Search

CN-121726074-B - Pressure injury monitoring system for multi-mode real-time sensing and dynamic risk assessment

CN121726074BCN 121726074 BCN121726074 BCN 121726074BCN-121726074-B

Abstract

The invention relates to the technical field of intelligent medical monitoring, in particular to a pressure injury monitoring system for multi-mode real-time perception and dynamic risk assessment, which comprises a multi-mode perception step, a digital twin analysis step, a pathology simulation injection step, a differential topology judgment step and a deep tissue injury early warning step, wherein the multi-mode perception step is used for receiving multi-source heterogeneous data streams and performing sequential synchronous alignment to generate real-time fusion characteristic data, the digital twin analysis step is used for constructing a virtual tissue dynamic model and resolving theoretical reference characteristic data in a health standard state, the pathology simulation injection step is used for calling a pathology evolution rule knowledge base and simulating to generate theoretical pathology characteristic data, the differential topology judgment step is used for executing double-track differential operation to generate actual observation and theoretical pathology residual vectors and calculating topology similarity measurement to generate risk assessment results, and the residual morphology matching mechanism is constructed through a double reference system to effectively filter environmental noise and physiological fluctuation, so that the false report rate is remarkably reduced and early warning of deep tissue injury is realized.

Inventors

  • HE JUAN

Assignees

  • 中国人民解放军空军军医大学

Dates

Publication Date
20260508
Application Date
20260212

Claims (7)

  1. 1. A system for monitoring pressure injury by multi-modal real-time perception and dynamic risk assessment, the system comprising: The multi-mode sensing terminal module is used for receiving and processing multi-source heterogeneous data streams of a target area, wherein the multi-source heterogeneous data streams comprise time sequence data of three modes of mechanics, heat and bioelectricity, and the multi-source heterogeneous data streams are subjected to time sequence synchronous alignment to generate standardized real-time fusion characteristic data; The digital twin analysis module is used for constructing a virtual tissue dynamic model based on the prior characteristic data of the individual, inputting the pressure related characteristic in the real-time fusion characteristic data into the virtual tissue dynamic model, and solving and outputting theoretical reference characteristic data under the corresponding health benchmark state; the pathology simulation injection module is used for calling a preset pressure injury pathology evolution rule knowledge base, extracting an evolution parameter set and injecting the evolution parameter set into the operation logic of the virtual tissue dynamic model so as to simulate and generate theoretical pathology characteristic data representing a pathology evolution process; The differential topology judgment module is used for executing double-track differential operation based on the real-time fusion characteristic data, the theoretical reference characteristic data and the theoretical pathological characteristic data to generate a real observation residual vector and a theoretical pathological residual vector, and generating a dynamic risk assessment result by calculating topology similarity measurement of the real observation residual vector and the theoretical pathological residual vector in a characteristic space; The digital twin analysis module comprises: The parameter initialization unit is used for loading priori feature data of an individual, wherein the priori feature data comprise body mass indexes, briden scores and historical blood perfusion reference values provided by an external system; The model construction unit is used for defining and initializing parameters of a dynamic system model with viscoelastic characteristics based on the prior characteristic data; the reference generating unit is used for inputting time sequence pressure characteristics in the real-time fusion characteristic data as a model, driving the dynamic system model to calculate, outputting a theoretical deformation sequence, a theoretical temperature change sequence and a theoretical impedance sequence without pathological disturbance, and combining the theoretical deformation sequence, the theoretical temperature change sequence and the theoretical impedance sequence into the theoretical reference characteristic data; The model construction unit adopts a backward Euler discretization scheme to construct the following recursive calculation formula to solve the theoretical deformation sequence : ; Wherein, the The source is real-time pressure data drive, the physical meaning is stress born by the tissue, and the unit is Pa; the method is characterized in that the method is a theoretical deformation sequence, the source is calculated by a recursive formula, the physical meaning is the compression degree of the tissue, and the unit is 1; For the effective elastic modulus, the nonlinear mapping function of BMI index is set as , wherein, As a reference modulus, the modulus of the material, For the stiffness, the physical meaning is the elastic stiffness of the tissue; for effective viscosity coefficient, the source is the correction of the Braden score, and the formula is Wherein, the method comprises the steps of, As a reference coefficient of viscosity, As a risk sensitivity coefficient, a risk-sensitive coefficient, For the theoretical maximum value of the braden score scale, The score is currently estimated for the patient.
  2. 2. The multi-modal real-time perception and dynamic risk assessment pressure injury monitoring system of claim 1 wherein the pathology simulation injection module comprises: the factor extraction unit is used for inquiring and extracting ischemia cascade coefficients, inflammatory thermal diffusivity and cell membrane permeability factors from the pressure injury pathological evolution rule knowledge base to form the evolution parameter set; and the superposition simulation unit is used for superposing the evolution parameter set into the operation logic of the virtual tissue dynamic model and simulating the conduction and diffusion effects defined by ischemia and inflammation rules so as to generate the theoretical pathological feature data.
  3. 3. The multi-modal real-time awareness and dynamic risk assessment pressure impairment monitoring system of claim 1, wherein the differential topology decision module comprises: the first difference unit is used for calculating the difference value between the real-time fusion characteristic data and the theoretical reference characteristic data and generating the actual observation residual vector containing the composite interference; The second difference unit is used for calculating the difference value between the theoretical pathological feature data and the theoretical reference feature data and generating a pure theoretical pathological residual error vector; And the coupling calculation unit is used for mapping the actual observation residual vector to a characteristic subspace formed by the theoretical pathological residual vector, and calculating cosine similarity of the actual observation residual vector and the theoretical pathological residual vector as the topological similarity measure.
  4. 4. The multi-modal real-time awareness and dynamic risk assessment pressure impairment monitoring system of claim 3, wherein the differential topology decision module further comprises: The logic judging unit is used for comparing the topological similarity measurement with a preset risk judging threshold value; Judging that the target area has pressure damage risk under the condition that the topological similarity measure is higher than the risk judging threshold value, and generating a high risk alarm instruction; And under the condition that the topological similarity measure is lower than the risk judging threshold value, judging that the actual observation residual vector is acceptable background fluctuation, and generating a safety state instruction.
  5. 5. The system for monitoring the pressure injury through multi-modal real-time sensing and dynamic risk assessment according to claim 1, wherein the multi-modal sensing terminal module comprises: the data interface unit is used for receiving time sequence pressure value data, skin surface temperature distribution map and environment reference temperature data and complex impedance spectrum data of biological tissues from the external sensing unit; and the characteristic synchronization unit is used for carrying out time stamp alignment and data fusion on the multi-source heterogeneous data stream to form standardized real-time fusion characteristic data.
  6. 6. The multi-modal real-time awareness and dynamic risk assessment pressure injury monitoring system of claim 1 further comprising: And the edge gateway module is connected to the upstream of the multi-mode sensing terminal module and is used for carrying out protocol analysis, format unification and noise filtering on the input original multi-source heterogeneous data stream, and packaging the cleaned data into a standardized message and transmitting the standardized message to the multi-mode sensing terminal module.
  7. 7. The multi-modal real-time awareness and dynamic risk assessment pressure injury monitoring system of claim 4 further comprising: And the feedback regulation module is used for responding to the high-risk alarm instruction output by the differential topology judgment module and generating a dynamic regulation strategy signal aiming at the nursing bed, wherein the signal can be used for driving an external execution mechanism to change the mechanical environment of a target area.

Description

Pressure injury monitoring system for multi-mode real-time sensing and dynamic risk assessment Technical Field The invention relates to the technical field of intelligent medical monitoring, in particular to a multi-mode real-time sensing and dynamic risk assessment pressure injury monitoring system. Background Along with the rapid iteration of the clinical monitoring technology, early prevention and treatment of pressure injury becomes a core pain point in the nursing of long-term bedridden patients, the complication seriously affects the life quality of the patients, how to accurately identify the risk of deep tissue injury in a complex clinical environment and effectively filter environmental interference and physiological fluctuation is a great difficulty currently faced, and whether noninvasive and accurate early warning can be realized on the premise of pursuing the refinement in the medical care field is a great focus of the monitoring technology. Traditional pressure injury monitoring schemes currently rely primarily on timed healthcare personnel manual skin assessment, single-dimensional pressure distribution pad monitoring, and body surface temperature alarms based on simple thresholds. However, the modes of manual evaluation, single pressure monitoring, temperature threshold alarming and the like have certain defects, for example, manual evaluation depends on naked eyes and often lags behind the occurrence of deep tissue injury, an intervention window is missed, the single pressure monitoring only reflects the external load condition and cannot acquire the actual pathological reaction inside the tissue, the temperature monitoring mode based on the threshold is easily interfered by the environment heat source such as an electric blanket or the normal physiological fluctuation such as reactive congestion and the like, the false alarm rate is extremely high, and the physiological compensation reaction cannot be effectively distinguished from the early pathological injury. Disclosure of Invention In order to solve the above technical problems, the present invention provides a pressure injury monitoring system for multi-mode real-time sensing and dynamic risk assessment, and specifically, the technical scheme of the present invention includes: The multi-mode sensing terminal module is used for receiving and processing multi-source heterogeneous data streams of a target area, wherein the multi-source heterogeneous data streams comprise time sequence data of three modes of mechanics, heat and bioelectricity, and the multi-source heterogeneous data streams are subjected to time sequence synchronous alignment to generate standardized real-time fusion characteristic data; The digital twin analysis module is used for constructing a virtual tissue dynamic model based on the prior characteristic data of the individual, inputting the pressure related characteristic in the real-time fusion characteristic data into the virtual tissue dynamic model, and solving and outputting theoretical reference characteristic data under the corresponding health benchmark state; the pathology simulation injection module is used for calling a preset pressure injury pathology evolution rule knowledge base, extracting an evolution parameter set and injecting the evolution parameter set into the operation logic of the virtual tissue dynamic model so as to simulate and generate theoretical pathology characteristic data representing a pathology evolution process; And the differential topology judgment module is used for executing double-track differential operation based on the real-time fusion characteristic data, the theoretical reference characteristic data and the theoretical pathological characteristic data to generate a real observation residual vector and a theoretical pathological residual vector, and generating a dynamic risk assessment result by calculating the topological similarity measurement of the real observation residual vector and the theoretical pathological residual vector in a characteristic space. Preferably, the digital twin analysis module includes: The parameter initialization unit is used for loading priori feature data of an individual, wherein the priori feature data comprise body mass indexes, briden scores and historical blood perfusion reference values provided by an external system; The model construction unit is used for defining and initializing parameters of a dynamic system model with viscoelastic characteristics based on the prior characteristic data; and the reference generation unit is used for inputting the time sequence pressure characteristic in the real-time fusion characteristic data as a model, driving the dynamic system model to calculate, outputting a theoretical deformation sequence, a theoretical temperature change sequence and a theoretical impedance sequence without pathological disturbance, and combining the theoretical deformation sequence, the theoretical temperature change sequen