Intelligent nursing bed remote monitoring system with early warning and monitoring method thereof

Abstract

This invention relates to the field of nursing monitoring technology, and discloses an intelligent nursing bed remote monitoring system with early warning and its monitoring method. The system includes a bed body, a bed surface zone pressure detection unit, a bed posture detection unit, a bedside area sensing unit, a ward area sensing unit, a bedside controller, a communication gateway, a server, a nurse station terminal, and a mobile nursing terminal. The server is equipped with an authorization token receiving unit, a nursing rule engine, a risk assessment unit, a conflict resolution decision unit, a return-to-bed verification unit, and an early warning interaction unit. By unifying and integrating bed surface zone pressure detection, bed posture detection, bedside area detection, ward area detection, nursing constraint information, and authorized bed leaving information, a complete multi-source monitoring input structure is formed, enabling continuous identification of the patient's bedside, bed leaving, temporary bed leaving, and bed returning processes.

Description

Technical Field

[0001] This invention relates to the field of nursing monitoring technology, specifically to a remote monitoring system for intelligent nursing beds with early warning capabilities and its monitoring method. Background Technology

[0002] In existing inpatient nursing settings, monitoring patient in-bed status, out-of-bed status, and nursing risks typically involves a combination of bedside sensors and nurse station information systems. Common technical solutions include using bed pressure sensors to detect patient presence, using guardrail status or infrared sensors to determine out-of-bed behavior, sending alarm information to the nurse station via the information system, and integrating a pressure ulcer risk assessment system to monitor patient pressure and provide turning reminders. This type of technical solution can monitor the patient's basic condition and, to some extent, assist nursing staff in risk management.

[0003] However, in the aforementioned existing technologies, bed pressure detection, bed posture detection, bedside area detection, and ward area detection are usually set up independently. Their monitoring data do not form a unified data fusion structure and are not associated with nursing constraint information and authorized bed leaving information. This makes it difficult to continuously identify and uniformly manage the patient's in-bed status, out-of-bed status, and temporary out-of-bed status, and it is difficult to form a complete monitoring link covering in-bed monitoring, out-of-bed monitoring, and return-to-bed process. Summary of the Invention

[0004] To address the shortcomings of existing technologies, this invention provides a remote monitoring system and method for intelligent nursing beds with early warning capabilities. This system solves the problem that existing technologies lack a unified data fusion structure for monitoring data and do not associate it with nursing constraint information and authorized bed leave information, making it difficult to continuously identify and uniformly manage patients' in-bed status, out-of-bed status, and temporary out-of-bed status.

[0005] To achieve the above objectives, the present invention is implemented through the following technical solution: a remote monitoring system for intelligent nursing beds with early warning, comprising: a bed body, and a bed surface zone pressure detection unit, a bed posture detection unit, a bedside area perception unit, a ward area perception unit, an authorization token receiving unit, a nursing rule engine, a risk assessment unit, a conflict resolution decision unit, a return-to-bed verification unit, and an early warning interaction unit connected to the bed body; The bed surface zone pressure detection unit is used to collect information on the zone pressure, pressure center of gravity changes, and duration of continuous pressure when the patient is in bed. The bed posture detection unit is used to collect information on the guardrail status, bed back angle, and bed posture changes. The bedside area sensing unit is used to detect whether the patient is in the bedside activity area; The ward area sensing unit is used to detect whether a patient has left the bedside area or left the preset ward area. The authorization token receiving unit is used to receive the bed exit authorization token issued by the nurse's terminal; The nursing rule engine is used to read patient nursing constraint information, which includes at least fall risk level, pressure ulcer risk level, nighttime bed leave restrictions, and nursing orders. The risk assessment unit is used to calculate the bed-leaving risk index and the stress relief liability index based on the zoning pressure information, bed posture change information, bedside area information, ward area information, bed-leaving authorization token, and nursing constraint information. The conflict resolution decision unit is used to output corresponding graded early warning results and nursing treatment strategies based on the bed closure risk index and the pressure reduction liability index; The bed return verification unit is used to verify whether the patient has completed a valid bed return after the patient is detected to have returned to the bed. The early warning interaction unit is used to send the graded early warning results and nursing treatment strategies to the nurse station and / or mobile nursing terminal.

[0006] Preferably, the bed leave authorization token includes at least one or more of the following: patient identifier, bed identifier, authorization start time, allowed bed leave duration, allowed activity area, caregiver status, time period strategy, and bed leave purpose type.

[0007] Preferably, the conflict resolution decision unit constructs a patient care state machine, which includes at least the following states: bed safety state, bedside transition state, unauthorized bed leave state, authorized temporary bed leave state, bed leave timeout state, suspected lost state, valid bed return state, and invalid bed return state.

[0008] Preferably, the risk index of leaving the bed is determined by any two or more of the following parameters: fall risk level, nighttime, nursing orders, caregiver status, distance from the bed, duration of time away from the bed, changes in the status of the guardrail, and bedside activity trajectory.

[0009] Preferably, the decompression liability index is determined by any two or more of the following parameters: duration of local continuous pressure, peak pressure in the zone, magnitude of pressure center shift, frequency of body position changes, duration of pressure relief after leaving the bed, and pressure recovery status after returning to the bed.

[0010] Preferably, the conflict resolution decision unit is configured as follows: When the risk index of leaving the bed is higher than the first threshold and the pressure relief debt index is lower than the second threshold, an early warning to restrict leaving the bed is output. When the decompression liability index is higher than the second threshold and the bed leave risk index is lower than the first threshold, a short-term controlled bed leave event is identified as a valid decompression event. When the risk index of leaving the bed and the stress relief liability index are both higher than the corresponding thresholds, a collaborative nursing warning will be issued indicating that the patient needs to be turned over, needs to be left in bed, or needs to be handled by a nurse at the bedside.

[0011] Preferably, the bed return verification unit uses at least two of the following conditions as valid bed return determination criteria: The pressure distribution on the bed surface is restored to the pressure distribution experienced by the human body. The center of pressure is stabilized in the preset lying or sitting position mode; Micro-motion signals and / or respiratory fluctuation signals consistent with those of the human body were detected; Maintain the bed return status until the preset verification time is reached.

[0012] Preferably, the early warning interaction unit outputs at least three levels of early warning, including bedside prompts, nurse-side reminders, and ward-level alarms, and the threshold for time spent away from the bed and the early warning level are higher during nighttime than during daytime.

[0013] Preferably, a remote monitoring method for intelligent nursing beds with early warning includes the following steps: Step 1: Collect information on the patient's bed surface pressure zones, bed posture changes, bedside area information, and ward area information; Step 2: Obtain patient care constraint information and receive or update the corresponding bed leave authorization token; Step 3: Based on the collected information, nursing constraint information, and bed exit authorization token, determine the patient's current nursing status and state. Step 4: Calculate the patient's risk index for leaving the bed and stress-relief debt index; Step 5: Based on the aforementioned bed-leaving risk index and stress-relief liability index, execute conflict resolution decisions to generate corresponding graded early warning results and nursing treatment strategies; Step 6: When a patient is detected returning to their bed, perform a bed return verification. Once the verification is successful, disable the bed exit event, restore monitoring arming, and update the nursing log and subsequent nursing tasks.

[0014] Preferably, in step five, when the patient is in an authorized temporary bed-out state and their bed-out behavior meets at least two of the authorized duration, authorized area, and accompanying conditions, the bed-out event is recorded as an effective decompression event, and the patient's pressure ulcer prevention task time is extended or recalculated based on the effective decompression event; when the conditions are not met, the bed-out event is upgraded to a bed-out timeout warning or a suspected missing person warning.

[0015] This invention provides a remote monitoring system and method for intelligent nursing beds with early warning capabilities. It offers the following advantages: 1. This invention integrates bed surface zone pressure detection, bed posture detection, bedside area detection, ward area detection, nursing restraint information, and authorized bed exit information to form a complete multi-source monitoring input structure, which can continuously identify the patient's bed-in, bed-out, temporary bed-out, and bed-back processes.

[0016] 2. By setting up an exit authorization token, this invention enables nursing staff's authorized operations to enter the monitoring system in the form of structured data and participate in the exit status determination and early warning output process, so that patient authorized exit, exit beyond the time limit, and exit beyond the boundary can be distinguished and processed.

[0017] 3. This invention constructs an out-of-bed risk index and a pressure relief liability index to achieve a unified calculation path for out-of-bed early warning and pressure ulcer pressure relief needs, enabling some authorized out-of-bed behaviors to be identified as effective pressure relief processes, thereby aligning pressure relief management with out-of-bed management.

[0018] 4. By setting up a conflict resolution decision unit, this invention can output different results such as restricting leaving the bed, providing caregiver services, nursing intervention, or stress reduction records based on the current dual-risk status in nursing scenarios where high fall risk and high pressure ulcer risk coexist. This avoids inconsistencies in handling caused by the independent nature of the warning logic.

[0019] 5. By setting up a bed return verification unit, the present invention continuously verifies the patient's status after returning to bed, avoiding the determination of bed return completion when items are placed on the bed, the patient is briefly placed near the bedside, or the patient has not yet formed a stable bed-sitting state. This improves the accuracy of determining the end of the bed-leaving event.

[0020] 6. This invention employs a tiered early warning mechanism that includes bedside prompts, nurse-side alerts, and ward-level alarms, enabling different monitoring states to correspond to different levels of early warning outputs. This facilitates nursing staff in handling and recording events according to their type. Attached Figure Description

[0021] Figure 1 This is a schematic diagram of the functional module composition structure of the server side in this invention; Figure 2 This is a schematic diagram of the state transitions of the nursing state machine in this invention; Figure 3 This is a schematic diagram of the process for issuing and monitoring off-bed authorization information in this invention; Figure 4 This is a schematic diagram of the bed return verification process in this invention; Figure 5 This is a schematic diagram of the disease area monitoring and judgment process in this invention. Detailed Implementation

[0022] The technical solution of the present invention will now be clearly and completely described 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] Please see the appendix Figure 1 To be continued Figure 5 This invention provides a remote monitoring system for intelligent nursing beds with early warning, comprising: a bed body, and a bed surface zone pressure detection unit, a bed posture detection unit, a bedside area perception unit, a ward area perception unit, an authorization token receiving unit, a nursing rule engine, a risk assessment unit, a conflict resolution decision unit, a return-to-bed verification unit, and an early warning interaction unit connected to the bed body; Among them, the bed surface zone pressure detection unit is used to collect information on the zone pressure, pressure center of gravity change, and duration of continuous pressure when the patient is in bed. The bed posture detection unit is used to collect information on the status of the guardrails, the angle of the headboard, and changes in the bed posture. The bedside area sensing unit is used to detect whether a patient is in the bedside activity area; The ward area sensing unit is used to detect whether a patient has left the bedside area or left a preset ward area. The authorization token receiving unit is used to receive the bed exit authorization token issued by the nurse's terminal; The nursing rule engine is used to read patient care constraint information, which includes at least the fall risk level, pressure ulcer risk level, nighttime bed leave restrictions, and nursing orders. The risk assessment unit is used to calculate the bed-leaving risk index and the stress relief liability index based on information such as zoning pressure, bed posture change, bedside area, ward area, bed-leaving authorization token, and nursing constraint information. The conflict resolution decision unit is used to output corresponding graded early warning results and nursing treatment strategies based on the bed closure risk index and the stress reduction liability index; The bed return verification unit is used to verify whether a patient has completed a valid bed return after the patient is detected returning to the bed. The early warning interaction unit is used to send the graded early warning results and nursing treatment strategies to the nursing station and / or mobile nursing terminal.

[0024] Specifically, it also includes a body temperature detection unit, which is installed on the bed itself or in the patient contact area to collect patient body temperature data and transmit it to the risk assessment unit. When performing risk calculations, the risk assessment unit also generates body temperature status parameters based on the temperature data. When the body temperature data exceeds a preset threshold, the abnormal body temperature status is included as an additional risk factor in the early warning classification, and it works in conjunction with the bed occupancy risk index and the stress relief liability index in the early warning interaction unit. In this embodiment, the intelligent nursing bed remote monitoring system with early warning includes the bed itself, a bedside controller, a bed surface zone pressure detection unit, a bed posture detection unit, a bedside area sensing unit, a ward area sensing unit, a communication gateway, a server, an authorization token receiving unit, a nursing rule engine, a risk assessment unit, a conflict resolution decision unit, a bed return verification unit, and an early warning interaction unit. The bed surface zone pressure detection unit, the bed posture detection unit, and the bedside area sensing unit are installed on the bed itself or in an area adjacent to the bed and are electrically connected to the bedside controller. The ward area sensing unit is located at the ward entrance and ward corridor and is communicatively connected to the communication gateway. The communication gateway communicates with the server, which in turn communicates with both the nurse station terminal and the mobile nursing terminal. The authorization token receiving unit, nursing rule engine, risk assessment unit, conflict resolution decision unit, bed return verification unit, and early warning interaction unit are deployed on the server, or within a distributed computing architecture formed by the server and the bedside controller. Status acquisition and preliminary judgment are performed by the bedside controller, while risk calculation, token verification, early warning classification, and log generation are performed by the server.

[0025] Furthermore, the risk index of leaving the bed is determined by any two or more of the following parameters: fall risk level, nighttime, nursing orders, caregiver status, distance from the bed, duration of time away from the bed, changes in the status of the guardrail, and bedside activity trajectory; The decompression liability index is determined by any two or more of the following parameters: duration of local continuous pressure, peak pressure in the zone, range of pressure center shift, frequency of body position changes, duration of pressure relief after leaving the bed, and pressure recovery status after returning to the bed. The bed return verification unit uses at least two of the following conditions as valid bed return criteria: The pressure distribution on the bed surface is restored to the pressure distribution experienced by the human body. The center of pressure is stabilized in the preset lying or sitting position mode; Micro-motion signals and / or respiratory fluctuation signals consistent with those of the human body were detected; Maintain the bed return status until the preset verification time is reached.

[0026] Specifically, in this embodiment, the bedside controller includes a microprocessor, a memory, a clock module, a signal conditioning interface, and a communication interface. The bed surface zone pressure detection unit is connected to the signal conditioning interface via a data acquisition board, while the bed posture detection unit and the bedside area sensing unit are connected to the bedside controller via a serial interface or a digital input interface, respectively. The bedside controller timestamps the data from each sensor, forming a data frame with a unified sampling period, and sends it to the communication gateway via the communication interface. The unified sampling period is set to Δt, and the state update period is set to... Δt is less than or equal to The server updates its status periodically. Perform status determination and early warning output for continuous data frames.

[0027] In this embodiment, the bed surface zone pressure detection unit consists of a pressure array positioned under the mattress or above the bed board. The bed body includes a split mattress structure, with the mattress divided into back area pads, hip area pads, and left and right lower limb area pads. Each area pad is independently installed and removable. Each area pad corresponds to a different detection area of ​​the bed surface zone pressure detection unit, allowing pressure data for each area to be collected and processed independently. When a local pad deforms or its performance changes, the corresponding area pad can be replaced individually. The pressure array is divided into five pressure areas: head and shoulder area, back area, hip and sacral area, thigh area, and calf area. The mattress area corresponding to the hip and sacral area uses a material structure designed to reduce local pressure, thereby improving the pressure distribution in the hip area. The material structure affects the pressure data in the hip and sacral area, and the pressure data is used as one of the input parameters for calculating the pressure relief liability index in the risk assessment process. The bedside controller reads the pressure values ​​of each sensing unit in each sampling cycle and generates the average pressure, peak pressure, pressure area, duration of continuous pressure, and pressure center coordinates for each pressure area. The pressure centroid coordinates are used to characterize the patient's center of load position on the bed surface, and the duration of continuous pressure is used to characterize the length of time that the corresponding area is continuously under pressure. The bedside controller combines the features of the five pressure areas to form the bed surface pressure vector at the current moment, which is used for subsequent bed status determination, bed edge transition determination, pressure ulcer risk calculation, and return-to-bed verification.

[0028] In this embodiment, the bed posture detection unit includes a guardrail status detector, a backrest angle detector, and a legboard angle detector. The guardrail status detector is installed at the rotating connection of the guardrail and outputs the guardrail raised and lowered states. The backrest angle detector and legboard angle detector are installed at the backrest hinge and legboard hinge, respectively, and output the backrest angle α(k) and legboard angle β(k). Here, k represents the k-th state update cycle. The backrest angle detector and legboard angle detector include angle measurement structures to acquire the backrest elevation angle and lower limb elevation angle; these angles are used to determine whether the patient has met the preset nursing posture requirements and serve as input parameters for the nursing state machine's state determination and risk assessment unit. The backrest angle α(k), legboard angle β(k), changes in guardrail status, and the lateral displacement of the bed surface pressure center of gravity are used together to identify whether the patient has transitioned from a supine position to a sitting position and from a sitting position to a bedside transition position.

[0029] In this embodiment, the bedside area sensing unit is located at the side or foot of the bed to detect whether the patient is within the near-field activity range of the bedside. The bedside area sensing unit outputs a bedside occupancy signal. When the patient is within a pre-defined area next to the bed, =1; When the patient is not within the preset bedside area =0. The ward area sensing unit includes a doorway positioning node installed at the ward entrance and a ward identification tag installed on the patient's wristband. Short-range identification is performed between the doorway positioning node and the patient's wristband to obtain the patient's ward area identifier Q(k). The ward identifier Q(k) includes at least the bedside area, the ward interior area, the doorway area, and the area outside the ward. Bedside occupancy signal. Together with the area identifier Q(k), it is used to determine whether a patient has only moved a short distance from the bed, whether they have left the ward, or whether they have entered an area outside the authorized zone.

[0030] Furthermore, the bed leave authorization token shall include at least one or more of the following: patient identifier, bed identifier, authorization start time, permitted bed leave duration, permitted activity area, accompanying status, time period strategy, and bed leave purpose type.

[0031] Specifically, in this embodiment, the authorization token receiving unit receives a bed-leaving authorization token issued by the nurse station terminal or mobile nursing terminal. The bed-leaving authorization token is denoted as A, and its data fields include the patient identifier. Bed signage Authorization start time Permissible time away from bed Permitted activity areas Requirements for accompanying persons Time-of-use strategy and types of bed separation The authorization deadline is recorded as follows: ,and Patient identification Bed identification A one-to-one correspondence is established, and the bed leave authorization token is only valid for the specified patient in the specified bed. The authorization token receiving unit verifies whether the bed leave authorization token satisfies the requirements of time validity, region validity, and accompanying person validity during each state update cycle. Time validity means that the current time t(k) satisfies... The valid representation of a region indicates that the region identifier Q(k) belongs to a permitted activity region. Effective caregiving indicates that the caregiving status is satisfactory. The corresponding constraints.

[0032] Furthermore, the conflict resolution decision unit constructs a patient care state machine, which includes at least the following states: bed safety state, bedside transition state, unauthorized bed leave state, authorized temporary bed leave state, bed leave timeout state, suspected lost state, effective bed return state, and ineffective bed return state. The conflict resolution decision unit is configured as follows: When the risk index of leaving the bed is higher than the first threshold and the pressure relief debt index is lower than the second threshold, an early warning to restrict leaving the bed is output. When the decompression liability index is higher than the second threshold and the bed leave risk index is lower than the first threshold, a short-term controlled bed leave event is identified as a valid decompression event. When the risk index of leaving the bed and the stress relief liability index are both higher than the corresponding thresholds, a collaborative nursing warning will be issued indicating that the patient needs to be turned over, needs to be left in bed, or needs to be handled by a nurse at the bedside.

[0033] Specifically, in this embodiment, the nursing rule engine is used to read patient nursing constraint information H. Nursing constraint information H includes the fall risk level L. f Pressure ulcer risk level L p Nighttime monitoring time window W n Nursing doctor's orders constraint O m Accompanying care requirements R c And the warning threshold group Θ. Fall risk level L f Pressure ulcer risk level L p The existing assessment results in the nursing system were normalized, with values ​​ranging from 0 to 1. Nighttime monitoring time window W n Used to determine whether the current time falls within the nighttime restricted monitoring period. Nursing doctor's orders constraint O m It includes at least four types: absolute bed rest, bedside activities, caregiver leaving the bed, and free leaving the bed, and converts these into constraint factors. Caregiver requirements R c This is used to determine whether a patient is allowed to leave the bed without a caregiver. The warning threshold group Θ includes the bed-leaving risk threshold Θ. FDebt reduction threshold Θ P Conflict handling threshold Θ H and Θ M and the return-to-bed verification threshold Θ R The nursing rule engine associates nursing constraint information H with authorization token A to form the monitoring rule context for the current patient.

[0034] In this embodiment, the risk assessment unit calculates the bed-leaving risk index F(k) and the pressure relief liability index P(k) in each status update cycle. The bed-leaving risk index F(k) characterizes the risks of falls, getting lost, and medical order violations corresponding to the current bed-leaving behavior, while the pressure relief liability index P(k) characterizes the degree of pressure ulcer risk accumulation corresponding to the current pressure state. The bed-leaving risk index F(k) is determined by the fall risk level L. f The factors used to determine the nighttime factor N(k), the doctor's order constraint factor O(k), the bed distance factor D(k), the caregiver compliance factor C(k), and the authorization abnormality factor E(k) are all included. Among them, the nighttime factor N(k) is located within the nighttime monitoring time window W at the current time. n The value is 1 if the condition is met, otherwise 0; the doctor's order constraint factor O(k) is based on the nursing doctor's order constraint O. m The following parameters are defined: The bed-leaving distance factor D(k) is determined based on the area identifier Q(k); the caregiver compliance factor C(k) is 1 if the caregiver status meets the requirements, otherwise it is 0; the authorization anomaly factor E(k) is 1 if the bed-leaving authorization token is missing, expired, the area is out of bounds, or the bed-leaving time is exceeded, otherwise it is 0. The bed-leaving risk index F(k) is generated using a weighted summation and normalization method, with a larger value indicating a higher bed-leaving risk.

[0035] In this embodiment, the decompression liability index P(k) is jointly determined by the local continuous pressure load S(k), the local peak pressure load V(k), the insufficient positional change M(k), and the effective decompression amount R(k). The local continuous pressure load S(k) is generated based on the number of regions exceeding the continuous pressure threshold and the corresponding duration in the five pressure regions; the local peak pressure load V(k) is generated based on the number of regions exceeding the peak pressure threshold and the over-threshold amplitude in the five pressure regions; the insufficient positional change M(k) is generated based on the magnitude of pressure center shift and bed angle change within a preset time window; and the effective decompression amount R(k) is generated based on the duration of the patient's authorized effective state of being out of bed. The decompression liability index P(k) is calculated using a cumulative update method, where continuous pressure and peak pressure increase P(k), and the effective decompression amount R(k) decreases P(k), with P(k) limited to the range of 0 to 1.

[0036] In this embodiment, to enable authorized bed leave events to participate in pressure ulcer risk calculation, the authorization token receiving unit generates an authorization validity coefficient Γ(k) in each state update cycle, the calculation formula of which is: In the formula, T(k) is the time validity flag, which is 1 when the current time is within the authorized time period, and 0 otherwise; Z(k) is the area validity flag, which is the flag when the area identifier Q(k) belongs to the permitted activity area Ω. a Z(k) = 1 when Z(k) is active, otherwise Z(k) = 0; C v (k) is a valid indicator of caregiving, when the caregiving status meets the requirements. v (k)=1, otherwise C v Γ(k) = 0. The value of Γ(k) ranges from 0 to 1. Γ(k) = 1 indicates that the time, region, and accompanying conditions are all met; Γ(k) = 0 indicates that the behavior of leaving the bed is not authorized.

[0037] In this embodiment, the conflict resolution decision-making unit generates a collaborative disposal index Ξ(k) based on the bedside risk index F(k), the pressure reduction liability index P(k), and the authorization effectiveness coefficient Γ(k). The calculation formula is as follows: In the formula, κ is the decompression compensation coefficient, with a value ranging from 0 to 1. In this formula, F(k) represents the direct risk of the current bed-leaving behavior, and P(k)·Γ(k) represents the decompression compensation amount that can be included under the condition of effective authorization. Since the effective authorization coefficient Γ(k) participates in the product term, when the patient's bed-leaving behavior does not meet the requirements of authorized time, authorized area, or caregiver, the decompression liability index P(k) does not participate in risk offsetting, and the collaborative treatment index Ξ(k) is equal to the bed-leaving risk index F(k); when the patient's bed-leaving behavior meets the authorization conditions and the decompression liability index is high, the collaborative treatment index Ξ(k) decreases. The conflict resolution decision unit combines the collaborative treatment index Ξ(k) with the threshold Θ. H Θ M For comparison, Ξ(k)≥Θ H Output a warning message that limits bed leaving. M ≤Ξ(k)<Θ H Timely output of caregiving reminders, Ξ(k)<Θ M And P(k)≥Θ P The current bed exit event is recorded as a valid decompression event and authorized bed exit monitoring is maintained.

[0038] In this embodiment, the system maintains a nursing state machine S(k). The nursing state machine includes bed safety state S0, bedside transition state S1, unauthorized bed leaving state S2, authorized temporary bed leaving state S3, bed leaving timeout state S4, suspected lost state S5, valid bed return state S6, and invalid bed return state S7. The bedside controller and server update the system in each state update cycle based on the bed surface pressure vector, bed back angle α(k), legboard angle β(k), and bedside occupancy signal B. bThe nursing status machine is updated using the following parameters: Q(k), ...

[0039] In this embodiment, when the total pressure on the bed surface exceeds the in-bed threshold and the center of pressure is located in the central area of ​​the bed surface, the nursing state machine enters the in-bed safety state S0. When the center of pressure continues to shift towards the edge of the bed, and the backrest angle α(k) exceeds the sit-up angle threshold or the guardrail is in the lowered state, the nursing state machine switches from the in-bed safety state S0 to the bed edge transition state S1. When the total pressure on the bed surface is lower than the leave-bed threshold, and the bedside occupancy signal B... b When Γ(k)=1 and Γ(k)=0, the nursing state machine switches to the unauthorized bed leave state S2. When the total pressure on the bed surface is lower than the bed leave threshold and Γ(k)>0, the nursing state machine switches to the authorized temporary bed leave state S3. When the duration of the authorized temporary bed leave state S3 exceeds the allowed bed leave duration τ... a Or, the area identifier Q(k) does not belong to the permitted activity area Ω. a Or the caregiver status does not meet C a The nursing state machine switches to the bed-out timeout state S4. When the area identifier Q(k) remains outside the ward or the authorization expires and the patient does not return to the ward within the preset time limit, the nursing state machine switches from the bed-out timeout state S4 to the suspected lost state S5.

[0040] In this embodiment, when the nursing status machine is in the unauthorized bed-avoidance state S2, authorized temporary bed-avoidance state S3, bed-avoidance timeout state S4, or suspected missing state S5, and the total pressure on the bed surface recovers to above the in-bed threshold, the system does not directly switch to the in-bed safety state S0, but instead first enters the bed-return verification process. The bed-return verification unit determines the bed-return verification based on the pressure distribution matching degree Y. p (k) Center of gravity stability Y g (k) Presence of micromotion Y m (k) and duration conformity Y t (k) Generate bed return verification value R v (k), its calculation formula is: In the formula, a1, a2, a3, and a4 are non-negative weighting coefficients, and a1 + a2 + a3 + a4 = 1. Pressure distribution matching degree Y p (k) represents the similarity between the current bed surface pressure vector and the patient's standard bed position vector; center of gravity stability Y g (k) indicates whether the pressure center fluctuation is within the threshold range within the preset verification time window; the degree of micro-motion Y m (k) indicates whether micro-movements or respiratory fluctuations consistent with human characteristics in bed were detected within the preset time window; duration of consistency Y t(k) indicates whether the current bed return status continues to reach the bed return verification time threshold. When R v (k)≥Θ R When, the nursing state machine switches to the effective return-to-bed state S6; when R v (k)<Θ R When the nursing state machine switches to the invalid return-to-bed state S7, the system closes the current bed-leaving event, resumes in-bed safety monitoring, and updates the nursing log. When the nursing state machine is in the invalid return-to-bed state S7, the system maintains the bed-leaving event in an unfinished state and continues to perform area monitoring and early warning judgment.

[0041] In this implementation, the system operates in the following sequence: continuous monitoring, rule matching, risk calculation, status update, conflict resolution, return-to-bed verification, and log generation. The bedside controller first collects data on bed surface pressure array, bed posture, bedside occupancy, and area identification, performing filtering, de-jittering, and timestamping in each sampling cycle. After receiving a unified data frame, the server calls the nursing rule engine to read the current patient's nursing constraint information H and the authorization token A. The risk assessment unit calculates F(k), P(k), and Γ(k) based on the current data frame and rule context, while the conflict resolution decision unit generates a resolution result based on Ξ(k) and the current nursing state machine status. Resolution results include continued monitoring, bedside prompts, nurse-side reminders, ward-level alarms, reminders for accompanying persons to be present, reminders for turning over, and checks on return-to-bed status. The system executes the above process in each status update cycle, ensuring that the risks of leaving the bed, pressure ulcers, and authorization conditions are processed within the same decision path.

[0042] Furthermore, a remote monitoring method for intelligent nursing beds with early warning capabilities includes the following steps: Step 1: Collect information on the patient's bed surface pressure zones, bed posture changes, bedside area information, and ward area information; Step 2: Obtain patient care constraint information and receive or update the corresponding bed leave authorization token; Step 3: Based on the collected information, nursing constraint information, and bed exit authorization token, determine the patient's current nursing status and state. Step 4: Calculate the patient's risk index for leaving the bed and stress-relief debt index; Step 5: Based on the bed exit risk index and stress reduction liability index, execute conflict resolution decisions and generate corresponding graded early warning results and nursing treatment strategies; Step 6: When a patient is detected returning to their bed, perform a bed return verification. Once the verification is successful, disable the bed exit event, restore monitoring arming, and update the nursing log and subsequent nursing tasks.

[0043] In this embodiment, the monitoring method includes the following steps: First, it collects information on the patient's bed surface pressure zones, bed posture changes, bedside area information, and ward area information, and performs time synchronization processing on the monitoring information from each source; then, it reads the patient's nursing restraint information H and receives the currently active bed-leaving authorization token A; subsequently, it calculates the total pressure on the bed surface, the position of the pressure center of gravity, the bed back angle α(k), and the bedside occupancy signal B. b The patient's current nursing status (S(k)) is determined by the region identifier Q(k) and the authorization validity coefficient Γ(k). After the status is determined, the bed abandonment risk index F(k) and the stress relief liability index P(k) are calculated, and the collaborative treatment index Ξ(k) is further generated. Then, the collaborative treatment index Ξ(k) is compared with the threshold Θ. H Θ M The system compares data to generate corresponding tiered early warning results and nursing intervention strategies. When a new distribution of human pressure is detected on the bed surface, the bed return verification process is initiated, and the bed return verification value R is calculated. v (k), when R v (k) reaches the threshold Θ R End the bed exit event and update the nursing task immediately; otherwise, maintain alert monitoring.

[0044] In this embodiment, the data acquisition steps include zero-point correction and outlier removal of the raw data from the bed surface pressure array, jitter reduction processing of the guardrail status and angle signals, and time continuity verification of the bedside occupancy signals and area markers. The status determination steps include bed determination, bedside transition determination, bed exit determination, authorized bed exit determination, time-out bed exit determination, and suspected wandering determination. The risk calculation steps include calculating the bed exit risk index F(k), the pressure reduction liability index P(k), and the authorized effectiveness coefficient Γ(k). The conflict resolution steps include generating a bed exit restriction warning, a caregiver intervention reminder, or a record of an effective pressure reduction event based on Γ(k). The bed return verification steps include reading the current bed surface pressure vector and calculating the pressure distribution matching degree Y. p (k) Calculate the centroid stability Y g (k) Identifying the presence of micro-motion Y m (k) Statistical Duration Consistency Y t (k), and according to R v (k) Complete the determination of effective and ineffective bed return.

[0045] Furthermore, in step five, when the patient is in an authorized temporary bed-out state and their bed-out behavior meets at least two of the authorized duration, authorized area, and accompanying conditions, the bed-out event is recorded as an effective decompression event, and the patient's pressure ulcer prevention task time is extended or recalculated based on the effective decompression event; when the conditions are not met, the bed-out event is upgraded to a bed-out timeout warning or a suspected wandering warning.

[0046] In this embodiment, when the patient is in an authorized temporary bed-off state S3, and Γ(k)=1, and P(k)≥Θ P When the duration of bed leave falls within the preset effective decompression time window, the system records this bed leave event as an effective decompression event. The effective decompression event is written to the nursing log and used to correct the update result of the decompression liability index P(k). The correction method is as follows: starting from the next status update cycle after the effective decompression event ends, P(k) is subtracted from the decompression amount R(k) corresponding to the duration of this effective decompression event, and the next turning reminder time is recalculated. This process is only performed when the bed leave authorization token is valid and the bed leave behavior meets the requirements of the allowed activity area and caregiver status; when the authorization time expires, the area is out of bounds, or the caregiver fails, the current bed leave event is not counted as an effective decompression event.

[0047] Furthermore, the early warning interaction unit outputs at least three levels of early warning, including bedside prompts, nurse-side reminders, and ward-level alarms, with the threshold for time spent away from the bed and the warning level being higher during the night than during the day.

[0048] Specifically, in this embodiment, the early warning interaction unit adopts a three-level early warning output mechanism. The first-level early warning is a bedside prompt, and its triggering conditions include the nursing state machine entering the bedside transition state S1 and the authorized temporary bed leave state S3 approaching the end of the authorization time t. e Alternatively, the stress-relief debt index P(k) reaches the turning-over reminder threshold but not the nurse-side reminder threshold. Bedside prompts are output via the bedside audio-visual module, and the prompt information includes the bed identifier, current status, and remaining allowed time to leave the bed. The second-level warning is a nurse-side reminder, triggered by conditions including the nursing status machine entering the unauthorized bed leaving state S2, the remaining authorized time under the authorized temporary bed leaving state S3 being lower than the reminder threshold, the stress-relief debt index P(k) reaching the nursing intervention threshold, or the return-to-bed verification failing. The nurse-side reminder is sent to the nurse station terminal and mobile nursing terminal, and the reminder content includes the patient identifier, bed identifier, current status, cumulative bed leaving time, F(k), P(k), Γ(k), and suggested treatment type. The third-level warning is a ward-level alarm, triggered by conditions including the nursing status machine entering the bed leaving timeout state S4, the suspected missing state S5, a bed leaving event occurring in a patient on absolute bed rest at night, or the unauthorized bed leaving state S2 continuing for more than the ward-level alarm delay. Ward-level alarms are displayed via nurse station pop-ups, on-duty terminal push notifications, and ward audio-visual modules. Alarm content includes patient identification, bed identification, current area identification Q(k), time spent away from bed, reason for authorization failure, and current warning level.

[0049] In this implementation, the outputs of the first-level, second-level, and third-level warnings do not overlap but are recorded sequentially according to time and level. The warning interaction unit generates an event number for each bed leaving event and associates the event number with the authorization token A, nursing state machine sequence, bed leaving risk index F(k), stress reduction liability index P(k), collaborative treatment index Ξ(k), and return-to-bed verification value R. v (k) The system associates and stores the handling results. After an out-of-bed event, the system generates a nursing log, which includes at least the start time, end time, type of reason for out-of-bed event, authorization status, boundary violation, timeout, return-to-bed verification result, and warning output record. The nursing log is used for subsequent nursing verification and parameter tuning, but does not change the generated out-of-bed event judgment result.

[0050] 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. A remote monitoring system for intelligent nursing beds with early warning capabilities, comprising: The system comprises a hospital bed body, and connected to the hospital bed body are a bed surface zone pressure detection unit, a bed posture detection unit, a bedside area perception unit, a ward area perception unit, an authorization token receiving unit, a nursing rule engine, a risk assessment unit, a conflict resolution decision unit, a return-to-bed verification unit, and an early warning interaction unit; characterized in that... The bed surface zone pressure detection unit is used to collect information on the zone pressure, pressure center of gravity changes, and duration of continuous pressure when the patient is in bed. The bed posture detection unit is used to collect information on the guardrail status, bed back angle, and bed posture changes. The bedside area sensing unit is used to detect whether the patient is in the bedside activity area; The ward area sensing unit is used to detect whether a patient has left the bedside area or left the preset ward area. The authorization token receiving unit is used to receive the bed exit authorization token issued by the nurse's terminal; The nursing rule engine is used to read patient nursing constraint information, which includes at least fall risk level, pressure ulcer risk level, nighttime bed leave restrictions, and nursing orders. The risk assessment unit is used to calculate the bed-leaving risk index and the stress relief liability index based on the zoning pressure information, bed posture change information, bedside area information, ward area information, bed-leaving authorization token, and nursing constraint information. The conflict resolution decision unit is used to output corresponding graded early warning results and nursing treatment strategies based on the bed closure risk index and the pressure reduction liability index; The bed return verification unit is used to verify whether the patient has completed a valid bed return after the patient is detected to have returned to the bed. The early warning interaction unit is used to send the graded early warning results and nursing treatment strategies to the nurse station and / or mobile nursing terminal.

2. The intelligent nursing bed remote monitoring system with early warning as described in claim 1, characterized in that, The leave authorization token includes at least one or more of the following: patient identifier, bed identifier, authorization start time, allowed leave duration, allowed activity area, accompanying person status, time period strategy, and leave purpose type.

3. The intelligent nursing bed remote monitoring system with early warning as described in claim 1, characterized in that, The conflict resolution decision unit constructs a patient care state machine, which includes at least the following states: bed safety state, bedside transition state, unauthorized bed leave state, authorized temporary bed leave state, bed leave timeout state, suspected lost state, valid bed return state, and invalid bed return state.

4. The intelligent nursing bed remote monitoring system with early warning as described in claim 1, characterized in that, The risk index for leaving the bed is determined by any two or more of the following parameters: fall risk level, nighttime, nursing orders, caregiver status, distance from the bed, duration of time away from the bed, changes in the status of the guardrail, and bedside activity trajectory.

5. The intelligent nursing bed remote monitoring system with early warning as described in claim 1, characterized in that, The pressure relief liability index is determined by any two or more of the following parameters: duration of local continuous pressure, peak pressure in the zone, range of pressure center shift, frequency of body position changes, duration of pressure relief after leaving the bed, and pressure recovery status after returning to the bed.

6. The intelligent nursing bed remote monitoring system with early warning as described in claim 1, characterized in that, The conflict resolution decision unit is configured as follows: When the risk index of leaving the bed is higher than the first threshold and the pressure relief debt index is lower than the second threshold, an early warning to restrict leaving the bed is output. When the decompression liability index is higher than the second threshold and the bed leave risk index is lower than the first threshold, a short-term controlled bed leave event is identified as a valid decompression event. When the risk index of leaving the bed and the stress relief liability index are both higher than the corresponding thresholds, a collaborative nursing warning will be issued indicating that the patient needs to be turned over, needs to be left in bed, or needs to be handled by a nurse at the bedside.

7. The intelligent nursing bed remote monitoring system with early warning as described in claim 1, characterized in that, The bed return verification unit uses at least two of the following conditions as valid bed return determination criteria: The pressure distribution on the bed surface is restored to the pressure distribution experienced by the human body. The center of pressure is stabilized in the preset lying or sitting position mode; Micro-motion signals and / or respiratory fluctuation signals consistent with those of the human body were detected; Maintain the bed return status until the preset verification time is reached.

8. The intelligent nursing bed remote monitoring system with early warning as described in claim 1, characterized in that, The early warning interaction unit outputs at least three levels of early warning, including bedside prompts, nurse-side reminders, and ward-level alarms, with the threshold for time spent out of bed and the warning level being higher at night than during the day.

9. A remote monitoring method for intelligent nursing beds with early warning, characterized in that, A remote monitoring system for intelligent nursing beds with early warning as described in any one of claims 1-8 includes the following steps: Step 1: Collect information on the patient's bed surface pressure zones, bed posture changes, bedside area information, and ward area information; Step 2: Obtain patient care constraint information and receive or update the corresponding bed leave authorization token; Step 3: Based on the collected information, nursing constraint information, and bed exit authorization token, determine the patient's current nursing status and state. Step 4: Calculate the patient's risk index for leaving the bed and stress-relief debt index; Step 5: Based on the aforementioned bed-leaving risk index and stress-relief liability index, execute conflict resolution decisions to generate corresponding graded early warning results and nursing treatment strategies; Step 6: When a patient is detected returning to their bed, perform a bed return verification. Once the verification is successful, disable the bed exit event, restore monitoring arming, and update the nursing log and subsequent nursing tasks.

10. A remote monitoring method for an intelligent nursing bed with early warning as described in claim 9, characterized in that, In step five, when a patient is in an authorized temporary bed-out state and their bed-out behavior meets at least two of the authorized duration, authorized area, and caregiver conditions, the bed-out event is recorded as an effective decompression event, and the patient's pressure ulcer prevention task time is extended or recalculated based on the effective decompression event. If the conditions are not met, the event of leaving the bed will be upgraded to a warning of timeout or a warning of suspected wandering.

Images