Catheter slippage risk assessment system and method based on drainage tube in-place status sensing and patient activity parameters

Abstract

The application provides a catheter slipping risk assessment system and method based on drainage tube in-situ state sensing and patient activity parameters, and relates to the technical field of drainage tube monitoring. The system comprises: collecting drainage tube in-situ micro-motion timing characteristics through an in-situ state sensing module, and synchronously analyzing patient activity vector sequences through an activity parameter acquisition module; an expected pulling direction calculation module calculates the expected tension pulling direction according to the above, and maps a dynamic displacement tolerance threshold; a slipping feature decoupling module projects the micro-motion characteristics to the direction in space, eliminates normal physiological body motion interference, and extracts the effective slipping displacement amount of the tube body; a risk assessment and early warning module triggers early warning according to the over-limit situation of the effective displacement amount, and performs negative attenuation update on the threshold based on the baseline drift rate when the over-limit situation does not occur. The application realizes accurate decoupling of real slipping features, and completely overcomes the high false alarm and implicit missing report defects of traditional schemes.

Description

Technical Field

[0001] This invention relates to the field of drainage tube monitoring technology, and in particular to a system and method for assessing the risk of catheter slippage based on the perception of the in-situ status of the drainage tube and patient activity parameters. Background Technology

[0002] Drainage tubes are widely used in intensive care units and postoperative rehabilitation wards. They are essential medical devices for draining accumulated fluid and monitoring changes in the patient's condition. Unplanned extubation is highly likely when patients experience postoperative delirium or unconscious turning. Current catheter dislodgement management primarily relies on physical fixation devices. For example, Chinese invention patent application CN113425988A discloses a chest drainage catheter fixation mechanism. This mechanism physically anchors the catheter to the patient's body surface using mechanical clips and a dressing layer. However, simple physical fixation cannot prevent continuous accidental pulling by the patient. Preventing dislodgement requires regular manual rounds by nursing staff. The article "Research and Application of Early Warning and Decision Support System for Unplanned Extubation of Endotracheal Intubation" published in the Chinese Journal of Nursing, Volume 57, Issue 1, 2022, describes the limitations of existing early warning management methods. This article points out that traditional timed rounds relying on manual experience have significant time lags. Static scale assessments cannot continuously and objectively obtain the real-time force state of the catheter.

[0003] To address the technical shortcomings of manual inspections due to their inherent lag, those skilled in the art have begun employing electronic sensing devices for continuous automated monitoring. Existing sensing solutions typically involve connecting a tension sensor in series with the external segment of the catheter. Current systems acquire tension parameters along the catheter's axial direction. The system processor compares these tension parameters with a preset static tension threshold. When the catheter tension parameter exceeds the static tension threshold, the system triggers an anti-slip warning signal. This threshold comparison scheme replaces manual visual judgment with a numerical threshold. This sensing solution has preliminarily established a hardware-based mechanism for sensing the catheter's safety status.

[0004] Real-world clinical applications demonstrate that the single-threshold approach has an extremely high false alarm rate. Normal patient positioning inevitably causes skin deformation, which exerts a transient tensile force on the drainage tube. The single force parameter collected by existing systems cannot distinguish between actual tube dislodgement and normal patient movement. Short-term traction caused by normal physiological activity easily exceeds the absolute force threshold. Existing systems frequently trigger false alarms during normal activity scenarios. Frequent false alarms lead to severe alarm fatigue among clinical nursing staff. In cases of latent slippage where the fixation dressing gradually loses its adhesiveness due to sweat, the catheter experiences millimeter-level continuous outward slippage. This slow slippage process does not generate a sudden tensile force along the catheter's axis. Existing threshold comparison mechanisms cannot effectively detect this slow slippage process. This results in missed alarms and ultimately, complete extubation. In summary, existing technologies suffer from the technical problem of difficulty in accurately decoupling microscopic displacement characteristics from macroscopic body movement interference. Summary of the Invention

[0005] In order to overcome the shortcomings of the prior art, the purpose of this invention is to provide a catheter slippage risk assessment system and method based on the perception of the in-situ status of the drainage tube and the patient's activity parameters. This invention solves the technical problem that the existing technology has difficulty in accurately decoupling micro-displacement characteristics and macro-motion interference.

[0006] To achieve the above objectives, the present invention provides the following solution: A catheter slippage risk assessment system based on drainage tube in-situ status sensing and patient activity parameters includes: The in-situ status sensing module is used to collect the displacement parameters of the drainage tube and extract the in-situ micro-motion timing features of the drainage tube based on the displacement parameters. The activity parameter acquisition module is used to synchronously collect the current patient's body movement status data and parse the body movement status data into an activity vector sequence; The expected tension direction calculation module is used to receive the activity vector sequence and calculate the expected tension direction of the drainage tube in the current body movement state based on the activity vector sequence. The dynamic threshold mapping module is used to receive the expected tension pulling direction and generate a dynamic displacement tolerance threshold corresponding to the body motion state data based on the expected tension pulling direction mapping. The slippage feature decoupling module is used to spatially project the in-situ micro-motion timing features onto the expected tension pulling direction, and extract the effective slippage displacement of the tube body that is decoupled from the body motion state data; The risk assessment and early warning module is used to receive the in-situ micro-motion timing characteristics, the dynamic displacement tolerance threshold, and the effective slippage displacement of the tube body. When the effective slippage displacement of the tube body is greater than the dynamic displacement tolerance threshold, a slippage early warning signal is obtained. If the effective slippage displacement of the tube body is less than or equal to the dynamic displacement tolerance threshold, the baseline drift rate of the in-situ micro-motion timing characteristics is extracted. When the baseline drift rate is greater than the latent slippage determination threshold, the dynamic displacement tolerance threshold is updated with negative decay.

[0007] A method for assessing catheter slippage risk based on drainage tube in-situ status perception and patient activity parameters includes: The displacement parameters of the drainage tube are collected, and the in-situ micro-motion timing characteristics of the drainage tube are extracted based on the displacement parameters; Simultaneously collect the patient's current body movement status data, and parse the body movement status data into an activity vector sequence; Receive the activity vector sequence, and calculate the expected tension direction of the drainage tube in the current body movement state based on the activity vector sequence; Receive the expected tension direction and generate a dynamic displacement tolerance threshold corresponding to the body motion state data based on the expected tension direction mapping. The in-situ micro-motion timing features are spatially projected onto the expected tension pulling direction to extract the effective slip displacement of the tube body, which is decoupled from the body motion state data. The system receives the in-situ micro-motion timing features, the dynamic displacement tolerance threshold, and the effective slip displacement of the tube body. When the effective slip displacement of the tube body is greater than the dynamic displacement tolerance threshold, a slip warning signal is obtained. If the effective slip displacement of the tube body is less than or equal to the dynamic displacement tolerance threshold, the baseline drift rate of the in-situ micro-motion timing features is extracted. When the baseline drift rate is greater than the latent slip judgment threshold, the dynamic displacement tolerance threshold is updated with negative decay.

[0008] The present invention discloses the following technical effects: This invention provides a catheter slippage risk assessment system and method based on in-situ state perception of the drainage tube and patient activity parameters. The invention simultaneously acquires the in-situ micro-motion temporal characteristics of the drainage tube and the patient's activity vector sequence, and calculates the expected tension direction by combining a human skeletal key point model and skin elastic damping coefficient. The in-situ micro-motion temporal characteristics are spatially projected onto the expected tension direction, and hysteresis effects are compensated. This accurately isolates the transient traction interference caused by the patient's normal positional changes on the drainage tube, extracting the effective slippage displacement of the tube body decoupled from macroscopic body motion, fundamentally addressing the issue. This system overcomes the frequent false alarms caused by the inability of existing technologies to decouple from normal physiological activities. Furthermore, for slow dislodgement scenarios where the effective slippage displacement of the catheter body does not exceed the dynamic displacement tolerance threshold, this system extracts the baseline drift rate of the in-situ micro-motion temporal characteristics and performs negative decay updates on the dynamic displacement tolerance threshold accordingly. This enables adaptive capture of hidden micro-slippage risks such as dressing loosening due to moisture, filling the blind spot of the existing static single absolute threshold monitoring mechanism. Thus, in the complex dynamic clinical context, it achieves high-precision, interference-resistant, all-weather accurate early warning of catheter slippage risk. Attached Figure Description

[0009] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0010] Figure 1 A schematic diagram of a catheter slippage risk assessment system based on the perception of the in-situ status of the drainage tube and patient activity parameters is provided for an embodiment of the present invention. Figure 2 The flowchart illustrates a catheter slippage risk assessment method based on the perception of the in-situ status of the drainage tube and patient activity parameters, as provided in this embodiment of the invention.

[0011] Figure label: 1-In-situ status perception module, 2-Activity parameter acquisition module, 3-Expected pulling direction calculation module, 4-Dynamic threshold mapping module, 5-Slippage feature decoupling module, 6-Risk assessment and early warning module. Detailed Implementation

[0012] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0013] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0014] like Figure 1 As shown, this invention provides a catheter slippage risk assessment system based on drainage tube in-situ status sensing and patient activity parameters, comprising: The in-situ status sensing module 1 is used to collect the displacement parameters of the drainage tube and extract the in-situ micro-motion timing features of the drainage tube based on the displacement parameters. Activity parameter acquisition module 2 is used to synchronously collect the current patient's body movement status data and parse the body movement status data into an activity vector sequence; The expected tension direction calculation module 3 is used to receive the activity vector sequence and calculate the expected tension direction of the drainage tube in the current body movement state based on the activity vector sequence. Dynamic threshold mapping module 4 is used to receive the expected tension pulling direction and generate a dynamic displacement tolerance threshold corresponding to the body motion state data based on the expected tension pulling direction mapping. The slippage feature decoupling module 5 is used to spatially project the in-situ micro-motion timing features onto the expected tension pulling direction, and extract the effective slippage displacement of the tube body that is decoupled from the body motion state data; The risk assessment and early warning module 6 is used to receive the in-situ micro-motion timing features, the dynamic displacement tolerance threshold, and the effective slip displacement of the tube body. When the effective slip displacement of the tube body is greater than the dynamic displacement tolerance threshold, a slip warning signal is obtained. If the effective slip displacement of the tube body is less than or equal to the dynamic displacement tolerance threshold, the baseline drift rate of the in-situ micro-motion timing features is extracted. When the baseline drift rate is greater than the latent slip judgment threshold, the dynamic displacement tolerance threshold is updated with negative decay.

[0015] Specifically, this embodiment provides a catheter slippage risk assessment system based on the perception of the drainage tube's in-situ status and patient activity parameters. This system is primarily used in intensive care units or post-surgical wards, aiming to address the shortcomings of anti-dislodgement systems, such as the tendency for normal drainage tube traction caused by patient repositioning to trigger false alarms, and the hidden underreporting caused by moisture-induced loosening of dressings.

[0016] In terms of hardware deployment architecture, this assessment system relies on a bedside monitoring main control unit. The bedside monitoring main control unit establishes data transmission links with flexible sensing terminals deployed on the patient's body surface and a depth vision camera suspended in front of the bed. At the logical control level, the system is divided into six functional modules: 1. In-situ status perception module; 2. Activity parameter acquisition module; 3. Expected pulling direction calculation module; 4. Dynamic threshold mapping module; 5. Slippage feature decoupling module; and 6. Risk assessment and early warning module. Each module communicates closely with the others via a unidirectional data flow: data acquisition, interference feature calculation, data decoupling, and risk determination. Through the coordinated operation of these modules, the system can isolate the interference of normal physiological activities from the actual slippage characteristics of the drainage tube, thereby achieving high-precision, interference-resistant catheter slippage prevention and early warning.

[0017] The specific operational logic and execution steps of the above six functional modules are explained in detail: During system operation, the in-situ state perception module 1 and the activity parameter acquisition module 2 perform synchronous data acquisition. Specifically, the in-situ state perception module 1 continuously reads the mechanical deformation at the skin-contact surface of the drainage tube using a flexible piezoelectric sensor attached to the fixed section of the drainage tube, and converts this mechanical deformation into displacement parameters of the drainage tube. Subsequently, the in-situ state perception module 1 uses a low-pass filter to filter out high-frequency physiological noise such as heartbeat and respiration, and extracts in-situ micro-motion temporal features reflecting the actual low-frequency movement trajectory of the drainage tube based on the displacement parameters. At the same time, the activity parameter acquisition module 2 synchronously acquires the patient's current body movement data (e.g., continuous video frames with depth information) using a depth vision camera, extracts the spatial displacement of key joints such as the patient's shoulder, chest, or waist using a human skeleton recognition algorithm, and parses the body movement data into an activity vector sequence composed of movement direction and amplitude.

[0018] After acquiring the aforementioned underlying data, the system enters the anti-interference parameter calculation stage. First, the expected tension direction calculation module 3 receives the activity vector sequence. This module has a pre-set patient skin deformation transmission model. When the patient performs actions such as turning over or raising their arm, this module calculates the corresponding skeletal movement trend based on the activity vector sequence, and then calculates the traction direction on the drainage tube when the skeletal movement is transmitted through the skin and soft tissue to the drainage tube anchor point, thereby obtaining the expected tension direction of the drainage tube in the current body movement state. Subsequently, the dynamic threshold mapping module 4 receives the expected tension direction. Since the skin extensibility varies in different directions of the human body (for example, the lateral stretch margin of chest skin is greater than that in the longitudinal direction), this module matches the corresponding maximum allowable deformation value in the pre-set skin elasticity tolerance matrix based on the expected tension direction, and then maps and generates a dynamic displacement tolerance threshold corresponding to the body movement state data. This dynamic displacement tolerance threshold represents the maximum legal displacement limit that the system allows the drainage tube to undergo normal deformation with the skin under the current patient activity.

[0019] To address the false alarm problem caused by "human movement and tube movement" in existing technologies, the slippage feature decoupling module 5 performs spatial decoupling calculations. This module receives the in-situ micro-motion timing features and the expected tension direction, establishes a three-dimensional decoupling coordinate system with the drainage tube fixation point as the origin, and projects the actual movement vector corresponding to the in-situ micro-motion timing features onto the axis containing the expected tension direction. Through this spatial projection operation, the overall displacement of the drainage tube is decomposed into a "physiological following displacement component" in the same direction as the expected traction and an "abnormal free displacement component" perpendicular to the traction direction. The slippage feature decoupling module 5 performs vector synthesis of the abnormal values ​​in the same direction that exceed the theoretical skin elastic deformation range and the normal abnormal free displacement component, thereby accurately eliminating physiological following displacement interference caused by the patient's normal turning over, and finally extracting the effective slippage displacement of the tube body decoupled from the body movement data.

[0020] After obtaining the actual displacement after decoupling, the risk assessment and early warning module 6 receives the in-situ micro-motion timing characteristics, the dynamic displacement tolerance threshold, and the effective slip displacement of the tube body, and executes a dual-track risk determination mechanism.

[0021] (1) Overt slippage determination: When the patient struggles violently or pulls directly on the catheter, the effective slippage displacement of the catheter body will increase rapidly. The risk assessment and early warning module 6 compares the two values. When the effective slippage displacement of the catheter body is greater than the dynamic displacement tolerance threshold, the system immediately determines that overt slippage has occurred and directly obtains a slippage early warning signal to drive the on-site audible and visual alarm.

[0022] (2) Dynamic Compensation for Latent Slippage: In some complex clinical scenarios (such as when sweat causes a slow decrease in the adhesiveness of medical dressings), the drainage tube may slip outwards at the millimeter level. At this time, the effective slippage displacement of the tube body may be temporarily less than or equal to the dynamic displacement tolerance threshold. Existing systems often miss the report because the fixed threshold is not triggered. For this scenario, the risk assessment and early warning module 6 extracts the time-domain waveform corresponding to the in-situ micro-motion time sequence characteristics and calculates the slope of the fitted line of the center of the time-domain waveform. The slope is used as the baseline drift rate (the larger the baseline drift rate, the more obvious the trend of the dressing gradually loosening). When the baseline drift rate is greater than the preset latent slippage judgment threshold, the risk assessment and early warning module 6 multiplies the dynamic displacement tolerance threshold by an attenuation coefficient less than 1 and performs a negative attenuation update on the dynamic displacement tolerance threshold. By continuously lowering the tolerance threshold, the system indirectly improves the sensitivity to the perception of micro-slippage, forcing the system to quickly trigger the slippage early warning signal in the next monitoring cycle, perfectly making up for the missed detection blind spot of traditional static monitoring.

[0023] Furthermore, the specific implementation of the in-situ state sensing module 1 is as follows: In order to capture minute displacement changes of the drainage tube with high precision and anti-interference, the in-situ state sensing module 1 is internally divided into a parameter acquisition submodule, a noise reduction preprocessing submodule, and a time-series feature construction submodule.

[0024] In actual operation, the parameter acquisition submodule is responsible for acquiring the displacement parameters of the drainage tube in a preset direction. Specifically, the acquisition method is as follows: the system senses the mechanical deformation of the drainage tube through a flexible piezoelectric sensor array attached to a fixed section of the drainage tube surface; subsequently, the mechanical deformation is sampled at a frequency of... The analog-to-digital conversion operation is performed to obtain a continuous digital electrical signal sequence; finally, the digital electrical signal sequence is used as the displacement parameter of the drainage tube.

[0025] It is important to note that when the drainage tube slips outward, either subtly or overtly, within the fixed dressing, the movement between the tube wall and the dressing is not perfectly smooth; instead, it induces microscopic stick-slip high-frequency frictional flutter. To convert the high-frequency voltage signal representing deformation acceleration into macroscopic physical displacement, the system employs a discretized energy equivalent integration algorithm: the system extracts the root mean square value of the signal in the resting state as the base environment noise floor voltage. Let the total number of flutter pulses detected within a single micro-motion cycle be... , No. The start and end sampling points of the next pulse are: and Cumulative displacement parameter The calculation formula is: ; In the formula, This is the calibrated piezoelectric-mechanical strain coupling constant. The algorithm eliminates the digital drift of traditional quadratic integrals through dead-time setting and pulse envelope integration.

[0026] Subsequently, the noise reduction and preprocessing submodule, which is communicatively connected to the parameter acquisition submodule, receives the displacement parameters and performs noise reduction processing on the displacement parameters using a low-pass filtering algorithm (effectively filtering out high-frequency electromyographic interference such as heartbeat and respiration) to obtain filtered displacement parameters. Finally, the timing feature construction submodule, which is communicatively connected to the noise reduction and preprocessing submodule, receives the filtered displacement parameters and extracts in-situ micro-motion timing features that can characterize the actual low-frequency physical movement of the drainage tube based on the distribution trajectory of the filtered displacement parameters in the time domain.

[0027] The specific implementation method of the activity parameter acquisition module 2 is as follows: This module is designed to accurately capture environmental interference benchmarks, and its internal components include: a state data synchronization submodule, a key point coordinate extraction submodule, and a vector sequence parsing submodule.

[0028] In practice, the status data synchronization submodule is used to synchronously collect the current patient's body movement status data. To overcome the failure problem of traditional two-dimensional RGB cameras under changes in ward lighting at night or when obscured by blankets, the body movement status data is specifically limited to depth image data (DepthMap) or three-dimensional point cloud data (3DPointCloud) with spatial depth information.

[0029] Subsequently, the joint coordinate extraction submodule, which is communicatively connected to the state data synchronization submodule, receives the body motion state data and inputs the body motion state data into a preset human skeleton key point model (such as a deep learning-based 3D pose estimation model). This accurately extracts the 3D coordinate parameters of multiple key joints throughout the patient's body (e.g., bilateral shoulder joints, sternal manubrium, pelvis, etc.) in the 3D camera coordinate system. ,in Number the joints.

[0030] Finally, the vector sequence parsing submodule, which is communicatively connected to the joint point coordinate extraction submodule, receives the three-dimensional coordinate parameters. The system compares adjacent sampling periods (such as the first sampling period). Frame and the The spatial motion vectors of each key joint point are calculated based on the three-dimensional coordinate parameters of the frame. Then, the corresponding motion direction (i.e., the spatial orientation angle of the vector) and displacement amplitude (i.e., the magnitude of the vector) are calculated, and the motion direction and displacement amplitude are structurally parsed into a sequence of motion vectors. Through the above rigorous lower-level steps, the system completes the accurate conversion of the patient's macroscopic irregular body motion into a calculable three-dimensional physical vector.

[0031] Furthermore, the specific implementation of the expected tension direction calculation module 3 is as follows: This module aims to deduce the theoretical traction effect of patient movement on the catheter by using biomechanical conduction laws. It specifically includes: anchor point spatial positioning submodule, local damping acquisition submodule, topology model adaptation submodule, traction force vector dimensionality reduction submodule, conduction phase alignment submodule, and expected direction generation submodule.

[0032] First, the anchor point spatial positioning submodule receives the activity vector sequence, extracts the initial body position reference data corresponding to the activity vector sequence, compares the initial body position reference data with the human skeleton key point model, and calculates the anchor point spatial coordinates of the drainage tube fixed on the patient's body surface.

[0033] Subsequently, the local damping acquisition submodule receives the spatial coordinates of the anchor point and extracts the skin elastic damping coefficient matching the spatial coordinates of the anchor point from a pre-set body surface elasticity feature library. It should be noted that the system obtains the patient's gender, age, and body mass index (BMI) by connecting to the hospital's electronic medical record system, and uses these physiological characteristics as index keys to look up the coefficient, ensuring individualized biomechanical parameter suitability.

[0034] Next, the topology model adaptive submodule receives the skin elastic damping coefficient and uses it to correct the mechanical transfer function of the initial skin deformation topology model, thereby obtaining a target skin deformation topology model that fits the individual skin characteristics of the current patient.

[0035] The traction vector dimensionality reduction submodule receives the active vector sequence and the target skin deformation topology model, inputs the active vector sequence into the target skin deformation topology model for spatial dimensionality reduction calculation, and obtains the surface traction vector acting on the spatial coordinates of the anchor point.

[0036] Furthermore, considering the viscoelasticity of subcutaneous fat and fascia, the conduction phase alignment submodule receives the surface traction force vector and, in conjunction with the preset skin soft tissue conduction delay time difference (which is determined by the system by looking up a table with parameters such as BMI, and is usually several hundred milliseconds), performs phase offset compensation on the surface traction force vector to eliminate the system phase difference caused by soft tissue conduction hysteresis, thereby obtaining a steady-state deformation prediction vector with precise time-scale alignment.

[0037] Finally, the expected direction generation submodule receives the steady-state deformation prediction vector, extracts the spatial transmission direction of the steady-state deformation prediction vector, and establishes the spatial transmission direction as the expected tension direction of the drainage tube in the current body motion state.

[0038] Furthermore, the specific implementation of the dynamic threshold mapping module 4 is as follows: This module is designed to set a safety boundary for the system that changes in real time with the patient's body movements. It includes a spatial angle matching submodule, a baseline tolerance extraction submodule, and a dynamic threshold generation submodule.

[0039] During operation, the spatial angle matching submodule receives the expected tension direction and calculates the target spatial angle of the expected tension direction in a preset three-dimensional coordinate system. Because the extensibility of local human skin (such as the chest and abdomen) exhibits significant anisotropy in different directions (for example, the lateral skin stretch tolerance is often greater than the longitudinal tolerance), the reference tolerance extraction submodule is communicatively connected to the spatial angle matching submodule. It receives the target spatial angle and searches for the reference maximum displacement margin corresponding to the target spatial angle in a preset anisotropic tension tolerance matrix.

[0040] Subsequently, the dynamic threshold generation submodule receives the baseline maximum displacement margin and performs weighted compensation on the baseline maximum displacement margin in conjunction with a preset activity intensity coefficient. The activity intensity coefficient is derived from the displacement acceleration of the aforementioned activity vector sequence and is used to characterize the intensity of the movement. Through this weighted compensation calculation, the system maps and generates a dynamic displacement tolerance threshold that closely corresponds to the body motion data. This threshold has a clear physical meaning; it represents the maximum legal displacement limit that the drainage tube can undergo due to natural skin deformation under the current specific movement.

[0041] Furthermore, the specific implementation of the slippage feature decoupling module 5 is as follows: This is the core computational component for accurately eliminating human-induced interference and ensuring a low false alarm rate in the system. Internally, it includes: a dynamic coordinate system establishment submodule, an initial projection decomposition submodule, a hysteresis stage division and evaluation submodule, a hysteresis-in-the-direction compensation submodule, an abnormal free-floating extraction submodule, and a true slip synthesis submodule.

[0042] First, the dynamic coordinate system establishment submodule receives the expected tension pulling direction and establishes a dynamic decoupled coordinate system with the drainage tube fixing needle point as the origin and the expected tension pulling direction as the first coordinate axis. The initial projection decomposition submodule receives the in-situ micro-motion timing features and the dynamic decoupled coordinate system, projects the in-situ micro-motion timing features onto the dynamic decoupled coordinate system, and decomposes them to obtain an initial unidirectional displacement component (i.e., the component following the skin pulling direction) and an initial normal displacement component perpendicular to the expected tension pulling direction. Here, let the first... The initial same-direction displacement components calculated by projection at time step are: To address the nonlinear physical hysteresis effect of subcutaneous tissue rebound under stress, the hysteresis stage segmentation and evaluation submodule receives the initial unidirectional displacement component and calculates the backward difference value between adjacent sampling points. Based on the sign of the first derivative of the initial unidirectional displacement component, the micro-motion period is divided into a deformation loading stage (derivative greater than zero). ) and deformation unloading stage (derivative less than zero, Furthermore, the system calculates the integral area difference between the deformation loading stage and the deformation unloading stage in the time domain, and maps the integral area difference into a hysteresis compensation parameter reflecting the skin's rebound delay. The corresponding discrete algebraic calculation formula is: ; In the formula, This is the fat deformation recovery constant retrieved from a table using the patient's BMI. The hysteresis-co-directional compensation submodule receives the initial co-directional displacement component and the hysteresis compensation parameters, and then... With the hysteresis compensation parameters The addition operation is performed to obtain the corrected same-direction displacement components that eliminate hysteresis residue.

[0043] Subsequently, the abnormal slippage extraction submodule receives the corrected unidirectional displacement component, calculates the difference between the corrected unidirectional displacement component and the preset theoretical following displacement reference value, and obtains the abnormal unidirectional slippage amount. Under ideal physiological following conditions, the unidirectional displacement after hysteresis compensation should perfectly cancel out skin deformation. Therefore, the theoretical following displacement reference value is set to 0 mm by default in engineering, and any part exceeding this value is judged as abnormal slippage.

[0044] Finally, the real slip synthesis submodule receives the anomalous same-direction free quantity and the initial normal displacement component, and combines the anomalous same-direction free quantity with the initial normal displacement component. Furthermore, the specific implementation method of risk assessment and early warning module 6 is as follows: This module is responsible for implementing a dual-track risk assessment mechanism to ensure the consistency between "explicit prevention of false alarms" and "implicit prevention of missed alarms." This module includes: a displacement threshold comparison submodule, an explicit early warning triggering submodule, and an implicit slippage compensation submodule.

[0045] The displacement threshold comparison submodule receives the dynamic displacement tolerance threshold and the effective slip displacement of the pipe body, compares the effective slip displacement of the pipe body with the dynamic displacement tolerance threshold, and obtains a judgment state that includes explicit over-limit results or implicit non-over-limit results.

[0046] When the determination state includes the explicit over-limit result that the effective slippage displacement of the tube body is greater than the dynamic displacement tolerance threshold, the explicit early warning triggering submodule receives the determination state, directly obtains the slippage early warning signal, and drives the bedside audible and visual alarm to operate.

[0047] If the drainage tube becomes damp with sweat and its adhesiveness slowly decreases, the tube may experience a very slight, subtle outward slippage. In this case, the determination status includes a subtle, non-exceeding result where the effective slippage displacement of the tube is less than or equal to the dynamic displacement tolerance threshold. To eliminate such blind spots, the subtle slippage compensation submodule receives the determination status, the in-situ micro-motion timing characteristics, and the dynamic displacement tolerance threshold. It extracts the baseline drift rate of the in-situ micro-motion timing characteristics, and when the baseline drift rate is greater than the subtle slippage determination threshold, it performs a negative decay update on the dynamic displacement tolerance threshold.

[0048] More specifically, to ensure the implicit slippage compensation logic possesses rigorous mathematical convergence and robustness, this embodiment meticulously designs the underlying logic for the extraction of the baseline drift rate and the negative decay update. Specifically, when extracting the baseline drift rate, the system first preprocesses the in-situ micro-motion temporal features using a median filtering algorithm to filter out abrupt interference signals, thereby generating smooth steady-state micro-motion temporal features. Based on this, the system iterates through and extracts the maximum and minimum value sequences of the steady-state micro-motion temporal features, and constructs a peak-valley envelope cluster based on these two. Furthermore, the system calculates the central weighted median of the peak-valley envelope cluster and extracts this central weighted median as the core baseline change trajectory of the in-situ micro-motion temporal features. Through the above envelope optimization extraction operation, the system can effectively isolate the random fluctuation interference caused by occasional deep breathing or postural adjustments by the patient on the signal baseline.

[0049] After acquiring the core baseline change trajectory, the system further performs slope fitting to quantify the degree of baseline drift. The length of the preset time window is set to... The discrete time series within the sliding window is The corresponding trajectory amplitude is The system uses the least squares method to calculate the linear fitting slope of the core baseline change trajectory within the preset time window, and establishes the absolute value of the linear fitting slope as the baseline drift rate. The specific fitting calculation formula is as follows: ; After quantizing the baseline drift rate, the system enters an adaptive negative decay update phase. When the baseline drift rate... Greater than the implicit slippage determination threshold At that time, the system will use the baseline drift rate Inputting into a pre-defined exponential decay model, a nonlinear decay coefficient with values ​​between zero and one is generated. The mapping formula is: ; In the formula, With empirical sensitivity constant All data were obtained through pre-physical calibration based on actual medical film peel strength degradation test curves under different humidity levels. Subsequently, the system calculates the current dynamic displacement tolerance threshold. With the nonlinear attenuation coefficient The product of the two values ​​is used to directly overwrite the original value (i.e., update to the original value). This completes the negative attenuation update of the dynamic displacement tolerance threshold. It is important to emphasize that, due to the mathematical properties of the natural base exponential function, no matter how severe the baseline drift caused by moisture-induced loosening of the patch is, the nonlinear attenuation coefficient... Always strictly bound to Within the open interval. This convergence operation ensures that the dynamic tolerance threshold is smoothly and safely reduced, thereby forcing the system to trigger an over-limit alarm instantly when the drainage tube experiences a slight outward detachment, perfectly compensating for the blind spot of missed detection caused by traditional static fixed threshold systems that cannot detect dressing failures.

[0050] like Figure 2 As shown, this embodiment also provides a method for assessing catheter slippage risk based on the perception of the drainage tube's in-situ status and patient activity parameters, including: Step 100: Collect the displacement parameters of the drainage tube, and extract the in-situ micro-motion timing features of the drainage tube based on the displacement parameters; Step 200: Synchronously collect the current patient's body movement data and parse the body movement data into an activity vector sequence; Step 300: Receive the activity vector sequence, and calculate the expected tension direction of the drainage tube in the current body movement state based on the activity vector sequence; Step 400: Receive the expected tension direction, and generate a dynamic displacement tolerance threshold corresponding to the body motion state data based on the expected tension direction mapping; Step 500: Spatially project the in-situ micro-motion timing features onto the expected tension direction to extract the effective slip displacement of the tube body that is decoupled from the body motion state data; Step 600: Receive the in-situ micro-motion timing features, the dynamic displacement tolerance threshold, and the effective slip displacement of the tube body. When the effective slip displacement of the tube body is greater than the dynamic displacement tolerance threshold, a slip warning signal is obtained. If the effective slip displacement of the tube body is less than or equal to the dynamic displacement tolerance threshold, the baseline drift rate of the in-situ micro-motion timing features is extracted. When the baseline drift rate is greater than the latent slip judgment threshold, the dynamic displacement tolerance threshold is updated with negative decay.

[0051] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on the differences from other embodiments. The same or similar parts between the various embodiments can be referred to each other.

[0052] This document uses specific examples to illustrate the principles and implementation methods of the present invention. The descriptions of the above embodiments are only for the purpose of helping to understand the method and core ideas of the present invention. Furthermore, those skilled in the art will recognize that, based on the ideas of the present invention, there will be changes in the specific implementation methods and application scope. Therefore, the content of this specification should not be construed as a limitation of the present invention.

Claims

1. A catheter slippage risk assessment system based on drainage tube in-situ status sensing and patient activity parameters, characterized in that, include: The in-situ status sensing module is used to collect the displacement parameters of the drainage tube and extract the in-situ micro-motion timing features of the drainage tube based on the displacement parameters. The activity parameter acquisition module is used to synchronously collect the current patient's body movement status data and parse the body movement status data into an activity vector sequence; The expected tension direction calculation module is used to receive the activity vector sequence and calculate the expected tension direction of the drainage tube in the current body movement state based on the activity vector sequence. The dynamic threshold mapping module is used to receive the expected tension pulling direction and generate a dynamic displacement tolerance threshold corresponding to the body motion state data based on the expected tension pulling direction mapping. The slippage feature decoupling module is used to spatially project the in-situ micro-motion timing features onto the expected tension pulling direction, and extract the effective slippage displacement of the tube body that is decoupled from the body motion state data; The risk assessment and early warning module is used to receive the in-situ micro-motion timing characteristics, the dynamic displacement tolerance threshold, and the effective slippage displacement of the tube body. When the effective slippage displacement of the tube body is greater than the dynamic displacement tolerance threshold, a slippage early warning signal is obtained. If the effective slippage displacement of the tube body is less than or equal to the dynamic displacement tolerance threshold, the baseline drift rate of the in-situ micro-motion timing characteristics is extracted. When the baseline drift rate is greater than the latent slippage determination threshold, the dynamic displacement tolerance threshold is updated with negative decay.

2. The catheter slippage risk assessment system based on drainage tube in-situ status sensing and patient activity parameters according to claim 1, characterized in that, The in-situ status sensing module includes: The parameter acquisition submodule is used to acquire the displacement parameters of the drainage tube in a preset direction; The noise reduction preprocessing submodule is communicatively connected to the parameter acquisition submodule. It is used to receive the displacement parameters and perform noise reduction processing on the displacement parameters using a low-pass filtering algorithm to obtain filtered displacement parameters. The timing feature construction submodule is communicatively connected to the noise reduction preprocessing submodule. It is used to receive the filtered displacement parameter and extract the in-situ micro-motion timing features of the drainage tube based on the distribution trajectory of the filtered displacement parameter in the time domain.

3. The catheter slippage risk assessment system based on drainage tube in-situ status sensing and patient activity parameters according to claim 2, characterized in that, The methods for obtaining the displacement parameters include: The mechanical deformation of the drainage tube is sensed by a flexible piezoelectric sensor array attached to the fixed section of the drainage tube body. An analog-to-digital conversion operation is performed on the mechanical deformation to obtain a continuous sequence of digital electrical signals; The digital electrical signal sequence is used as the displacement parameter of the drainage tube.

4. The catheter slippage risk assessment system based on drainage tube in-situ status sensing and patient activity parameters according to claim 1, characterized in that, The activity parameter acquisition module includes: The status data synchronization submodule is used to synchronously collect the current patient's body motion status data, wherein the body motion status data is depth image data or three-dimensional point cloud data; The joint point coordinate extraction submodule is communicatively connected to the state data synchronization submodule. It is used to receive the body motion state data, input the body motion state data into a preset human skeleton key point model, and extract the three-dimensional coordinate parameters of multiple key joint points of the patient's whole body. The vector sequence parsing submodule is communicatively connected to the joint coordinate extraction submodule. It is used to receive the three-dimensional coordinate parameters, compare the three-dimensional coordinate parameters of adjacent sampling periods to calculate the corresponding motion direction and displacement amplitude, and parse the motion direction and displacement amplitude into an active vector sequence.

5. The catheter slippage risk assessment system based on drainage tube in-situ status sensing and patient activity parameters according to claim 1, characterized in that, The expected pulling direction calculation module includes: Anchor point spatial positioning submodule is used to receive the activity vector sequence, extract the initial body position reference data corresponding to the activity vector sequence, compare the initial body position reference data with the human skeleton key point model, and calculate the anchor point spatial coordinates of the drainage tube fixed on the patient's body surface. The local damping acquisition submodule is communicatively connected to the anchor point spatial positioning submodule and is used to receive the anchor point spatial coordinates and extract the skin elastic damping coefficient that matches the anchor point spatial coordinates from a preset body surface elastic feature library. The topology model adaptive submodule is communicatively connected to the local damping acquisition submodule. It is used to receive the skin elastic damping coefficient and use the skin elastic damping coefficient to correct the mechanical transfer function of the initial skin deformation topology model to obtain the target skin deformation topology model. The traction force vector dimensionality reduction submodule is communicatively connected to the topology model adaptive submodule. It is used to receive the active vector sequence and the target skin deformation topology model, input the active vector sequence into the target skin deformation topology model for spatial dimensionality reduction calculation, and obtain the surface traction force vector acting on the spatial coordinates of the anchor point. The conduction phase alignment submodule is communicatively connected to the traction force vector dimensionality reduction submodule. It is used to receive the surface traction force vector and perform phase offset compensation on the surface traction force vector by combining the preset skin soft tissue conduction delay time difference to obtain the steady-state deformation prediction vector. The expected direction generation submodule is communicatively connected to the conduction phase alignment submodule. It is used to receive the steady-state deformation prediction vector, extract the spatial conduction direction of the steady-state deformation prediction vector, and use the spatial conduction direction as the expected tension pulling direction of the drainage tube in the current body motion state.

6. The catheter slippage risk assessment system based on drainage tube in-situ status sensing and patient activity parameters according to claim 1, characterized in that, The dynamic threshold mapping module includes: The spatial angle matching submodule is used to receive the expected tension stretching direction and calculate the target spatial angle of the expected tension stretching direction in a preset three-dimensional coordinate system. The reference tolerance extraction submodule is communicatively connected to the spatial angle matching submodule. It is used to receive the target spatial angle and find the reference maximum displacement margin corresponding to the target spatial angle in the preset anisotropic tension tolerance matrix. The dynamic threshold generation submodule is communicatively connected to the baseline tolerance extraction submodule. It is used to receive the baseline maximum displacement margin, and perform weighted compensation on the baseline maximum displacement margin in combination with a preset activity intensity coefficient to map and generate a dynamic displacement tolerance threshold corresponding to the body motion state data.

7. The catheter slippage risk assessment system based on drainage tube in-situ status sensing and patient activity parameters according to claim 1, characterized in that, The slippage feature decoupling module includes: The dynamic coordinate system establishment submodule is used to receive the expected tension pulling direction and establish a dynamic decoupled coordinate system with the fixed anchor point of the drainage tube as the origin and the expected tension pulling direction as the first coordinate axis. The initial projection decomposition submodule establishes a communication connection with the dynamic coordinate system submodule. It is used to receive the in-situ micro-motion timing features and the dynamic decoupled coordinate system, and to spatially project the in-situ micro-motion timing features onto the dynamic decoupled coordinate system to decompose them into initial unidirectional displacement components and initial normal displacement components perpendicular to the expected tension direction. The hysteresis stage division and evaluation submodule is communicatively connected to the initial projection decomposition submodule. It is used to receive the initial same-direction displacement component, divide the micro-motion period into deformation loading stage and deformation unloading stage based on the positive and negative signs of the first derivative of the initial same-direction displacement component, calculate the integral area difference between the deformation loading stage and the deformation unloading stage in the time domain, and map the integral area difference into a hysteresis compensation parameter that reflects the skin rebound delay. The hysteresis co-directional compensation submodule is communicatively connected to the hysteresis stage division and evaluation submodule. It is used to receive the initial co-directional displacement component and the hysteresis compensation parameter, and to add the initial co-directional displacement component and the hysteresis compensation parameter to obtain the corrected co-directional displacement component. An abnormal free displacement extraction submodule is communicatively connected to the hysteresis co-directional compensation submodule. It is used to receive the corrected co-directional displacement component, calculate the difference between the corrected co-directional displacement component and the preset theoretical following displacement reference value, and obtain the abnormal co-directional free displacement amount. The real slip synthesis submodule is communicatively connected to the abnormal free ion extraction submodule and the initial projection decomposition submodule. It is used to receive the abnormal unidirectional free ion quantity and the initial normal displacement component, perform vector superposition of the abnormal unidirectional free ion quantity and the initial normal displacement component, and extract the modulus of the vector superposition result as the effective slip displacement of the tube body decoupled from the body dynamic state data.

8. The catheter slippage risk assessment system based on drainage tube in-situ status sensing and patient activity parameters according to claim 1, characterized in that, The risk assessment and early warning module includes: The displacement threshold comparison submodule is used to receive the dynamic displacement tolerance threshold and the effective slip displacement of the pipe body, compare the magnitude of the effective slip displacement of the pipe body with the dynamic displacement tolerance threshold, and obtain a judgment state including explicit limit exceedance results or implicit non-limit exceedance results. The explicit warning triggering submodule is communicatively connected to the displacement threshold comparison submodule and is used to receive the judgment state. When the judgment state includes the explicit over-limit result that the effective slip displacement of the pipe body is greater than the dynamic displacement tolerance threshold, a slip warning signal is obtained. The implicit slip compensation submodule is used to receive the determination state, the in-situ micro-motion timing features, and the dynamic displacement tolerance threshold. When the determination state includes the implicit non-exceeding result where the effective slip displacement of the pipe body is less than or equal to the dynamic displacement tolerance threshold, the baseline drift rate of the in-situ micro-motion timing features is extracted. When the baseline drift rate is greater than the implicit slip determination threshold, the dynamic displacement tolerance threshold is updated with negative decay.

9. A catheter slippage risk assessment system based on drainage tube in-situ status sensing and patient activity parameters according to claim 8, characterized in that, The methods for extracting and negatively decaying the baseline drift rate include: The sudden jump interference signal in the in-situ micro-motion timing features is filtered out using the median filtering algorithm to generate steady-state micro-motion timing features; Extract the maximum and minimum point sequences of the steady-state micro-motion time series features, and construct a peak-valley envelope cluster based on the maximum and minimum point sequences; Calculate the center weighted median of the peak-valley envelope cluster, and extract the center weighted median as the core baseline change trajectory of the in-situ micro-motion temporal characteristics; The slope of the linear fitting of the core baseline change trajectory within a preset time window is calculated using the least squares method, and the absolute value of the linear fitting slope is taken as the baseline drift rate. When the baseline drift rate is greater than the implicit slip determination threshold, the baseline drift rate is input into a preset exponential decay model to generate a nonlinear decay coefficient with a value between zero and one. Calculate the product of the current dynamic displacement tolerance threshold and the nonlinear attenuation coefficient, and use the product value to directly overwrite the original value to complete the negative attenuation update of the dynamic displacement tolerance threshold.

10. A method for assessing catheter slippage risk based on drainage tube in-situ status perception and patient activity parameters, characterized in that, include: The displacement parameters of the drainage tube are collected, and the in-situ micro-motion timing characteristics of the drainage tube are extracted based on the displacement parameters; Simultaneously collect the patient's current body movement status data, and parse the body movement status data into an activity vector sequence; Receive the activity vector sequence, and calculate the expected tension direction of the drainage tube in the current body movement state based on the activity vector sequence; Receive the expected tension direction and generate a dynamic displacement tolerance threshold corresponding to the body motion state data based on the expected tension direction mapping. The in-situ micro-motion timing features are spatially projected onto the expected tension pulling direction to extract the effective slip displacement of the tube body, which is decoupled from the body motion state data. The system receives the in-situ micro-motion timing features, the dynamic displacement tolerance threshold, and the effective slip displacement of the tube body. When the effective slip displacement of the tube body is greater than the dynamic displacement tolerance threshold, a slip warning signal is obtained. If the effective slip displacement of the tube body is less than or equal to the dynamic displacement tolerance threshold, the baseline drift rate of the in-situ micro-motion timing features is extracted. When the baseline drift rate is greater than the latent slip judgment threshold, the dynamic displacement tolerance threshold is updated with negative decay.

Images