CN-122025066-A - Intelligent monitoring nursing bed control method based on behavior habit and physiological data fusion

CN122025066ACN 122025066 ACN122025066 ACN 122025066ACN-122025066-A

Abstract

The application discloses an intelligent monitoring nursing bed control method based on behavior habit and physiological data fusion, which comprises the steps of obtaining behavior habit data and physiological monitoring data of a target user, constructing a plurality of user state units based on the behavior habit data and the physiological monitoring data, identifying time attribute categories of each user state unit, carrying out comfort and health risk assessment according to a preset first control factor by taking each user state unit as a basic unit to obtain a corresponding health risk index, calculating a nursing intervention level, calculating nursing response priority data according to a preset second control factor, and generating a nursing bed control instruction with an adaptive time period characteristic. The application can sense the user state more comprehensively and respond to individual demands more flexibly.

Inventors

LI ZHIYE
LI ZHIFA
LU CHUNXIAN
LIANG CHAOHUI

Assignees

广东中匠福健康产业股份有限公司

Dates

Publication Date: 20260512
Application Date: 20260131

Claims (10)

1. The intelligent monitoring nursing bed control method based on the integration of behavior habit and physiological data is characterized by comprising the following steps: Acquiring behavior habit data and physiological monitoring data of a target user; Constructing a plurality of user state units based on the behavior habit data and the physiological monitoring data, and identifying a time attribute category of each user state unit, wherein the time attribute category comprises a sleep period and a non-sleep period; Taking each user state unit as a basic unit, performing comfort and health risk assessment according to a preset first control factor to obtain a corresponding health risk index, and calculating a corresponding nursing intervention level according to the health risk index and the time attribute type; taking each user state unit as a basic unit, and calculating nursing response priority data according to a preset second control factor; And generating nursing bed control instructions adapting to time period characteristics based on the nursing intervention level, the nursing response priority data and the time attribute category corresponding to each user state unit.
2. The method of claim 1, wherein constructing a plurality of user state units based on the behavioral habit data and physiological monitoring data and identifying a category of time attribute for each of the user state units comprises: Performing data cleaning and time alignment on the behavior habit data and the physiological monitoring data to obtain processed multisource fusion user data; Performing sliding window segmentation and time sequence clustering based on the processed multi-source fusion user data to generate an initial user state sequence; And extracting state boundary characteristics according to the initial user state sequence, and determining whether each user state unit belongs to a sleep period or not through a sleep state judging model by combining heart rate variability, body movement frequency, breathing regularity and historical work and rest regularity, so as to determine a time attribute type.
3. The method according to claim 1, wherein the step of performing comfort and health risk assessment based on the preset first control factor by using each user status unit as a basic unit to obtain a corresponding health risk index, and calculating a corresponding care intervention level based on the corresponding health risk index includes: Dividing the using time period of the target user into long-time segments, determining all the time segments of the single user state unit, and quantitatively scoring each time segment according to a first control factor to obtain a first time period score; And taking a weighted average value of the first time period scores of all the time slices covered by the single user state unit, calculating a corresponding first control factor integrated score, and taking the first control factor integrated score as a health risk index.
4. A method according to claim 3, wherein said calculating a corresponding care intervention level from said health risk index and said temporal attribute category comprises: acquiring historical care record data; Constructing a labeling data set based on the historical care record data, and extracting a physiological feature subset and a behavioral feature subset from the labeling data set to form a multidimensional input feature vector; utilizing the labeling data set to train a gradient lifting tree model, and establishing a nonlinear mapping relation between the multidimensional input characteristic vector and the optimal nursing parameter set; and in the operation stage, the multidimensional input feature vector corresponding to the current user state unit is input into the gradient lifting tree model, an output optimal nursing parameter set is obtained, and the nursing intervention level is determined according to the intervention intensity corresponding to the optimal nursing parameter set.
5. The method according to claim 1, wherein, with each of the user status units as a basic unit, care response priority data is calculated according to a preset second control factor, and specifically comprising: dividing the using time period of the target user into long-time segments, determining all the time segments covered by a single user state unit, and scoring the weighting of each time segment according to a second control factor to obtain a second time period score; and taking a dynamic weighted average of second time interval scores of all time slices covered by the single user state unit, calculating a corresponding second control factor priority score, and calculating nursing response priority data by using the second control factor priority score and a priority judgment rule.
6. The method of claim 1, wherein generating the care bed control instructions for adapting the time period characteristics based on the care intervention level, the care response priority data, and the time attribute categories corresponding to each of the user status units comprises: taking each user state unit as a basic unit, taking a corresponding nursing intervention level as row vector data and a corresponding nursing response priority data as column vector data, and inputting the row vector data and the column vector data into a preset nursing action decision matrix to obtain a corresponding nursing bed control action type and execution intensity; And generating and issuing a nursing bed control instruction according to the type and the execution intensity of the nursing bed control action corresponding to each user state unit.
7. The method of claim 6, wherein the method further comprises: Receiving a nursing effect score submitted by a user through a man-machine interaction interface; And when the nursing effect score is lower than a preset threshold value, triggering a model fine-tuning mechanism, performing disturbance adjustment on the basis of original nursing parameters, and incorporating the adjustment result and the new nursing effect score into an incremental training data set of the gradient lifting tree model.
8. Intelligent monitoring nursing bed based on behavioral habit fuses with physiological data, its characterized in that includes: The memory is used for storing the physiological monitoring data and behavior habit data of the target user, and a preset first control factor and a preset second control factor; A processor, coupled to the memory, configured to: Acquiring the behavior habit data and the physiological monitoring data; Constructing a plurality of user state units based on the behavior habit data and the physiological monitoring data, and identifying a time attribute category of each user state unit, wherein the time attribute category comprises a sleep period and a non-sleep period; Taking each user state unit as a basic unit, performing comfort and health risk assessment according to the first control factors to obtain corresponding health risk indexes, and calculating corresponding nursing intervention grades according to the health risk indexes and the time attribute categories; Calculating nursing response priority data according to the second control factors by taking each user state unit as a basic unit; Generating a nursing bed control instruction adapting to time period characteristics based on the nursing intervention level, the nursing response priority data and the time attribute category corresponding to each user state unit; and the executing mechanism is coupled with the processor and used for driving the bed body to execute corresponding actions according to the nursing bed control instruction.
9. A computer device comprising a memory, a processor and a computer program stored in the memory and executable on the processor, characterized in that the processor, when executing the computer program, carries out the steps of the intelligent monitoring care bed control method based on a fusion of behavioural habits and physiological data as claimed in any one of claims 1 to 7.
10. A computer readable storage medium storing a computer program, characterized in that the computer program when executed by a processor implements the steps of the intelligent monitoring care bed control method based on behavioral habit and physiological data fusion according to any one of claims 1 to 7.

Description

Intelligent monitoring nursing bed control method based on behavior habit and physiological data fusion Technical Field The application relates to the technical field of health monitoring, in particular to an intelligent monitoring nursing bed control method based on behavior habit and physiological data fusion. Background In the field of intelligent medical treatment and intelligent health care, a nursing bed system gradually introduces sensing monitoring and automatic control functions, monitors surface physiological parameters such as body position angles, local temperatures, off-bed time length and the like of a patient in real time through basic sensing units such as an integrated pressure sensor, an angle encoder and a temperature and humidity probe, and triggers nursing actions according to the surface physiological parameters. However, the prior art is generally based on a preset time interval or a single physiological threshold, and the control logic is mostly implemented by using a rule engine. The method still faces a plurality of common challenges in practical application, namely firstly, a care strategy generally adopts uniform parameter configuration, and is difficult to adapt to individual differences of different users in terms of body position preference, activity rhythm, physiological tolerance and the like, secondly, the dependent monitoring data are limited to static indexes, comprehensive perceptibility of dynamic physiological state change is lacked, risk early warning is delayed, and the current behavior state of the users is not considered in a control decision process, so that intervention actions can be executed at unsuitable occasions, and user experience is affected. Thus, there is a need for a new control method that more fully senses the status of the user and more flexibly responds to individual needs. Disclosure of Invention In order to provide a novel nursing bed control method capable of more comprehensively sensing the state of a user and responding to the needs of an individual more flexibly, the application provides an intelligent monitoring nursing bed control method based on the integration of behavior habit and physiological data. In a first aspect, the object of the application is achieved by the following technical scheme: an intelligent monitoring nursing bed control method based on behavior habit and physiological data fusion comprises the following steps: Acquiring behavior habit data and physiological monitoring data of a target user; Constructing a plurality of user state units based on the behavior habit data and the physiological monitoring data, and identifying a time attribute category of each user state unit, wherein the time attribute category comprises a sleep period and a non-sleep period; Taking each user state unit as a basic unit, performing comfort and health risk assessment according to a preset first control factor to obtain a corresponding health risk index, and calculating a corresponding nursing intervention level according to the health risk index and the time attribute type; taking each user state unit as a basic unit, and calculating nursing response priority data according to a preset second control factor; And generating nursing bed control instructions adapting to time period characteristics based on the nursing intervention level, the nursing response priority data and the time attribute category corresponding to each user state unit. By adopting the technical scheme, the limitations of the conventional intelligent nursing bed in the aspects of personalized adaptation and situational awareness are effectively overcome. The continuous use process is decomposed into user state units with semantic meaning through fusing behavior habit data and physiological monitoring information, and fine granularity division and context perception of the user state are realized through identifying time attribute categories. In order to establish a multi-dimensional and hierarchical risk and response joint evaluation mechanism, comfort and health risks are quantified based on a first control factor to generate a health risk index reflecting potential hazard degrees, urgency and feasibility of current care resources are evaluated by combining a second control factor, and care response priority is output. The nursing bed control instruction realizes the dynamic adaptation and parameterization output of the nursing control strategy, and maintains the comfort of the user to the maximum extent while ensuring the safety intervention. In a preferred example, the present application constructs a plurality of user state units based on the behavior habit data and the physiological monitoring data, and identifies a time attribute category of each user state unit, including: Performing data cleaning and time alignment on the behavior habit data and the physiological monitoring data to obtain processed multisource fusion user data; Performing sliding window segmentatio