An intelligent ward nursing system

Abstract

This invention relates to the field of interdisciplinary technology of artificial intelligence and medical information, and discloses an intelligent ward nursing system aimed at solving the problems of scattered nursing task information, lack of dynamic scheduling, and high-risk early warning mechanisms in existing technologies. The system includes: constructing a structured nursing needs database; dynamically generating a unified nursing task list for the entire ward; visually displaying task status through a graphical interface in multiple dimensions; establishing a linkage early warning mechanism for high-risk states such as severe / critical illness; and integrating dynamic physiological data such as fluid intake and output, and weight to drive adaptive nursing scheduling. By adopting the above technical solution, this application can achieve centralized management, intelligent early warning, and proactive intervention of nursing tasks, significantly improving nursing efficiency and preventing emergencies.

Description

Technical Field

[0001] This invention relates to the field of interdisciplinary technology of artificial intelligence and medical information, specifically to an intelligent ward nursing system. Background Technology

[0002] With the continuous advancement of smart healthcare systems, the intelligentization and efficiency of ward nursing work have become key aspects of improving hospital service quality and patient safety. The nursing projects involved in modern wards are becoming increasingly complex and diverse, including vital sign monitoring, tube management, intake and output recording, and special treatment support. The timeliness and accuracy of these nursing tasks are directly related to the patient's recovery process and the ability to prevent and control sudden risks.

[0003] Among them, the intelligent ward nursing system aims to centrally identify, dynamically track and visualize patients' nursing needs through information technology. Its core objective is to standardize and structure the scattered nursing tasks and provide nursing staff with real-time and comprehensive decision support in combination with the frequency and timeliness requirements.

[0004] Existing technologies still have significant shortcomings in ward nursing management: First, nursing task information is scattered across paper records, electronic medical records, or independent monitoring devices, lacking a unified and integrated display platform, making it difficult for nurses to quickly grasp the nursing priorities of all patients in the ward; second, most systems only support static medical order entry and cannot automatically associate nursing frequency, execution window, and urgency level according to the type of medical order, resulting in omissions, duplications, or delays; third, the systems generally lack a linkage and early warning mechanism for high-risk states (such as severe or critical illness) and key nursing operations (such as ECG monitoring, oxygen administration, and maintenance of gastric / urinary catheters), making it difficult to achieve proactive intervention; finally, existing solutions fail to effectively integrate dynamic physiological data such as fluid intake and output, weight changes, and other correlation logic with nursing tasks, and cannot support adaptive nursing scheduling based on the evolution of patient status.

[0005] The aforementioned deficiencies severely restrict the rational allocation of nursing resources and the early identification of sudden illnesses, necessitating an intelligent ward nursing system capable of intelligently aggregating multi-dimensional nursing needs, providing real-time visualization, and enabling dynamic response. Summary of the Invention

[0006] (a) Technical problems to be solved To address the shortcomings of existing technologies, this invention provides an intelligent ward nursing system that can effectively solve the problems mentioned in the background section.

[0007] (II) Technical Solution To achieve the above objectives, the present invention provides the following technical solution: an intelligent ward nursing system, comprising the following modules: The module for constructing a structured database of patient care needs is used to extract each patient's medical orders, condition status, and nursing operation type based on hospital information system and electronic medical record data. It standardizes and codes nursing tasks such as blood glucose measurement, blood pressure measurement, electrocardiogram monitoring, oxygen administration, gastric tube maintenance, urinary catheter maintenance, critical illness labeling, critical illness labeling, fluid intake and output records, and weight measurement, and associates them with corresponding execution frequency, execution window, and time priority. The module for dynamically generating nursing task lists is used to calculate the pending execution status, remaining execution time window, and urgency of each patient's nursing tasks in real time based on the structured database combined with the current timestamp and historical execution records, and generate a unified dynamic nursing task list for the entire ward. The module for visually presenting nursing task information is used to simultaneously display the dynamic nursing task list on nurse workstation terminals and mobile terminals through a graphical interface. It is classified and sorted in multiple dimensions according to bed number, task type, and urgency, and the task status is distinguished by color coding. It supports filtering and focusing by nursing item, patient status, or execution frequency. The high-risk status linkage early warning module is used to automatically increase the priority of all related nursing tasks when a patient is identified as being in a critical or serious condition, and to trigger mandatory reminders for critical operations such as electrocardiogram monitoring, oxygen administration, and tubing maintenance. At the same time, it pushes early warning information to the responsible nurse and the on-duty doctor. The adaptive scheduling module continuously collects dynamic physiological parameters such as patient fluid intake and output and weight changes. It combines these parameters with a preset threshold model to determine the trend of patient status evolution. If abnormal fluctuations are detected, it automatically adjusts the execution frequency of relevant nursing tasks or adds temporary nursing instructions to achieve adaptive nursing scheduling based on patient status.

[0008] Preferably, in the module for constructing a structured database of patient care needs, the standardized coding of nursing tasks adopts a unified medical terminology system, and each task is bound to a unique task identifier, a standard operating procedure description, a recommended execution time, and the type of nursing resources required.

[0009] Preferably, in the module that dynamically generates the nursing task list, the execution frequency includes multiple fixed cycles and personalized non-periodic frequencies set according to medical orders. The system dynamically determines whether a task is in a pending execution, timed-out execution, or completed state through a time sliding window algorithm.

[0010] Preferably, in the module that visualizes nursing task information, the graphical interface adopts a ward floor plan layout, with each bed location corresponding to an interactive card. The card displays the number and type of nursing items to be performed in the form of icons. Clicking on the card will expand a detailed task list, including the task name, last execution time, next planned time, execution requirements, and historical completion records.

[0011] Preferably, the high-risk status linkage early warning module is configured with a three-level response strategy: Level 1 is a critical illness status, which triggers a red flashing warning and a forced pop-up reminder; Level 2 is a serious illness status, which triggers a yellow highlight and places the task list at the top; Level 3 is a normal high-risk operation, which triggers a blue indicator and starts polling reminders during a specific time period before the task expires.

[0012] Preferably, in the adaptive scheduling module, fluid intake and output records are automatically acquired by connecting to the intelligent infusion pump, urine collection device and drinking water recording terminal, and weight data is uploaded in real time by the networked intelligent scale. When the difference between intake and output exceeds a preset threshold or the weight change within a specific time period exceeds a preset threshold, the system automatically triggers a fluid balance assessment task and suggests adjusting the care plan.

[0013] Preferably, the dynamic nursing task list supports multi-terminal collaborative operation. After the nurse confirms the task on the mobile terminal, the system immediately updates the server status and automatically records the executor, execution time, operation notes and abnormal situations, forming a complete nursing process traceability chain.

[0014] Preferably, it also includes a nursing resource load assessment module, which is used to calculate the real-time workload index of each nurse based on the number of nurses, skill level and task distribution density of the current shift, and to issue a manpower allocation suggestion to the nursing team leader when the load index exceeds a preset threshold.

[0015] Preferably, it also includes a nursing resource load assessment module, which is used to calculate the real-time workload index of each nurse based on the number of nurses, skill level and task distribution density of the current shift, and to issue a manpower allocation suggestion to the nursing team leader when the load index exceeds a preset threshold.

[0016] Compared with the prior art, the present invention provides an intelligent ward nursing system, which has the following beneficial effects: 1. Achieve centralized and structured management of nursing tasks. By constructing a unified structured nursing needs database, nursing information scattered in paper records, electronic medical records and independent devices is integrated into standardized and computable task units, solving the problem of information silos. This enables nurses to fully grasp the nursing needs of patients in the entire ward within a single interface, significantly improving the efficiency and accuracy of task identification.

[0017] 2. Supports dynamic and intelligent task scheduling. The system not only statically displays medical orders, but also dynamically generates task lists based on time, execution records and patient status. Through an adaptive scheduling mechanism, it automatically adjusts the frequency of nursing care based on changes in physiological parameters such as fluid intake and output and weight, realizing the transformation from passive execution to proactive intervention and effectively preventing the deterioration of the condition due to nursing delays.

[0018] 3. Establish a proactive early warning system for high-risk states. For high-risk states such as serious illness or critical illness, the system automatically increases the priority of related tasks and triggers multi-channel early warnings to ensure that critical nursing operations are not missed, significantly reduce response delays in emergencies, and improve the level of patient safety.

[0019] 4. Enhance the scientific allocation of nursing resources. By calculating nurses' workload index in real time and providing manpower allocation suggestions, the system assists managers in optimizing shift scheduling and task allocation, avoiding local overload or idle resources, and improving the overall work efficiency and sustainability of the nursing team.

[0020] 5. Construct a complete and traceable nursing execution closed loop. From task generation and execution confirmation to abnormal record, the system leaves a complete record, forming an immutable nursing process data chain. This provides a reliable basis for quality control, performance evaluation, and adverse event analysis, and promotes the development of nursing services towards refinement and standardization. Attached Figure Description

[0021] Figure 1 This is a schematic diagram of the overall technical architecture of the intelligent ward nursing system proposed in this invention; Figure 2 This is a schematic diagram of the core principle framework of the dynamic physiological data-driven adaptive nursing scheduling mechanism in this invention; Detailed Implementation

[0022] The technical solutions of the embodiments of the present invention 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.

[0023] Example 1 This embodiment is applied to the intensive care unit (ICU) or high-dependency ward (HDU) of a tertiary-level general hospital. In this scenario, patients' conditions are complex, nursing tasks are intensive and time-sensitive, and there are a large number of high-risk status indicators. To support efficient, accurate, and traceable nursing work, this invention deploys a complete intelligent ward nursing system, whose hardware architecture and software logic are deeply integrated to form a closed-loop intelligent nursing scheduling platform.

[0024] At the system architecture level, this embodiment constructs a distributed electromechanical-information system consisting of a central server, nurse workstation terminals, mobile nursing terminals, intelligent sensing devices, and network infrastructure. The central server adopts a dual-machine hot standby architecture, with both the primary and standby servers configured with Intel Xeon Silver 4310 processors, 64GB DDR4 ECC memory, and 2TB NVMe solid-state drives. It runs a customized containerized operating system based on the Linux kernel and deploys microservice architecture applications through the Docker engine. The server connects to the hospital's intranet via a 10 Gigabit fiber optic switch and interacts with the Hospital Information System (HIS), Electronic Medical Record System (EMR), and Laboratory Information System (LIS) via the HL7 v2.8 standard protocol to ensure real-time synchronization of patient basic information, medical orders, and test results.

[0025] The nurse workstation terminal is deployed at the nurse station's control panel, featuring a 27-inch 4K resolution industrial-grade touchscreen all-in-one machine. It incorporates an NVIDIA T1000 GPU to support smooth rendering of the graphical interface and runs desktop client applications developed based on the Electron framework. This terminal connects to the central server via Gigabit Ethernet and is equipped with a USB-C port for connecting external auxiliary devices such as barcode scanners and printers. The mobile nursing terminal is a customized Android 12 tablet device, powered by a Qualcomm Snapdragon 680 processor, 6GB RAM, and 128GB storage. It integrates an NFC module, a 5G / 4G / Wi-Fi 6 tri-mode communication module, and an IP67-rated dustproof and waterproof casing, facilitating task confirmation and data entry by nurses at the bedside. All terminals access the system via the hospital's Wi-Fi 6 wireless network, with signal coverage utilizing an 802.11ax standard AP array to ensure low-latency, highly reliable data transmission.

[0026] At the perception layer, the system integrates various intelligent medical devices as data acquisition sources. These include: a networked ECG monitor (model: Mindray BeneVision N12), which uploads vital signs such as heart rate, blood oxygen saturation, and respiratory rate every 5 seconds via HL7 over TCP / IP protocol; an intelligent infusion pump (B. Braun Space Infusion Pump) with a built-in flow sensor and communication module, recording infusion rate and cumulative volume in real time; an intelligent urine collection device (custom-developed, including a weighing sensor and liquid level detection module) that uploads urine volume data every 10 minutes; a networked intelligent scale (Seca mBCA 515) that transmits encrypted weight data to a mobile terminal via Bluetooth 5.0; and an RFID tag reader / writer installed beside the bed to identify patient wristband information, ensuring accurate matching of the patient.

[0027] The aforementioned hardware components are interconnected through a three-layer network topology: the perception layer devices connect to the edge gateway (an embedded gateway based on ARM Cortex-A53, running the OpenWrt system) via wired (RS485 / Modbus RTU) or wireless (BLE / Zigbee) methods. After the edge gateway completes protocol conversion and data preprocessing, it uploads structured data to the central server via an HTTPS RESTful API. The server and each terminal communicate with each other using bidirectional TLS 1.3 encrypted communication to ensure data security.

[0028] Based on this system architecture, this invention executes the following dynamic workflow to achieve a complete closed loop from data acquisition to adaptive scheduling: First, the data acquisition module constructs a structured database of patient care needs. This module periodically retrieves the latest medical orders from the HIS / EMR system (every 5 minutes), including both long-term and temporary orders, from all patients in the ward. The medical order text is parsed by a Natural Language Processing (NLP) engine and mapped to a pre-defined standardized nursing task library. This task library uses the SNOMED CT (Systematized Nomenclature of Medicine – Clinical Terms) international medical terminology system for encoding, with each task bound to a unique task identifier (Task ID). For example, "blood glucose measurement" corresponds to Task ID: GLU-001, and "gastric tube maintenance" corresponds to Task ID: NGT-003. The standard operating procedure (SOP) description associated with each Task ID, the recommended execution time (e.g., 3 minutes for blood glucose measurement), and the required nursing resource type (e.g., blood glucose meter, lancet) are all stored in a task metadata table in a PostgreSQL relational database. Meanwhile, the system automatically parses the frequency descriptions in the medical orders (such as "q2h", "bid", "prn") into structured execution rules, including fixed-cycle or personalized non-cycle rules (such as "once every 30 minutes within 6 hours after surgery"), and calculates the first execution time window and subsequent sliding windows.

[0029] Subsequently, the task scheduling engine module generates a dynamic nursing task list. This module uses the current system timestamp as a baseline, combined with historical execution records (stored in a MongoDB time-series database), to determine the status of each task for each patient. The determination logic employs a time-sliding window algorithm: for periodic tasks, if the current time falls within the interval [last execution time + period - tolerance window, last execution time + period + tolerance window], it is marked as "pending execution"; if it exceeds the upper limit, it is marked as "timed out and not executed"; if it is completed within the tolerance window, it is marked as "completed". The tolerance window is set according to the task type, such as ±5 minutes for vital sign monitoring and ±15 minutes for tubing maintenance. The urgency level is determined by the task's priority (e.g., critical illness-related tasks are P0 level) and the remaining execution time window, using a weighted scoring model: Urgency = α × basic task priority + β × (1 - remaining time / total window), where α = 0.7 and β = 0.3. Finally, the engine generates a JSON-formatted task list containing fields such as patient ID, bed number, task ID, status, urgency, and next scheduled time, and caches it in a Redis in-memory database for fast access.

[0030] Next, the visualization module performs a multi-terminal graphical display. This module renders the ward floor plan layout on the nurse workstation terminal, with each bed location corresponding to an interactive card component. The cards are developed using the React framework and maintain a long-term connection with the server via WebSocket to receive task update events in real time. The card background color dynamically changes according to the highest urgency task: red (P0), yellow (P1), blue (P2), and gray (no tasks). The center of the card displays an icon showing the total number of tasks to be executed. Clicking it brings up a modal box listing detailed tasks, including task name, last execution time (accurate to the second), next planned time, execution requirements (e.g., "fasting," "lying down"), and historical completion records (including executor, time spent, and remarks). The mobile terminal interface adopts Material Design specifications and supports filtering by task type (e.g., "pipeline-related," "monitoring-related"), patient status (severe / critical), or execution frequency (high-frequency / low-frequency). The filtered results are displayed in a waterfall layout and support pull-to-refresh and swipe-to-load.

[0031] When the system detects that a patient has been marked as critically ill or seriously ill, the high-risk alert module immediately triggers a response mechanism. This module continuously monitors status change events in the EMR system. Upon receiving a "critically ill" flag, it initiates a Level 1 response: on all terminals, the patient's bed number card is highlighted with a flashing red animation, and a full-screen alert window is forcibly popped up, listing all associated critical tasks (ECG monitoring, oxygen administration, tube maintenance, etc.), which must be manually confirmed by a nurse before it can be closed. Simultaneously, the system pushes a high-priority notification to the responsible nurse's mobile terminal via Firebase Cloud Messaging (FCM) and sends an alert SMS to the on-duty doctor via the hospital's internal SMS gateway. For "critically ill" status, a Level 2 response is triggered: the task list is pinned to the top, and the card is highlighted in yellow, but the pop-up window is not forcibly displayed. For routine high-risk procedures (such as gastric tube maintenance), the system begins polling reminders 30 minutes before the task's due date, displaying a blue notification bubble in the mobile terminal's status bar every 5 minutes until the task is completed.

[0032] During daily operation, the adaptive scheduling module continuously executes physiological data-driven scheduling. This module receives fluid intake and output data streams and weight data from smart devices. Fluid intake and output data include: intake reported by the infusion pump (unit: mL), oral intake recorded by the drinking water recording terminal (entered by nurses via mobile terminals or by patients via bedside tablets), and output reported by the urine collection device. The system calculates the cumulative 24-hour fluid intake and output difference ΔV every 15 minutes. If |ΔV| > 1000 mL, or the 24-hour weight change ΔW > 2 kg, a fluid balance assessment sub-process is triggered. This sub-process calls the built-in rule engine (customized based on the Drools open-source rule engine) to match preset clinical rules. For example, the rule base defines: "IF Patient status = severity of illness AND 2 consecutive ΔV < -800 mL AND systolic blood pressure < 90 mmHg THEN It is recommended to add a 'blood pressure monitoring every 2 hours' task, frequency = 2h, duration 24h." The rules engine outputs temporary nursing instructions, which, after being reviewed and confirmed by the nursing team leader on the workstation terminal, are automatically injected into the task scheduling engine to generate new tasks to be executed and synchronized to all terminals. The rule base is stored in JSON format, allowing clinical experts to add, delete, and modify rules through a web management backend, enabling dynamic evolution of knowledge.

[0033] In addition, the system includes a nursing resource load assessment module, which obtains the current list of nurses on duty, their skill levels (e.g., N1 / N2 / N3), and the number of beds they are responsible for at the start of each shift (e.g., 8:00 AM). The task scheduling engine categorizes tasks to be performed according to skill requirements (e.g., N2 and above can perform ECG monitoring) and calculates the expected workload for each nurse.

[0034] Finally, the traceability module ensures that all operations are fully traceable. When a nurse confirms a task on their mobile device, the system automatically records the executor's ID, precise execution time, operation notes, and any abnormalities, and writes the data to a blockchain-based log database to ensure data immutability. This traceability chain can be used for subsequent quality audits, performance evaluations, and root cause analysis of adverse events.

[0035] In summary, this embodiment, through the close integration of the above system architecture and dynamic processes, achieves a comprehensive intelligent upgrade of nursing tasks from static medical orders to dynamic scheduling, from scattered information to centralized visibility, and from passive execution to proactive early warning, effectively solving the problems of information silos, delayed response, and resource mismatch in the existing technology.

[0036] Example 2 This embodiment is applied to a general internal medicine ward in a primary care hospital. In this scenario, patients' conditions are relatively stable, and nursing tasks mainly consist of routine monitoring and basic care. However, there is a shortage of nurses and a weak information technology infrastructure. To adapt to this environment, this invention simplifies the hardware and reduces the functionality based on Embodiment 1, forming a lightweight deployment solution.

[0037] In terms of system architecture, the dual-machine hot standby server was eliminated, replaced by a single domestic Phytium FT-2000 / 4 processor server running the Kylin V10 operating system, while retaining containerized deployment capabilities. The nurse workstation terminal uses a standard commercial Windows 10 all-in-one machine with a 21.5-inch screen and a 1920×1080 resolution, connecting to the central server via RDP remote desktop protocol to reduce local computing burden. Mobile nursing terminals use lower-cost Android 10 tablets, retaining only Wi-Fi 5 communication capabilities and eliminating the 5G module. The sensing layer devices were significantly simplified: only a networked scale and a manual data entry terminal were retained. Fluid intake and output data are entered by nurses using structured forms on mobile terminals, with the system providing intelligent verification (e.g., a confirmation prompt pops up when a single fluid intake exceeds 1000mL).

[0038] Despite hardware simplification, the core software logic remains intact. The task scheduling engine has optimized its algorithm for grassroots applications: the tolerance of the time sliding window has been widened to ±30 minutes to accommodate execution delays caused by insufficient manpower; the high-risk warning mechanism retains only the first-level critical illness response, eliminating the distinction between severe illness and ordinary high-risk; the adaptive scheduling module's rule base has five pre-set rules (such as "weight loss >1.5kg in 24 hours → add nutritional assessment"), rather than being open for editing, thus lowering the barrier to entry.

[0039] The visual interface has also been adapted: the ward floor plan uses a simplified SVG vector graphic; cards only display the number of tasks and the color of the highest priority; and the details page that expands after clicking removes the history charts, retaining only the text list. Night mode is enabled by default, and the font size is increased by 20% to suit the visual needs of middle-aged and elderly nurses.

[0040] In terms of workflow, the system has significantly enhanced offline operation capabilities. When Wi-Fi is disconnected, mobile terminals can cache up to 4 hours of task operation data, which will automatically synchronize once the network is restored. The synchronization process uses incremental hash verification to ensure data consistency. Furthermore, the system has added a "batch confirmation" function: nurses can select and submit multiple tasks for the same patient all at once, reducing the number of operations required.

[0041] In this scenario, the resource load assessment module is adjusted to a "task density heatmap": on the workstation interface, different shades of red blocks display the task density of beds in each area, assisting the head nurse in verbal scheduling rather than generating precise manpower suggestions.

[0042] Through the aforementioned lightweight design, this embodiment significantly reduces deployment costs and operational complexity while ensuring core intelligent nursing functions, enabling the intelligent ward nursing system to be implemented in resource-limited primary healthcare institutions, demonstrating the broad applicability and scalability of the present invention.

[0043] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. An intelligent ward nursing system, characterized in that, Includes the following modules: The module for constructing a structured database of patient care needs is used to extract each patient's medical orders, condition status, and nursing operation type based on hospital information system and electronic medical record data. It standardizes and codes nursing tasks such as blood glucose measurement, blood pressure measurement, electrocardiogram monitoring, oxygen administration, gastric tube maintenance, urinary catheter maintenance, critical illness labeling, critical illness labeling, fluid intake and output records, and weight measurement, and associates them with corresponding execution frequency, execution window, and time priority. The module for dynamically generating nursing task lists is used to calculate the pending execution status, remaining execution time window, and urgency of each patient's nursing tasks in real time based on the structured database combined with the current timestamp and historical execution records, and generate a unified dynamic nursing task list for the entire ward. The module for visually presenting nursing task information is used to simultaneously display the dynamic nursing task list on nurse workstation terminals and mobile terminals through a graphical interface. It is classified and sorted in multiple dimensions according to bed number, task type, and urgency, and the task status is distinguished by color coding. It supports filtering and focusing by nursing item, patient status, or execution frequency. The high-risk status linkage early warning module is used to automatically increase the priority of all related nursing tasks when a patient is identified as being in a critical or serious condition, and to trigger mandatory reminders for critical operations such as electrocardiogram monitoring, oxygen administration, and tubing maintenance. At the same time, it pushes early warning information to the responsible nurse and the on-duty doctor. The adaptive scheduling module continuously collects dynamic physiological parameters such as patient fluid intake and output and weight changes. It combines these parameters with a preset threshold model to determine the trend of patient status evolution. If abnormal fluctuations are detected, it automatically adjusts the execution frequency of relevant nursing tasks or adds temporary nursing instructions to achieve adaptive nursing scheduling based on patient status.

2. The intelligent ward nursing system according to claim 1, characterized in that, In the module that constructs a structured database of patient care needs, the standardized coding of nursing tasks adopts a unified medical terminology system. Each task is bound to a unique task identifier, a standard operating procedure description, a recommended execution time, and the type of nursing resources required.

3. The intelligent ward nursing system according to claim 1, characterized in that, In the module that dynamically generates nursing task lists, the execution frequency includes multiple fixed cycles as well as personalized non-periodic frequencies set according to medical orders. The system dynamically determines whether a task is in a pending execution, timed-out, or completed state through a time sliding window algorithm.

4. The intelligent ward nursing system according to claim 1, characterized in that, In the module that visualizes nursing task information, the graphical interface adopts a ward floor plan layout. Each bed location corresponds to an interactive card. The card displays the number and type of nursing items to be performed in the form of icons. Clicking on the card will expand a detailed task list, including the task name, last execution time, next planned time, execution requirements, and historical completion records.

5. The intelligent ward nursing system according to claim 1, characterized in that, The high-risk status linkage early warning module is set with a three-level response strategy: Level 1 is a critical illness status, which triggers a red flashing warning and a forced pop-up reminder; Level 2 is a serious illness status, which triggers a yellow highlight and puts the task list at the top. Level 3 is a normal high-risk operation, which triggers a blue indicator and will start polling and reminding users during a specific period before the task expires.

6. The intelligent ward nursing system according to claim 1, characterized in that, In the adaptive scheduling module, fluid intake and output records are automatically acquired by connecting to the intelligent infusion pump, urine collection device and drinking water recording terminal. Weight data is uploaded in real time by the networked intelligent scale. When the difference between intake and output exceeds a preset threshold or the weight change within a specific time period exceeds a preset threshold, the system automatically triggers a fluid balance assessment task and suggests adjustments to the care plan.

7. The intelligent ward nursing system according to claim 1, characterized in that, The dynamic nursing task list supports multi-terminal collaborative operation. After the nurse completes the task confirmation on the mobile terminal, the system immediately updates the server status and automatically records the executor, execution time, operation notes and abnormal situations, forming a complete nursing process traceability chain.

8. The intelligent ward nursing system according to claim 1, characterized in that, It also includes a nursing resource load assessment module, which calculates the real-time workload index of each nurse based on the number of nurses on the current shift, skill level, and task distribution density. When the load index exceeds a preset threshold, it sends a manpower allocation suggestion to the nursing team leader.

9. The intelligent ward nursing system according to claim 1, characterized in that, The adaptive scheduling module has a built-in rule engine that allows clinical experts to configure nursing logic rules, and the rule base is dynamically expanded as clinical guidelines are updated. The module that visualizes nursing task information supports night mode and high-contrast display, and integrates voice broadcast function to issue targeted reminders through headphones or speakers when a task is about to expire.

Images