Search

CN-122004831-A - Intelligent reminding system and method for preventing pressure sores

CN122004831ACN 122004831 ACN122004831 ACN 122004831ACN-122004831-A

Abstract

The invention discloses an intelligent reminding system and method for preventing pressure sores, and relates to the technical fields of medical health and Internet of things. The method comprises the steps of collecting pressure distribution data, local microenvironment data and physiological time sequence data of a target object in real time, combining static attribute information and historical intervention records, generating a dynamic pressure sore risk index and a predictive intervention time point by utilizing a personalized dynamic risk assessment model, carrying out layered decision according to the risk index, the predictive time point and the current state of the target object, generating a personalized intervention prompt scheme, initiating a graded prompt or intervention instruction to the target object and/or a nursing terminal, and recording feedback data and optimizing the model after an effective pressure release event is monitored. The system comprises a multi-mode sensing module, a data processing and communication module, an intelligent analysis decision module, a reminding and interaction module and a feedback learning module, so that accurate, personalized and self-adaptive pressure sore prevention is realized, nursing burden is reduced, and comfort and safety of patients are improved.

Inventors

  • WANG NING

Assignees

  • 西安交通大学医学院第一附属医院

Dates

Publication Date
20260512
Application Date
20260318

Claims (10)

  1. 1. The intelligent reminding method for preventing pressure sores is characterized by comprising the following steps of: s1, acquiring multi-mode monitoring data of a target object in real time, wherein the multi-mode monitoring data at least comprises pressure distribution data and local microenvironment data which are acquired by a sensing array paved on a supporting surface; S2, calculating through a personalized dynamic risk assessment model based on the multi-mode monitoring data, static attribute information of a target object and a historical intervention record, and generating a dynamic pressure sore risk index and a predictive intervention time point at the current moment; s3, performing hierarchical decision according to the dynamic pressure sore risk index, the predictive intervention time point and the current state of the target object estimated by physiological data, and generating a personalized intervention prompt scheme corresponding to the decision level; s4, according to the personalized intervention prompt scheme, prompting or intervention instructions of corresponding levels are initiated to a target object and/or a nursing terminal; and S5, after an effective pressure release event is monitored, recording validity data of the intervention, and feeding back the validity data to the personalized dynamic risk assessment model for self-adaptive optimization of the model.
  2. 2. The intelligent reminding method for preventing pressure sores according to claim 1, wherein in the step S1, the local microenvironment data comprises contact interface temperature and/or humidity, the multi-mode monitoring data further comprises physiological time sequence data acquired through wearable equipment, and the physiological time sequence data comprises at least one of heart rate, heart rate variability and body movement data.
  3. 3. The intelligent reminding method for preventing pressure sores according to claim 1 or 2, wherein in step S2, the personalized dynamic risk assessment model is a time sequence deep learning model, and the input features at least include: time sequence characteristics extracted from the pressure distribution data, including pressure intensity, pressure gradient and duration of the specific region; The time sequence change characteristics of the local micro-environment data; the static attribute information comprises age, weight and skin condition score; time, frequency, and effectiveness of historical pressure relief events.
  4. 4. The intelligent warning method for pressure sore prevention according to claim 2, wherein in step S3, the current state of the target subject includes sleep stage or awake/sleep state, which is presumed by analyzing body movement data, heart rate and heart rate variability data in the physiological time series data, and the logic of the hierarchical decision includes: when the dynamic pressure sore risk index is lower than a first threshold value or the target object is in a deep sleep period, delaying or temporarily not initiating active reminding; When the dynamic pressure sore risk index is between a first threshold value and a second higher threshold value and the target object is in a light sleep or awake state, a first-level reminder for guiding autonomous inching is preferentially initiated to the target object end; and when the dynamic pressure sore risk index exceeds a second threshold value or the first-level reminding is invalid, initiating a second-level alarm containing a specific posture adjustment suggestion to the nursing terminal.
  5. 5. The intelligent reminding method for preventing pressure sores according to claim 1, wherein in the step S4, the reminding mode initiated to the target object end comprises gradual lamplight, mild vibration or voice guidance, and the intervention instruction initiated to the nursing terminal comprises a recommended turning body position, an angle and a body part identifier needing key decompression.
  6. 6. An anti-pressure sore intelligent reminding system for implementing the method of any one of claims 1-4, comprising: the multi-mode sensing module is used for collecting pressure distribution data and local microenvironment data of a target object in real time; the data processing and communication module is used for preprocessing and transmitting the data acquired by the multi-mode sensing module; An intelligent analysis decision module comprising a processor storing a personalized dynamic risk assessment model for performing the calculation and decision process of steps S2 and S3 of claim 1; the reminding and interaction module is used for executing the reminding operation of the step S4 in the claim 1 according to the output of the intelligent analysis decision module; and the feedback learning module is used for recording the pressure release event data of the dry and dry state and carrying out optimization updating on the personalized dynamic risk assessment model.
  7. 7. The pressure sore prevention intelligent reminding system of claim 6, wherein the multi-modal sensing module comprises: The flexible pressure distribution sensing array is embedded into the mattress or the cushion to form a pressure lattice; The temperature and humidity sensors are distributed among the nodes of the pressure distribution sensing array; And an optional bioimpedance sensing unit for monitoring local tissue electrical impedance changes.
  8. 8. The pressure sore prevention intelligent reminding system according to claim 6, further comprising a physiological monitoring module, which is a stand-alone wearable device or a contact sensor integrated in a mattress, for collecting heart rate, heart rate variability and body movement data and transmitting the physiological time series data to the intelligent analysis decision module.
  9. 9. The intelligent pressure sore prevention reminding system according to claim 6, wherein the intelligent analysis decision module adopts a cloud-edge cooperative architecture: The data processing and communication module comprises an edge computing unit which is used for carrying out local data fusion, primary risk judgment and sending out a local alarm when the risk exceeds the limit; the personalized dynamic risk assessment model is deployed on a cloud server and is used for receiving data uploaded by edges and carrying out depth feature fusion, model calculation and long-term data storage.
  10. 10. The pressure sore prevention intelligent reminding system according to claim 6, wherein the reminding and interaction module comprises: The patient end interaction unit is integrated on the bed body or is independently arranged and comprises a light array, a vibration motor and/or a loudspeaker; The nursing end interaction unit provides a graphical interface for displaying a risk thermodynamic diagram, a risk index trend, a predictive intervention time point and a specific turning scheme for the mobile terminal or the fixed workstation.

Description

Intelligent reminding system and method for preventing pressure sores Technical Field The invention relates to the technical fields of medical health and Internet of things, in particular to an intelligent reminding system and method for preventing pressure sores. Background Pressure sores, also known as pressure injuries, are common serious complications of long-term bedridden or sedentary patients, mainly caused by long-term compression of body local tissues. The existing prevention method mainly relies on the nursing staff to manually turn over at regular time, and has the following defects: the rigidity is fixed, and the individual difference and the real-time physiological state change of the patient are ignored by the unified turning interval. Passive response, the early warning can not be carried out before the pressure really reaches the dangerous threshold. Increasing the burden that frequent invalid reminders or turning over can interfere with patient rest and increase the workload of nursing staff. Lack of quantification, relying on subjective experience, lack of continuous monitoring of objective data such as stress, microenvironment, etc. In the prior art, some intelligent mattresses monitor pressure distribution through a pressure lattice, but most of the mattresses only perform static threshold alarming, cannot be deeply combined with physiological signals, body position comfort level and historical data of patients, and have limited intelligent degree and insufficient early warning accuracy. Disclosure of Invention Aiming at the defects of the prior art, the invention provides an intelligent system and a method which can evaluate the risk of pressure sores in real time and in multiple dimensions, dynamically generate the optimal turn-over reminding time and scheme based on a personalized model, and realize the radical transition from timing to on-demand prevention. In order to achieve the purpose, the invention provides the following technical scheme that the intelligent reminding method for preventing pressure sores comprises the following steps: s1, acquiring multi-mode monitoring data of a target object in real time, wherein the multi-mode monitoring data at least comprises pressure distribution data and local microenvironment data which are acquired by a sensing array paved on a supporting surface; S2, calculating through a personalized dynamic risk assessment model based on the multi-mode monitoring data, static attribute information of a target object and a historical intervention record, and generating a dynamic pressure sore risk index and a predictive intervention time point at the current moment; s3, performing hierarchical decision according to the dynamic pressure sore risk index, the predictive intervention time point and the current state of the target object estimated by physiological data, and generating a personalized intervention prompt scheme corresponding to the decision level; s4, according to the personalized intervention prompt scheme, prompting or intervention instructions of corresponding levels are initiated to a target object and/or a nursing terminal; and S5, after an effective pressure release event is monitored, recording validity data of the intervention, and feeding back the validity data to the personalized dynamic risk assessment model for self-adaptive optimization of the model. Preferably, in step S1, the local microenvironment data comprises contact interface temperature and/or humidity, the multi-modal monitoring data further comprises physiological time series data acquired through the wearable device, and the physiological time series data comprises at least one of heart rate, heart rate variability and body movement data. Preferably, in step S2, the personalized dynamic risk assessment model is a time-series deep learning model, and the input features thereof at least include time-series features extracted from pressure distribution data, including pressure intensity, pressure gradient and duration of a specific region, time-series variation features of the local microenvironment data, static attribute information including age, weight and skin condition score, and time, frequency and validity record of historical pressure release events. Preferably, in step S3, the current state of the target object includes a sleep stage or awake/sleep state, which is presumed by analyzing body movement data, heart rate and heart rate variability data in the physiological time sequence data, and the logic of the hierarchical decision includes delaying or temporarily not initiating an active alert when the dynamic pressure sore risk index is lower than a first threshold value or the target object is in a deep sleep stage, preferentially initiating a first-level alert for guiding autonomous jog to the target object when the dynamic pressure sore risk index is between the first threshold value and a higher second threshold value and the target object is in a light slee