keyword, at least one object intent of the object to be cared for is determined to obtain object intent information.

3. The nursing task scheduling method according to claim 2, characterized in that, The step of extracting the audio data of the object to be cared for from the object data includes: Acquire reference audio data of the object to be cared for, and extract features from the reference audio data to obtain reference acoustic features of the object to be cared for; The audio data of the audio modality is filtered out from the object data to obtain the original audio data; Based on the reference acoustic features, audio data associated with the object to be cared for is extracted from the original audio data.

4. The nursing task scheduling method according to claim 2, characterized in that, Determining the current state of the object to be cared for includes: The object physiological data of the object is filtered out from the object data, and the initial object state of the object to be cared for is determined based on the object physiological data; At least one dimension of emotional features is extracted from the audio data, and the current emotional type of the person to be cared for is determined based on the emotional features. Based on the emotion type, the initial object state is adjusted to obtain the current object state of the object to be cared for.

5. The nursing task scheduling method according to claim 4, characterized in that, Determining the current emotional type of the person requiring care based on the emotional characteristics includes: Extract the object video in the video modality from the object data, and extract the facial image sequence of the object to be cared for from the object video; Feature extraction is performed on the facial image sequence to obtain the facial expression features of the subject to be cared for, and the facial expression features indicate the changes in the facial expression of the subject to be cared for. The body movement information of the person to be cared for is identified in the video of the object, and the current emotional type of the person to be cared for is determined based on the emotional features, the facial expression features and the body movement information.

6. The nursing task scheduling method according to claim 1, characterized in that, The step of generating the current nursing task for the object to be cared for based on the nursing intent information and the current object status includes: Based on the nursing intent information and the current object status, the nursing task type of the object to be cared for is determined; Obtain at least one historical nursing task of the patient to be cared for, and query the target historical nursing task corresponding to the nursing task type from the historical nursing tasks; Based on the query results of the target historical nursing tasks, the current nursing task for the patient to be cared for is generated.

7. The nursing task scheduling method according to claim 6, characterized in that, The step of generating the current nursing task for the patient based on the query results of the target historical nursing tasks includes: Extract at least one object task data associated with the nursing task from the object data, and determine the trigger time of the nursing request; When the target historical nursing task exists, the target historical nursing task is updated based on the at least one object task data and the trigger time to obtain the current nursing task; When the target historical nursing task does not exist, at least one nursing task for the object to be cared for is constructed based on the nursing task type, the at least one object task data and the trigger time, to obtain the current nursing task.

8. The nursing task scheduling method according to claim 6, characterized in that, Determining the task priority of the current nursing task includes: The waiting time for the nursing request is determined based on the triggering time. When the waiting time exceeds a preset time threshold, the preset task priority of the current nursing task is adjusted based on the waiting time to obtain the task priority of the current nursing task; When the waiting time does not exceed the preset time threshold, the preset task priority of the current nursing task is used as the task priority of the current nursing task.

9. The nursing task scheduling method according to claim 1, characterized in that, Determining the task priority of the current nursing task includes: Obtain at least one historical nursing request for the subject within a preset time range, and calculate the request similarity between the nursing request and the at least one historical nursing request; The number of historical nursing requests whose similarity exceeds a preset similarity threshold is counted from the at least one historical nursing request. Based on the number of requests, the preset task priority model of the current nursing task is adjusted to obtain the task priority of the current nursing task.

10. The nursing task scheduling method according to claim 1, characterized in that, Determining the task priority of the current nursing task includes: Anomaly detection is performed on the object data of the at least one modality to obtain anomaly detection results, wherein the anomaly detection results indicate the degree of abnormality of the object to be cared for; When the anomaly detection result indicates that an anomaly exists, the preset task priority of the current nursing task is adjusted based on the anomaly detection result to obtain the task priority of the current nursing task. When the anomaly detection result indicates that there is no anomaly, the preset task priority of the current nursing task is used as the task priority of the current nursing task.

11. The nursing task scheduling method according to claim 1, characterized in that, The step of selecting the target care object from the at least one care object based on the task priority and the object attribute information includes: Based on the task priority, at least one filtering condition corresponding to the current nursing task is determined from the preset set of filtering conditions; Based on the object attribute information and the at least one filtering condition, the target nursing object is selected from the at least one nursing object.

12. The nursing task scheduling method according to claim 1, characterized in that, Also includes: When a status update request for the current nursing task is received from the nursing terminal, the status of the current nursing task is updated based on the status update request. Filter out the remaining nursing terminals from the at least one nursing terminal, excluding those that returned the status update request; The status update request is sent to the remaining nursing terminals so that the remaining nursing terminals can update the status of the current nursing task based on the status update request.

13. A method for scheduling nursing tasks, characterized in that, include: When object data of at least one modality of the object to be cared for is collected, a nursing request for the object to be cared for is generated based on the object data; The nursing request is sent to the server, and the server returns the task information of the current nursing task corresponding to the nursing request. When the target nursing object is detected to have entered the object area where the object to be cared for is located, the task information of the current nursing task is displayed; When the target care object is detected to have left the object area, a status update request for the current care task is sent to the server so that the server can update the current care task based on the status update request.

14. The nursing task scheduling method according to claim 13, characterized in that, Before displaying the task information of the current nursing task when the target nursing object is detected to have entered the object area where the object to be cared for is located, the method further includes: Display the nursing page corresponding to the object area. The nursing page includes a nursing task display area, which includes historical task information of at least one historical nursing task. When the target nursing object is detected to have entered the object area where the object to be cared for is located, the task information of the current nursing task is displayed, including: displaying the task information of the current nursing task in the nursing task display area.

15. The nursing task scheduling method according to claim 14, characterized in that, After sending a status update request for the current care task to the server when the target care object is detected to have left the object area, the method further includes: Based on the status update request, the task information of the current nursing task is updated; The updated task information for the current nursing task is displayed in the nursing task display area.

16. A nursing task scheduling device, characterized in that, include: The first receiving unit is configured to receive a nursing request for a subject to be cared for, the nursing request carrying object data of at least one modality of the subject to be cared for. The identification unit is used to identify the intention of the object to be cared for based on the object data, obtain the nursing intention information of the object to be cared for, and determine the current object status of the object to be cared for. The determining unit is used to generate the current nursing task for the object to be cared for based on the nursing intention information and the current object status, and to determine the task priority of the current nursing task; A filtering unit is used to obtain object attribute information of at least one nursing object, and to filter out a target nursing object from the at least one nursing object according to the task priority and the object attribute information; A sending unit is configured to send the current nursing task and the object identifier of the target nursing object to at least one nursing terminal, so that the nursing terminal prompts the target nursing object to perform the current nursing task.

17. A nursing task scheduling device, characterized in that, include: The generation unit is used to generate a nursing request for the object to be cared for based on the object data when at least one modality of the object to be cared for is collected. The second receiving unit is used to send the nursing request to the server and receive the task information of the current nursing task corresponding to the nursing request returned by the server. The display unit is used to display the task information of the current nursing task when the target nursing object is detected to have entered the object area where the object to be cared for is located; An update unit is configured to send a status update request for the current care task to the server when the target care object is detected to have left the object area, so that the server updates the current care task based on the status update request.

18. An electronic device, characterized in that, It includes a processor and a memory, the memory storing an application program, and the processor running the application program within the memory to perform the steps of the nursing task scheduling method according to any one of claims 1 to 15.

19. A computer program product comprising a computer program / instructions, characterized in that, When the computer program / instructions are executed by the processor, they implement the steps of the nursing task scheduling method according to any one of claims 1 to 15.

20. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a plurality of instructions adapted for loading by a processor to perform the steps of the nursing task scheduling method according to any one of claims 1 to 15.

Images