Anti-dislodgement thoracic drainage tube securement device and system for cardiothoracic surgery

Abstract

This invention relates to the field of medical device technology, specifically to a device and system for fixing anti-slip chest drainage tubes in cardiothoracic surgery. The system's execution steps include: based on the inherent dimensions of the sensor array in the axial and circumferential directions, performing a fusion analysis on the dispersion of sensor coordinates relative to the centroid coordinates, and combining this with pressure data from each sensor to determine the pressure distribution state factor at each moment; constructing gradient vectors and centroid displacement vectors, and calculating the matching degree between them; adaptively establishing a personalized set of reference parameters; and calculating the pressure dispersion deviation and centroid deviation in real time based on the reference parameter set, combining this with the matching degree for fusion analysis to determine the comprehensive risk assessment coefficient at each moment. This invention significantly enhances the system's anti-interference capability and judgment confidence in complex clinical environments, making the final warning output both timely and accurate, and is suitable for fixing medical drainage tubes.

Description

Technical Field

[0001] This invention relates to the field of medical device technology, specifically to a device and system for fixing an anti-slip chest drainage tube used in cardiothoracic surgery. Background Technology

[0002] As the core instrument for achieving this function, the stable placement of the chest drainage tube directly affects the treatment outcome and patient safety. However, in clinical practice, accidental slippage of the drainage tube is a common and extremely risky complication. Once slippage occurs, it can lead to a series of serious consequences, such as partial slippage causing poor drainage, subcutaneous emphysema, or local infection. In clinical medicine, anti-slip devices with mechanical locking structures are commonly used to prevent accidental slippage of the drainage tube. These devices achieve tube fixation by providing a rated clamping pressure at a set position, and their function focuses on physical restraint under static conditions.

[0003] With increasing demands for medical safety, the industry has begun exploring the integration of sensing and alarm functions into fixation devices to proactively monitor the risk of slippage. However, existing technologies still face challenges in terms of the dimensions of perception and the intelligence of signal processing. On the one hand, common single-point or discrete pressure sensors struggle to continuously and comprehensively acquire the spatial distribution of contact pressure between the drainage tube and the fixation interface and its temporal variation, making it easy to miss crucial mechanical information reflecting early micro-displacement of the tube. On the other hand, the clinical environment is complex; patients' physiological and voluntary activities such as breathing, heartbeat, and changes in body position generate continuous and variable pressure fluctuations. The signal characteristics of these fluctuations overlap in the time domain with those caused by abnormal traction on the tube, making effective differentiation difficult using only simple threshold judgments.

[0004] Therefore, current systems often face a dilemma in sensitivity settings: if the sensitivity is set too high to improve the timeliness of early warnings, it will be easily interfered with, leading to frequent false alarms; if the sensitivity is set too low to reduce false alarms, it may delay the early warning of real slip risk. Summary of the Invention

[0005] To address the technical problem mentioned above regarding existing fixation methods for cardiothoracic surgical pleural drainage scenarios, which suffer from the inability to achieve accurate sensing and intelligent early warning under continuous interference such as breathing and heartbeat, resulting in a difficulty in balancing sensitivity and reliability in safety systems, the present invention aims to provide a pleural drainage tube fixation device and system for cardiothoracic surgery to prevent slippage. The specific technical solution adopted is as follows: One embodiment of the present invention provides a fixation system for an anti-slip chest drainage tube in cardiothoracic surgery, comprising a memory and a processor, wherein the processor is used to process instructions stored in the memory to implement the following detection process: The pressure data time series and two-dimensional parameterized coordinates of several flexible thin-film pressure sensors 3 are acquired in real time; the two-dimensional parameterized coordinates include the axial coordinates and circumferential coordinates of the sleeve. Determine the centroid coordinates of the sleeve; based on the inherent dimensions of the sensor array in the axial and circumferential directions, perform a fusion analysis on the dispersion of the sensor coordinates relative to the centroid coordinates, and combine the pressure data of each sensor to determine the pressure distribution state factor at each moment; Construct a gradient vector representing the direction of local pressure change and a centroid displacement vector representing the direction of motion of the overall pressure center, and calculate the degree of matching between the gradient vector and the centroid displacement vector; The statistical characteristics of the pressure distribution state factor and the axial position of the center of mass are obtained by initializing the learning period, and the mechanical design parameters of the sleeve are integrated to adaptively establish a personalized set of reference parameters including state reference, fluctuation range and physical tolerance. Based on the benchmark parameter set, the pressure dispersion deviation and centroid deviation are calculated in real time. Combined with the matching degree, a fusion analysis is performed to determine the comprehensive risk assessment coefficient at each moment, and then the output result of the signal analysis and transmission module 9 is obtained.

[0006] Furthermore, the steps for determining the centroid coordinates include: At the same moment, the pressure data of each sensor is used as the weight of the corresponding sensor's two-dimensional parameterized coordinates. The weighted average method is used to calculate the weighted average of the axial coordinates and the circumferential coordinates, which are used as the centroid coordinates at the corresponding moment. The centroid coordinates include the centroid axial coordinates and the centroid circumferential coordinates.

[0007] Further, determining the pressure distribution state factor at each moment includes: Obtain the total axial span and total circumferential width of the sensor array; The axial coordinates and centroid axial coordinates of each sensor are normalized using the total axial span, and the circumferential coordinates and centroid circumferential coordinates of each sensor are normalized using the total circumferential width. For any given moment, based on the normalized two-dimensional parameterized coordinates and centroid coordinates of each sensor, calculate the squared Euclidean distance from each sensor to the centroid. The contribution value of each sensor is obtained based on the squared Euclidean distance and pressure data of each sensor; The pressure distribution state factor at the corresponding moment is determined based on the contribution value of each sensor.

[0008] Further, determining the pressure distribution state factor at the corresponding time based on the contribution value of each sensor includes: Calculate the sum of the contribution values ​​of all sensors at the same time, as the first sum; and calculate the sum of the pressure data of all sensors at the same time, as the second sum. The pressure distribution state factor at the corresponding time is determined based on the ratio of the first sum to the second sum.

[0009] Furthermore, the construction of the gradient vector characterizing the direction of local pressure change and the centroid displacement vector characterizing the direction of motion of the overall pressure center includes: For any given moment, based on the distribution of the sensor array in the two-dimensional parameter space, the first and second coefficients of the two-dimensional pressure plane model are obtained by fitting the two-dimensional parameterized coordinates of each sensor and the pressure data using the least squares method. The first coefficient and the second coefficient are used as the partial derivatives of the pressure field in the axial and circumferential directions, respectively, and the gradient vector is constructed based on the partial derivatives of the pressure field in the axial and circumferential directions. The axial centroid coordinate difference and circumferential centroid coordinate difference at the first and second moments are obtained to form a centroid displacement vector; the first moment is any of the moments mentioned above, and the second moment is a moment that differs from the first moment by a preset time window.

[0010] Further, calculating the degree of matching between the gradient vector and the centroid displacement vector includes: Calculate the product of the gradient vector and the centroid displacement vector, and use it as the first product; Obtain the magnitudes of the gradient vector and the centroid displacement vector, and determine the second product based on the product of the two magnitudes; The degree of matching is determined based on the first product and the second product.

[0011] Furthermore, the adaptive establishment of a personalized benchmark parameter set, including state benchmarks, fluctuation ranges, and physical tolerances, includes: During the initialization learning period, the centroid axial position and pressure distribution state factor at each moment are acquired; the initialization learning period is at least used to characterize the time period during which the patient remains at rest and the drainage tube is in a stable and normal clamping state. The average values ​​of the centroid axial position and pressure distribution state factor at all times during the initial learning period are calculated to determine the state reference; the state reference includes the position reference and the pressure state reference. Based on the pressure distribution state factor at all times during the initial learning period, the fluctuation range of the pressure distribution state factor is determined using the standard deviation function; Obtain the mechanical design parameters of the sleeve and the sensor axis vector range to determine the physical tolerance of the axial displacement; the mechanical design parameters include at least the mechanical design length and the safety sliding margin.

[0012] Furthermore, the real-time calculation of the pressure dispersion deviation and the centroid deviation based on the reference parameter set includes: For any given moment, calculate the first difference value between the pressure distribution state factor and the pressure state benchmark, and use the ratio of the first difference value to the fluctuation range as the degree of pressure dispersion deviation. Calculate a second difference value between the axial position of the centroid and the position reference, and use the ratio of the second difference value to the physical tolerance as the degree of centroid deviation.

[0013] Further, the output results of the signal analysis and transmission module 9 are obtained, including: For each moment, a risk contribution coefficient is determined, and the product of the risk contribution coefficient and the degree of matching is taken as the degree of directional matching; the risk contribution coefficient is determined by comparing the degree of matching with a preset matching threshold. Based on the degree of directional matching, the degree of pressure dispersion deviation, and the degree of centroid deviation, a comprehensive risk assessment coefficient is determined for each moment. If the comprehensive risk assessment coefficient is greater than the assessment threshold for a consecutive preset number of time periods, it is determined to be fixed abnormal; otherwise, it is determined to be fixed normal. The determination result is used as the output result of the signal analysis and transmission module 9.

[0014] Another embodiment of the present invention provides a fixation device for an anti-slip chest drainage tube in cardiothoracic surgery, comprising: Wearing mechanism 5 is used to fix the device to the patient's body via end Velcro 8; An airbag-type clamping mechanism 1 is fixedly installed on the wearable mechanism 5, and a fixing groove 6 is provided on one or both sides of it. A detachable unlocking block 2 is provided, which is equipped with a silicone clamping mechanism 4. The unlocking block 2 is configured to be able to be inserted into the fixing groove 6, so that the silicone clamping mechanism 4 fits with the airbag clamping mechanism 1 to form a sleeve for clamping the drainage tube. A press-type elastic self-locking mechanism 7 is provided between the unlocking block 2 and the fixing groove 6, which is used to lock the two when the unlocking block 2 is inserted into the position, and to release the lock when it is pressed again so that the unlocking block 2 can pop out. Multiple flexible thin-film pressure sensors 3 are arrayed on the inner surface of the sleeve formed by the airbag clamping mechanism 1 and the silicone clamping mechanism 4, for monitoring the contact pressure between the drainage tube and the sleeve. The signal analysis and transmission module 9 is electrically connected to the plurality of flexible thin-film pressure sensors 3 and is used to implement the execution steps of the anti-slip chest drainage tube fixation system used in cardiothoracic surgery. The inflation module 11 is connected to the airbag clamping mechanism 1 via the air tube 10; and the control unit is configured to: when the signal analysis and transmission module 9 analyzes and determines that the contact pressure signal is abnormal, control the inflation module 11 to inflate the airbag clamping mechanism 1 with gas to increase the volume of the airbag clamping mechanism 1, thereby enhancing the clamping force on the drainage tube.

[0015] The present invention has the following beneficial effects: This invention provides a fixation device and system for an anti-slip chest drainage tube in cardiothoracic surgery. The device integrates a passive elastic fixation silicone clamping mechanism and an active controllable pressurization airbag clamping mechanism. The silicone and airbag together provide stable static friction. In case of warning, the airbag can be inflated instantly to significantly and rapidly increase the radial clamping force, forming a powerful active braking. In other words, once the risk is confirmed, physical intervention can be automatically implemented to suppress slippage in its early stages, fundamentally improving the safety limit.

[0016] This system acquires real-time pressure data time series and two-dimensional parameterized coordinates, which is beneficial for establishing a data foundation for spatiotemporal analysis. It solves the problem of a single sensing dimension, enabling subsequent analysis to extract information from spatial distribution patterns. This provides the possibility of capturing local distortions and overall shifts in the pressure field caused by drainage tube displacement, a fundamental prerequisite for early and accurate warnings. Determining the pressure distribution state factor allows for dynamic quantification of the uniformity of the two-dimensional pressure field distribution, enabling sensitive detection of early, localized contact anomalies and significantly advancing the warning window, compensating for the shortcomings of relying solely on centroid movement for judgment. Constructing and calculating the matching degree between the gradient vector and the centroid displacement vector fundamentally solves the technical bottleneck of distinguishing between physiological interference and abnormal traction. True slippage is characterized by highly consistent directions of the two vectors, while interferences such as breathing and coughing do not possess this stable directional correlation. Through this analysis, the system can effectively distinguish between directional slippage signals and non-directional interference signals, fundamentally reducing false alarms caused by interference. By establishing a personalized baseline parameter set during the initial learning period, a personalized quantitative baseline and physical safety boundary for normal state are adaptively established for each patient and each installation. This completely solves the clinical problem that fixed and uniform thresholds cannot adapt to individual differences, making the warning threshold both individually adaptable and objectively safe. This significantly improves the system's universality for different patients and the clinical acceptability of the warning results. Based on the multi-source information fusion calculation of the comprehensive risk assessment coefficient, the analysis results of all dimensions are integrated, greatly enhancing the system's anti-interference ability and judgment confidence in complex clinical environments. This makes the final warning output both timely and accurate, providing a reliable basis for initiating active braking, thus solving the ultimate problems of limited braking capacity and poor warning reliability in a closed loop. Attached Figure Description

[0017] To more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art 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.

[0018] Figure 1 This is a schematic diagram of the overall structure of an anti-slip chest drainage tube fixation device for cardiothoracic surgery, according to an embodiment of the present invention. Figure 2 This is a top view of the device in an embodiment of the present invention; Figure 3 This is a front view of the device in an embodiment of the present invention; Figure 4 This is a front view of the device in an embodiment of the present invention; Figure 5 An embodiment of the present invention provides an execution flowchart of an anti-slippage chest drainage tube fixation system for cardiothoracic surgery; Reference numerals: 1 is an airbag clamping mechanism, 2 is a detachable unlocking block, 3 is a flexible thin film pressure sensor, 4 is a silicone clamping mechanism, 5 is a wearable mechanism, 6 is a fixing groove, 7 is a press-type elastic self-locking mechanism, 8 is a Velcro strap, 9 is a signal analysis and transmission module, 10 is an air tube, and 11 is an inflation module. Detailed Implementation

[0019] To further illustrate the technical means and effects adopted by the present invention to achieve its intended purpose, the specific implementation methods, structures, features, and effects of the technical solution proposed according to the present invention are described in detail below with reference to the accompanying drawings and preferred embodiments. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. Furthermore, specific features, structures, or characteristics in one or more embodiments can be combined in any suitable form.

[0020] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains.

[0021] The application scenarios targeted by this invention can be: In cardiothoracic surgery and intensive care, thoracic drainage is a crucial treatment method used to drain gas, blood, exudate, or pus from the patient's pleural cavity to maintain negative pressure, promote lung re-expansion, and prevent infection. Accidental slippage of the thoracic drainage tube is a common and extremely risky complication. Existing anti-slip devices only have a locking mechanism with a fixed clamping pressure on the drainage tube, offering limited braking capability in the event of slippage or other unexpected situations. Furthermore, because they cannot achieve accurate sensing and intelligent early warning under continuous disturbances such as breathing and heartbeat, the safety system struggles to balance sensitivity and reliability.

[0022] One embodiment of the present invention provides a fixation device for an anti-slip chest drainage tube in cardiothoracic surgery, the overall structure of which is shown in the schematic diagram below. Figure 1 As shown, the top view of the device is as follows Figure 2 As shown, the main view of the device is as follows: Figure 3 As shown, the front view of the device is as follows Figure 4 As shown.

[0023] In this embodiment, the device includes: a wearing mechanism 5, used to fix the device to the patient's body via end Velcro 8; An airbag-type clamping mechanism 1 is fixedly installed on the wearable mechanism 5 and cannot be moved. It has a fixing groove 6 on one or both sides. A detachable unlocking block 2 is provided, which is equipped with a silicone clamping mechanism 4. The unlocking block 2 is configured to be inserted into the fixing groove 6, so that the silicone clamping mechanism 4 fits against the airbag clamping mechanism 1 to form a sleeve for clamping the drainage tube. A press-type elastic self-locking mechanism 7 is provided between the unlocking block 2 and the fixing groove 6 to lock the two when the unlocking block 2 is inserted into place, and to release the lock when pressed again so that the unlocking block 2 can pop out. The inner diameter of the near-circular hole formed by the silicone clamping mechanism 4 and the airbag clamping mechanism 1 is slightly smaller than the outer diameter of the drainage tube, which can play a role in fixing and clamping the drainage tube. Multiple flexible thin-film pressure sensors 3 are arrayed on the inner surface of the sleeve formed by the airbag clamping mechanism 1 and the silicone clamping mechanism 4 to monitor the contact pressure between the drainage tube and the sleeve. The flexible thin-film pressure sensors 3 can be arranged in two groups of eight uniform arrays that are parallel to each other and parallel to the vertical cross section of the inner hole. The real-time pressure monitored can be transmitted to the signal analysis and transmission module 9 for analysis and processing. The signal analysis and transmission module 9 is electrically connected to the plurality of flexible thin-film pressure sensors 3 and is used to implement the steps of the anti-slippage chest drainage tube fixation system for cardiothoracic surgery; wherein, the processed information can be transmitted quickly through the Internet of Things. An inflation module 11 is installed in the wearable mechanism 5 and connected to the airbag clamping mechanism 1 via an air tube 10. Its working principle is similar to that of the inflation module of a portable blood pressure monitor. A control unit is configured to: when the signal analysis and transmission module 9 analyzes and determines that the contact pressure signal is abnormal, control the inflation module 11 to inflate the airbag clamping mechanism 1 with gas to increase the volume of the airbag clamping mechanism 1, thereby enhancing the clamping force of the clamping structure formed by the airbag clamping mechanism 1 and the silicone clamping mechanism 4 on the drainage tube.

[0024] In summary, one embodiment of the present invention provides a fixation device for an anti-slippage chest drainage tube in cardiothoracic surgery. This device integrates a passive elastic fixation silicone clamping mechanism and an active, controllable pressurization airbag clamping mechanism. The silicone and airbag jointly provide stable static friction. Upon warning, the airbag can instantly inflate, significantly and rapidly increasing the radial clamping force, forming a powerful active braking effect. In other words, once a risk is confirmed, automatic physical intervention can be implemented to suppress slippage in its early stages, fundamentally improving the safety ceiling. Multiple flexible thin-film pressure sensors are arranged in a two-dimensional array on the inner surface of the sleeve. This allows for real-time acquisition of the spatial distribution map of the contact pressure between the drainage tube and the fixation interface, rather than single-point values. It also facilitates the detection of early abnormal patterns: by analyzing the morphological changes in the pressure distribution map, early signs of slippage such as unilateral tension and localized detachment can be sensitively identified before macroscopic displacement of the drainage tube, greatly advancing the warning window. By setting up signal analysis and transmission modules, the system can process multi-dimensional spatiotemporal data, fundamentally solving the paradox of the incompatibility between sensitivity and specificity, and achieving high-sensitivity early warning with low false alarm rate.

[0025] An embodiment of the present invention also provides a fixation system for an anti-slip thoracic drainage tube in cardiothoracic surgery, including a memory and a processor, wherein the processor is used to process instructions stored in the memory to implement the following detection process, such as... Figure 5 As shown, it includes: S1, real-time acquisition of pressure data time series and two-dimensional parameterized coordinates from several flexible thin-film pressure sensors 3.

[0026] Drainage tube slippage is essentially an unexpected axial displacement or rotation of the tube within the sleeve. This physical movement inevitably leads to a specific spatial change in the pressure distribution at the contact interface between the tube and the sleeve. For example, axial traction causes the overall pressure distribution to shift to one side; localized compression causes localized spikes in the pressure distribution. Therefore, the spatial properties of this shift and spikes can be quantified using location information. Furthermore, pressure fluctuations caused by patient breathing and coughing are usually global and periodic, while early slippage or localized traction of the drainage tube is usually localized and trend-like. Only by simultaneously collecting and analyzing the pressure values ​​and location from the sensors can the spatial distribution of the pressure field be analyzed, thereby monitoring the current fixation status of the drainage tube.

[0027] The first step is to acquire the time series of pressure data from several flexible thin-film pressure sensors 3 in real time.

[0028] In this embodiment, a multi-channel synchronous data acquisition method is used to acquire signals from eight flexible thin-film pressure sensors evenly distributed on the inner wall of the clamping sleeve. This obtains the voltage signals from all pressure sensors, and each pressure sensor then converts its voltage signal into an instantaneous pressure value, thereby acquiring the raw pressure data of each sensor at different times. The analysis can be performed on any patient requiring real-time monitoring of the chest drainage tube's fixation. The sampling period can be set to the current 5 minutes, and the sampling frequency can be set to 50Hz.

[0029] For the raw pressure data time series of each pressure sensor, a low-pass digital filter is used for filtering to eliminate high-frequency noise and obtain smooth pressure data, which constitutes the pressure data time series of the corresponding pressure sensor.

[0030] The cutoff frequency of the low-pass digital filter is determined based on the upper limit of physiological activity frequency, and can be empirically set to 3Hz. This frequency aims to retain low-frequency trend signals caused by drainage tube displacement while filtering out high-frequency physiological noise caused by heartbeat, muscle tremors, etc. In practical applications, this cutoff frequency can be adaptively adjusted according to the patient's physiological characteristics (such as average heart rate) to optimize the signal-to-noise ratio.

[0031] It should be noted that all pressure data mentioned in the subsequent implementation process are filtered pressure data.

[0032] The second step is to acquire the two-dimensional parameterized coordinates of several flexible thin-film pressure sensors 3 in real time.

[0033] In this embodiment, the sensor array is distributed in a non-closed area. Each pressure sensor has a fixed two-dimensional parametric coordinate on the inner wall of the clamping sleeve. These two-dimensional parametric coordinates include the axial and circumferential coordinates of the sleeve, which can be denoted as... .in, This indicates the position coordinates of the pressure sensor along the axial direction of the sleeve (the length direction of the drainage tube), and the range of values ​​is consistent with the axial deployment range of the sensor array; This represents the maximum envelope angle of the sensor array after circumferential deployment, i.e., the circumferential deployment coordinates. This value is determined by the physical curvature of the device and is less than 360°.

[0034] Step S1 described above acquires the pressure data time series and two-dimensional parameterized coordinates in real time, establishing the data foundation for spatiotemporal analysis. It binds the pressure reading of each sensor to its precise spatial position on the inner wall of the sleeve, upgrading discrete point-like perception to two-dimensional field perception of the contact surface. It solves the problem of single perception dimension by organizing the pressure data into a two-dimensional array with spatial attributes, enabling subsequent analysis to extract information from the spatial distribution pattern. This provides the possibility of capturing local distortion and overall shift of the pressure field caused by the displacement of the drainage tube, which is the fundamental premise for achieving early and accurate early warning.

[0035] S2, determine the centroid coordinates of the sleeve; based on the inherent dimensions of the sensor array in the axial and circumferential directions, perform a fusion analysis on the dispersion of the sensor coordinates relative to the centroid coordinates, and combine the pressure data of each sensor to determine the pressure distribution state factor at each moment.

[0036] To accurately monitor the relative spatial position of the drainage tube within the clamping sleeve, a global characteristic analysis of the contact pressure distribution between the tube and the inner surface of the sleeve is necessary. Patients can still perform limited movements after the drainage tube is placed, which may cause slight movement of the tube within the sleeve. Furthermore, when the drainage tube is subjected to abnormal traction, its contact with the inner wall of the sleeve changes from a uniform fit to a non-uniform state of localized tightness and looseness. While relying solely on the centroid coordinates of the pressure distribution can reflect the overall translation of the distribution center, it is not sensitive to changes in the shape of the pressure distribution and is difficult to effectively detect early, localized contact abnormalities.

[0037] Therefore, this embodiment introduces a pressure distribution state factor to simultaneously characterize the dispersion and uniformity changes of the pressure distribution. When the drainage tube slips or undergoes abnormal displacement, it will simultaneously cause two key changes in the pressure distribution: first, a significant shift in the pressure distribution center; and second, an abnormal increase in the dispersion of the pressure distribution. By constructing a pressure distribution state factor to quantify the spatial dispersion characteristics of the pressure field, abnormal changes in the pressure distribution morphology can be comprehensively captured. This allows for sensitive early warning of slippage trends even before a significant shift in the pressure distribution centroid, improving the system's ability to detect local abnormal states.

[0038] The first step is to determine the coordinates of the sleeve's center of mass.

[0039] The centroid coordinates are also the pressure center coordinates. When determining the pressure center, the greater the pressure at a point (pressure data from a sensor), the tighter the drainage tube is compressed at that point, and the greater the contribution of that location to the overall displacement. By using pressure weighting, the centroid coordinates can reflect the point of force concentration on the drainage tube. When traction slippage occurs, the force point will move in the direction of traction, and the centroid coordinates will shift accordingly, thus allowing the slippage direction to be monitored through the centroid displacement.

[0040] Specifically, for the same moment, the pressure data of each sensor is used as the weight of the corresponding sensor's two-dimensional parameterized coordinates. The weighted average method is used to calculate the weighted average of the axial coordinates and the circumferential coordinates, which are used as the centroid coordinates at the corresponding moment. The centroid coordinates include the centroid axial coordinates and the centroid circumferential coordinates.

[0041] In this embodiment, based on the pressure data and two-dimensional parameterized coordinates of each sensor at time t, the pressure data of each sensor is used as the weight of the corresponding sensor's two-dimensional parameterized coordinates. A weighted average method is used to calculate the weighted average values ​​of the axial and circumferential coordinates, and the weighted average result is used as the centroid coordinate at time t, denoted as . Where t is a positive integer, This represents the centroid axis coordinate at time t. Let t represent the circumferential coordinates of the centroid at time t; the implementation process of the weighted average method is existing technology and will not be described in detail here.

[0042] The second step involves performing a fusion analysis on the dispersion of sensor coordinates relative to the centroid coordinates based on the inherent dimensions of the sensor array in the axial and circumferential directions, and combining the pressure data from each sensor to determine the pressure distribution state factor at each moment.

[0043] Here, the pressure distribution state factor is used to characterize at least the degree of dispersion of the pressure distribution, that is, the degree of dispersion of the contact pressure between the drainage tube and the tube wall. When normally clamped, the pressure distribution state factor is relatively stable, but when slippage or pulling occurs, the morphology of the contact surface is distorted, such as unilateral compression or local detachment, causing the pressure distribution to deviate from the reference value.

[0044] As an exemplary implementation, determining the pressure distribution state factor for any given time includes: The first sub-step is to obtain the total axial span and total circumferential width of the sensor array.

[0045] Here, the total axial span and total circumferential width are both ranges, used for subsequent normalization processing.

[0046] In this embodiment, the total axial span refers to the difference between the maximum and minimum axial coordinates among all sensors; the total circumferential width refers to the difference between the maximum and minimum circumferential coordinates among all sensors. These two values ​​are inherent physical parameters of the sensor array, used to eliminate dimensional differences in the coordinate axes.

[0047] The second sub-step involves normalizing the axial coordinates and centroid axial coordinates of each sensor using the total axial span, and normalizing the circumferential coordinates and centroid circumferential coordinates of each sensor using the total circumferential width.

[0048] Here, coordinate normalization is performed to address the inconsistency between the axial and circumferential dimensions after the cylindrical surface is unfolded, stretching it to be isotropic for distance calculation.

[0049] In this embodiment, the axial coordinate and centroid axial coordinate of each sensor are divided by the total axial span, such as... , Let L represent the axial coordinate of the i-th sensor after normalization, and L represent the total axial span. Simultaneously, divide the circumferential coordinate and centroid circumferential coordinate of each sensor by the total circumferential width, as shown below. , This represents the circumferential coordinates of the i-th sensor after normalization. This indicates the total circumferential width.

[0050] It should be noted that, in subsequent calculations involving two-dimensional parametric coordinates and centroid coordinates, normalized two-dimensional parametric coordinates and centroid coordinates will be used by default to avoid invalid analysis caused by different dimensions.

[0051] The third sub-step involves calculating the squared Euclidean distance from each sensor to the centroid for each sensor at any given time, based on the normalized two-dimensional parameterized coordinates and centroid coordinates of each sensor.

[0052] Here, the squared Euclidean distance refers to the geometric deviation of each sensor relative to the pressure center within a unit space. Due to normalization, the distance reflects the deviation of the relative position.

[0053] In this embodiment, the formula for calculating the squared Euclidean distance from the i-th sensor to the centroid can be: In the formula, This represents the squared Euclidean distance from the i-th sensor to the centroid. This represents the normalized centroid axis coordinates. This represents the normalized circumferential coordinates of the centroid.

[0054] The fourth sub-step involves obtaining the contribution value of each sensor based on the squared Euclidean distance of each sensor and the pressure data.

[0055] Here, we introduce pressure weights. The greater the pressure, the greater the impact of the point's positional deviation on the overall shape.

[0056] In this embodiment, the formula for calculating the contribution value of the i-th sensor can be: In the formula, This represents the contribution value of the i-th sensor. This represents the pressure data from the i-th sensor.

[0057] It should be noted that the contribution of a sensor to the overall dispersion depends on the magnitude of its pressure data and its distance from the center of mass. Points with high pressure and far from the center of mass are the main factors causing uneven distribution and have a larger contribution value.

[0058] The fifth sub-step involves determining the pressure distribution state factor at the corresponding moment based on the contribution value of each sensor.

[0059] Here, the pressure distribution state factor reflects the contact topology of the drainage tube inside the sleeve. When slippage causes non-uniform distortion of the contact surface (such as unilateral dead pressure or local voiding), this index will deviate significantly from the resting reference value.

[0060] In this embodiment, the pressure distribution state factor at any given moment is calculated based on the contribution value of each sensor and the pressure data.

[0061] Specifically, the sum of the contribution values ​​of all sensors at the same time is calculated as the first sum; and the sum of the pressure data of all sensors at the same time is calculated as the second sum; based on the ratio of the first sum and the second sum, the pressure distribution state factor at the corresponding time is determined.

[0062] Furthermore, when the cumulative pressure data from all sensors is not zero, the formula for calculating the pressure distribution state factor at time t can be: In the formula, Let I represent the pressure distribution state factor at time t, and let I represent the number of sensors. This represents the first sum at time t. This represents the second sum value at time t.

[0063] It should be noted that a smaller pressure distribution state factor indicates that the pressure is concentrated at a single center, while a larger pressure distribution state factor indicates that the pressure distribution is dispersed, or that there are high-pressure points far from the center. This may lead to unilateral traction or localized detachment, and a greater likelihood of early slippage. It is also worth noting that when the cumulative value of the pressure data from all sensors is zero, it indicates an abnormal clamping state, and a warning can be issued directly.

[0064] Referring to the calculation process of the pressure distribution state factor at time t above, the pressure distribution state factor at each time can be obtained.

[0065] Step S2 above determines the pressure distribution state factor, which can dynamically quantify the uniformity of the two-dimensional pressure field distribution. By calculating the pressure weighted centroid and analyzing the dispersion of each sensor position relative to the centroid, a dynamic index characterizing the degree of concentration or dispersion of pressure distribution is generated. It enables sensitive detection of early and local contact anomalies, greatly advancing the warning window period and making up for the shortcomings of relying solely on the movement of the centroid for judgment.

[0066] S3. Construct a gradient vector representing the direction of local pressure change and a centroid displacement vector representing the direction of motion of the overall pressure center, and calculate the degree of matching between the gradient vector and the centroid displacement vector.

[0067] In its physical nature, drainage tube slippage is a continuous traction process with a clear direction, which creates a pressure gradient pattern with specific spatial orientation in the pressure field. Specifically, when the drainage tube is subjected to abnormal traction, the contact pressure between it and the sleeve will exhibit a regular gradient change along the traction direction, and the main direction of this pressure gradient is highly consistent with the overall displacement direction of the drainage tube. Conversely, pressure changes caused by patient positioning, local compression, or physiological activities (such as breathing and coughing), although they may also cause instantaneous shifts in the centroid coordinates or temporary increases in distribution dispersion, usually do not possess a stable and consistent spatial orientation in their pressure field evolution, or are not significantly correlated with the direction of centroid movement.

[0068] Therefore, relying solely on global statistics such as centroid location and dispersion is insufficient to effectively distinguish the fundamentally different physical modes mentioned above. This invention achieves a quantitative analysis of the directional characteristics of the pressure field by constructing and calculating the degree of matching between the pressure gradient vector and the centroid displacement vector. This degree of matching characterizes the consistency between the direction of the fastest local pressure change and the overall direction of pressure center movement. When the degree of matching remains high, it indicates that the changes in the pressure field have clear directional cooperative characteristics, highly consistent with the physical process of slippage; while when the degree of matching is low, it indicates that the pressure changes are more likely to originate from disturbances without clear directionality.

[0069] By introducing and analyzing this matching degree, the system can effectively distinguish between real slip and various non-directional interferences from the spatial correlation dimension of the signal, thereby significantly reducing the false alarm rate and improving the overall accuracy and reliability of state determination while maintaining high early warning sensitivity.

[0070] The first step is to construct a gradient vector representing the direction of local pressure change and a centroid displacement vector representing the direction of motion of the overall pressure center.

[0071] As an exemplary implementation, constructing the gradient vector and centroid displacement vector includes: The first sub-step involves, for any given moment, using the distribution of the sensor array in the two-dimensional parameter space, and employing the two-dimensional parameterized coordinates of each sensor and pressure data, fitting the first and second coefficients of the two-dimensional pressure plane model using the least squares method.

[0072] In this embodiment, based on the two-dimensional distribution of the sensor array on the inner wall of the sleeve, the pressure field can be approximated as a continuous two-dimensional plane in the parameter space. Correspondingly, in the two-dimensional parameter space... The two-dimensional parametric coordinates of each sensor and their corresponding pressure data are used to construct a local two-dimensional pressure plane model through the least squares fitting method.

[0073] The local two-dimensional pressure plane model specifically includes: In the formula, and Let represent the normalized axial and circumferential coordinates, and a, b, and c be the fitting coefficients to be determined. The fitting coefficients a, b, and c can be solved by the least squares method to minimize the sum of squared residuals between the fitting plane and the actual pressure values ​​P of all sensors.

[0074] The second sub-step involves using the first and second coefficients as the partial derivatives of the pressure field in the axial and circumferential directions, respectively, and constructing a gradient vector based on the partial derivatives of the pressure field in the axial and circumferential directions.

[0075] Here, the gradient vector can reflect the principal direction of the local pressure gradient, which in turn reflects the direction of the fastest pressure change in a local region of the pressure field and is the core indicator characterizing the directional characteristics of the pressure field; the magnitude of the gradient vector can reflect the rate of pressure change.

[0076] In this embodiment, the fitting coefficients a and b have explicit physical meanings, corresponding to the partial derivatives of the pressure field in the axial and circumferential directions, respectively. Based on the partial derivatives, an approximate gradient vector of the pressure field at the corresponding moment can be constructed, i.e. If the fitting coefficient 'a' can be the first coefficient, then the fitting coefficient 'b' can be the second coefficient.

[0077] The third sub-step involves obtaining the difference between the axial and circumferential centroid coordinates at the first and second time points, which forms the centroid displacement vector.

[0078] Here, the first moment is any moment, and the second moment is the moment that differs from the first moment by a preset time window. The preset time window can be set to 0.2s. This duration aims to balance response speed and signal smoothness; it needs to be short enough to capture the onset of the slip event, and long enough to avoid misinterpreting instantaneous noise as displacement trends. This value can be adjusted according to clinical needs and data sampling rate. The preset time window can contain 10 sampling points. The centroid displacement vector can reflect the movement trend of the global pressure center.

[0079] In this embodiment, the normalized axial centroid coordinates at the first time point (time t) and the second time point (time t) are calculated. The difference between the normalized axial centroid coordinates is denoted as . And calculate the difference between the normalized circumferential centroid coordinates at the first time step and the normalized circumferential centroid coordinates at the second time step, denoted as . Therefore, the centroid is obtained within the preset time window. Displacement vector within, displacement vector of the centroid .

[0080] The direction of the centroid displacement vector is the centroid within the preset time window. The direction of movement within the centroid, and the magnitude of the centroid displacement vector can reflect the distance the centroid has moved.

[0081] The second step is to calculate the degree of matching between the gradient vector and the centroid displacement vector.

[0082] Analysis of the physical process of slippage reveals that actual drainage tube slippage manifests as continuous traction along a specific direction. This process induces a pressure change with clear spatial directionality on the sensor array, generating a directionally stable pressure gradient vector. The gradient direction points towards the region of increased pressure, and its opposite direction is the direction of the slippage force. Simultaneously, this pulling process inevitably leads to a change in the overall spatial position of the drainage tube, manifested as a displacement vector with the center of mass moving in the same direction. .

[0083] For non-directional disturbances (such as postural adjustments, localized compression, or physiological activities), although they may cause instantaneous pressure gradients... The modulus increases, but its gradient direction mostly points towards the instantaneous point of compression, rather than the overall displacement direction of the center of mass. There is no stable and significant spatial correlation between them.

[0084] Therefore, this embodiment calculates the pressure gradient vector. With the centroid displacement vector The degree of matching between them is used to quantify the consistency of their directions.

[0085] As an exemplary implementation, calculating the degree of matching includes: The first sub-step is to calculate the product of the gradient vector and the centroid displacement vector, which is then used as the first product.

[0086] The second sub-step involves obtaining the magnitudes of the gradient vector and the centroid displacement vector, and determining the second product based on the product of the two magnitudes.

[0087] The third sub-step determines the degree of matching based on the first and second products.

[0088] As an example, the formula for calculating the degree of matching between the gradient vector and the centroid displacement vector at time t can be: In the formula, This indicates the degree of matching between the gradient vector and the centroid displacement vector at time t. This represents the modulus function. Represents the first product. Indicates the second product. This represents a non-zero constant, used to avoid the case where the denominator of a fraction is zero. Its empirical value can be 0.1.

[0089] In the formula for calculating the degree of matching, the value of the degree of matching ranges from -1 to 1. The closer the degree of matching is to 1, the more consistent the direction of the pressure gradient is with the direction of the centroid movement, and the more likely there is a risk of directional traction.

[0090] Step S3 above constructs and calculates the matching degree between the gradient vector and the centroid displacement vector. It extracts the vector features of pressure field changes and performs correlation analysis. Specifically, it fits the local pressure gradient and calculates the centroid displacement, and quantifies the consistency of the two directions. Its core solution is to overcome the technical bottleneck of the difficulty in distinguishing between physiological interference and abnormal traction. True slippage is characterized by a high degree of consistency between the two vector directions, while interference such as breathing and coughing does not have this stable directional correlation. Through this analysis, the system can effectively distinguish between directional slippage signals and non-directional interference signals, fundamentally reducing false alarms caused by interference.

[0091] S4 obtains the statistical characteristics of the pressure distribution state factor and the axial position of the center of mass during the initial learning period, and integrates the mechanical design parameters of the sleeve to adaptively establish a personalized set of reference parameters including state reference, fluctuation range and physical tolerance.

[0092] From the perspective of the physical mechanism of slippage, the initial slippage will lead to two typical pressure distribution changes: first, the contact area between the drainage tube and the sensor will be distorted, resulting in an abnormal increase in the pressure distribution state factor; second, the drainage tube will shift along the axis, causing the axial position of the centroid to continuously deviate from the initial position. However, different patients have different body sizes, drainage tube models and clamping forces, and a fixed uniform threshold cannot be adapted to all clinical scenarios.

[0093] To achieve accurate and personalized monitoring of the risk of drainage tube slippage, this embodiment provides an adaptive baseline parameter establishment method. This method aims to overcome the problem of poor adaptability of a fixed, uniform threshold caused by differences in individual patient characteristics, drainage tube specifications, and clamping force.

[0094] As an exemplary implementation, obtaining a personalized set of benchmark parameters includes: The first step is to obtain the centroid axial position and pressure distribution state factor at each moment during the initial learning period.

[0095] Here, the initial learning period is used to characterize the time during which the patient remains at rest and the drainage tube is in a stable, normal clamping position. The centroid axial position and pressure distribution state factor are calculated at each moment during the initial learning period to facilitate subsequent statistical analysis and establish three core physical baseline parameters.

[0096] In this embodiment, after the drainage tube fixation device is installed and the patient is at rest, a preset initialization learning period begins. During this period, the system continuously collects data from multiple pressure sensors and calculates the centroid axial position and pressure distribution state factor in real time based on this data.

[0097] The system enters an initial learning phase, continuously monitoring the statistical characteristics (such as mean and variance) of the centroid axial position and pressure distribution state factors. The learning phase ends when these statistical parameters converge under a preset stability criterion, or when a preset maximum learning time (e.g., 5 minutes) is reached, with the first condition met becoming the benchmark. This adaptive mechanism ensures the validity of the baseline, while the patient remains at rest during this phase, and the drainage tube is in a stable, normal clamping position.

[0098] The second step is to calculate the average value of the centroid axial position and pressure distribution state factor at all times during the initial learning period to determine the state baseline.

[0099] Here, the condition reference includes the position reference and the pressure condition reference.

[0100] In this embodiment, after the learning period ends, the system performs statistical analysis on all collected centroid axial positions and pressure distribution state factors, calculates the average value of the centroid axial position at all times during the initial learning period, and calculates the pressure distribution state factor at all times during the initial learning period to obtain individualized statistical baseline parameters, including: centroid axial position baseline and pressure distribution state factor baseline values.

[0101] The third step is to determine the fluctuation range of the pressure distribution state factor based on the pressure distribution state factor at all times during the initial learning period using the standard deviation function.

[0102] In this embodiment, k=3 is set based on empirical values, which represents k times the standard deviation of the dispersion data during the initial learning period. This allows us to obtain the normal fluctuation range of the pressure distribution state factor, which can be expressed as: In the formula, This indicates the fluctuation range of the pressure distribution state factor during the initial learning period. Represents the standard deviation function. Let T represent the pressure distribution state factor at time t, and let T represent the number of all times during the initial learning period.

[0103] The k value is set with reference to the 3-sigma principle in statistical process control. This range can cover about 99.7% of normal physiological fluctuations. The k value can be adjusted to change the sensitivity of the warning. A smaller k value will make the system more sensitive to small changes.

[0104] The fourth step is to obtain the mechanical design parameters of the sleeve and the sensor axis vector range, and determine the physical tolerance of the axial displacement.

[0105] Here, the mechanical design parameters include at least the mechanical design length and the safety sliding margin. The physical tolerance, also known as the axial displacement tolerance, is a core physical parameter indicating whether the axial displacement of the center of mass is abnormal. It maps the maximum safe sliding distance allowed by the sleeve design to a threshold value in the sensor coordinate system. Its value is objectively determined by the mechanical design, rather than by empirical fitting.

[0106] The closer the actual axial displacement of the drainage tube is to the device's allowable safe sliding margin, the higher the physical risk of complete slippage. In other words, the smaller the safety margin, the smaller the allowable change in sensor readings should be. By determining the physical tolerance of the axial displacement, the actual physical safe sliding distance can be converted into a threshold value in the sensor coordinate system.

[0107] Specifically, the effective axial length of the sleeve and its allowable safe sliding margin are obtained, and the axial displacement tolerance is calculated by combining the axial vector range of the sensor array and the proportional relationship.

[0108] In this embodiment, the effective axial length of the inner wall of the sleeve that can effectively clamp the drainage tube is first obtained according to the prior design specifications, and is denoted as the mechanical design length of the clamping sleeve. Next, obtain the maximum safe sliding distance that the drainage tube can slide within the sleeve, and record it as the allowable safe sliding margin. And the range of coordinate values ​​of the sensor array along the axial direction, denoted as the sensor axial direction. Determine the range; then calculate the allowable safety sliding margin. With mechanical design length The ratio, and calculate the ratio and sensor axial direction. The product of the ranges is used as the axial displacement tolerance, denoted as . .

[0109] As an example, the formula for calculating the physical tolerance of axial displacement can be: In the formula, Indicates the physical tolerance for axial displacement. This represents the axial vector path of the sensor array.

[0110] Ultimately, the system integrates individualized statistical baseline parameters and physical tolerance parameters to form a personalized baseline parameter set for a specific patient's usage scenario.

[0111] It should be noted that by constructing a personalized set of benchmark parameters, the monitoring benchmark can be transformed from a fixed threshold to a combination of individual adaptation and physical constraints, which significantly improves the applicability, early warning accuracy and safety of the system in different clinical scenarios.

[0112] Step S4 above establishes a personalized baseline parameter set during the initialization learning period. This set adaptively establishes a personalized quantitative baseline and physical safety boundary for each patient and each installation. During the resting period, it learns the statistical baseline and calculates the physical tolerance in combination with the sleeve design parameters. This completely solves the clinical problem that a fixed uniform threshold cannot adapt to individual differences. It makes the warning threshold both individual adaptability and objective safety, significantly improving the system's universality for different patients and the clinical acceptability of the warning results.

[0113] S5 calculates the degree of pressure dispersion and centroid deviation in real time based on the benchmark parameter set, and performs fusion analysis in combination with the degree of matching to determine the comprehensive risk assessment coefficient at each moment, thereby obtaining the output results of the signal analysis and transmission module 9.

[0114] By fusing and comprehensively evaluating multi-dimensional information, the risk of drainage tube slippage can be accurately and reliably determined. Its core function is to address the problems of susceptibility to interference and high false alarm rates associated with single-feature assessments, thereby balancing the sensitivity and specificity of early warning in complex clinical environments. The fusion analysis here refers to integrating evidence from three dimensions. A high probability of slippage risk is only determined when the pressure distribution dispersion significantly increases, the centroid position continuously shifts, and the gradient vector and centroid displacement vector show a high degree of matching.

[0115] This embodiment calculates the comprehensive risk assessment coefficient at each moment by weighting and combining the three parameters or by integrating them logically based on rules. This coefficient integrates multiple information such as abnormal amplitude, directional correlation and persistence, forming a more robust decision indicator.

[0116] As an exemplary implementation, the output of the signal analysis and transmission module 9 is obtained, including: The first step is to calculate the degree of pressure dispersion and centroid deviation in real time based on the benchmark parameter set.

[0117] During real-time monitoring, the system calculates the normalized deviation of two key physical quantities based on a pre-established set of personalized reference parameters: first, the deviation of the pressure distribution dispersion from its individualized reference (i.e., the normal fluctuation range); and second, the deviation of the centroid axial position from its individualized reference (i.e., the physical safety tolerance). These two deviations quantify the difference between the current state and the safety baseline from two independent dimensions: contact uniformity and overall position.

[0118] Specifically, for any given moment, the first difference value between the pressure distribution state factor and the pressure state reference is calculated, and the ratio of the first difference value to the fluctuation range is taken as the degree of pressure dispersion deviation; the second difference value between the centroid axial position and the position reference is calculated, and the ratio of the second difference value to the physical tolerance is taken as the degree of centroid deviation.

[0119] As an example, the formula for calculating the degree of pressure dispersion can be: In the formula, This indicates the degree of pressure dispersion deviation at time t. This represents the pressure distribution state factor at time t. Indicates the pressure condition reference. This represents the function for finding the absolute value. Indicates the first difference value. This indicates the fluctuation range of the pressure distribution state factor.

[0120] The formula for calculating the degree of centroid deviation is as follows: In the formula, This indicates the degree of centroid deviation at time t. This represents the axial position of the centroid at time t. Indicates the axial position reference of the center of mass. Indicates the second difference value. This indicates the physical tolerance for axial displacement.

[0121] The second step involves conducting a fusion analysis based on the degree of pressure dispersion deviation and the degree of centroid deviation, combined with the degree of matching, to determine the comprehensive risk assessment coefficient for each moment.

[0122] However, various non-slip factors in clinical practice (such as patient turning over, coughing, or local compression) can also cause temporary abnormalities in the aforementioned single dimension, leading to misdiagnosis. To address this, the present invention further introduces a third dimension, reflecting the degree of matching of the consistency of pressure change direction. This parameter characterizes the spatial correlation between the direction of local pressure gradient and the direction of overall centroid movement, and can effectively distinguish between continuous traction with a clear direction and random interference without direction.

[0123] As an exemplary implementation, determining the comprehensive risk assessment coefficient at each time point includes: The first sub-step involves determining the risk contribution coefficient at each time point and using the product of the risk contribution coefficient and the degree of matching as the degree of directional matching.

[0124] Here, the risk contribution coefficient is determined by comparing the degree of matching with the preset matching threshold.

[0125] In this embodiment, a risk contribution coefficient is set to characterize the consistency of directional traction. When the degree of matching between the gradient vector and the centroid displacement vector at a certain moment is greater than 0, the risk contribution coefficient is set to 1. When the degree of matching between the gradient vector and the centroid displacement vector at a certain moment is not greater than 0, the risk contribution coefficient is set to 0.

[0126] It should be noted that the risk contribution coefficient is similar to a one-way filter. Slippage inevitably leads to the pressure gradient being consistent with the direction of movement, in which case the matching degree is greater than 0. Non-slippage interference (such as body position adjustment) usually results in directions that are unrelated or opposite, in which case the matching degree is not greater than 0. This interference factor does not need to be considered when calculating the comprehensive risk assessment coefficient. Without restrictions, a negative matching degree (opposite direction) will mathematically lower the overall risk coefficient, leading to missed alarms. Therefore, this embodiment determines the risk contribution coefficient to allow only cases with consistent directions (high risk) to participate in the risk assessment, while filtering out interference with inconsistent directions (no risk), thereby improving alarm accuracy.

[0127] The second sub-step involves determining the comprehensive risk assessment coefficient for each moment based on the degree of directional matching, the degree of pressure dispersion deviation, and the degree of centroid deviation.

[0128] Here, the larger the comprehensive risk assessment coefficient at a certain moment, the greater the probability of slippage risk at that moment.

[0129] In this embodiment, the cumulative values ​​of the directional matching degree, pressure dispersion deviation degree, and centroid deviation degree at the same moment are calculated, and the cumulative values ​​of the three degrees are normalized. The normalized value is then used as the comprehensive risk assessment coefficient for the corresponding moment. The normalization process can employ the maximum-minimum normalization method, which is an existing technology and will not be elaborated upon here.

[0130] In another embodiment, the degree of directional matching, the degree of pressure dispersion deviation, and the degree of centroid deviation at the same time are weighted and summed, and the weighted sum is normalized. The normalized value is used as the comprehensive risk assessment coefficient at the corresponding time. The sum of the weight coefficients for different dimensions is 1. For example, the degree of directional matching, the degree of pressure dispersion deviation, and the degree of centroid deviation are 0.2, 0.4, and 0.4, respectively. These values ​​can be calibrated based on clinical trial data to reflect the differences in the contribution of different characteristics to slippage risk.

[0131] The third step is to obtain the output results of the signal analysis and transmission module 9 based on the comprehensive risk assessment coefficient at each moment.

[0132] Because patients may experience intense but transient disturbances, such as rapid turning over and occasional limb tremors, leading to non-slippage abnormalities, it is necessary to further integrate local directional features and duration judgments when obtaining the final output results.

[0133] Specifically, if the comprehensive risk assessment coefficient is greater than the assessment threshold for a consecutive preset number of time periods, it is determined to be fixed abnormal; otherwise, it is determined to be fixed normal. The determination result is used as the output result of the signal analysis and transmission module 9.

[0134] In this embodiment, the comprehensive risk assessment coefficient for a preset number of consecutive time intervals is the short duration period, which can be set to 3 seconds. The system monitors two conditions through a real-time timing module. When the comprehensive risk assessment coefficient is greater than the assessment threshold of 0.6 and the duration exceeds the short duration period, it is judged as abnormal. If an abnormality occurs, the system immediately sends a signal to the bedside remote monitoring system based on the signal transmission module. The bedside remote monitoring system then alarms the nurse station and sends a locking signal to the airbag clamping mechanism 1. The gas in the inflation module 11 is injected into the airbag clamping mechanism 1 to instantly increase the pressure and improve the friction between the clamping mechanism and the drainage tube to resist slippage.

[0135] The preset number of moments, or the number of moments within 3 seconds, is a short duration designed to effectively distinguish between persistent slip-through risks and transient artifacts (such as coughing or postural changes). Its length should be greater than the duration of a typical disruptive event. This value can be calibrated based on clinical observation data. The assessment threshold can be set to 0.6. This threshold is not a fixed value but is calibrated through receiver operating characteristic (ROC) curve analysis of a large amount of clinical data to achieve the optimal balance between warning sensitivity and specificity. Furthermore, this value can be preset and adjusted by medical institutions according to their risk control strategies.

[0136] Step S5 above calculates the comprehensive risk assessment coefficient based on multi-source information fusion, integrates the analysis results of all dimensions, and generates a unified and robust comprehensive risk assessment index through weighted or logical fusion. It ultimately achieves a balance between high sensitivity and high reliability, solving the dilemma of traditional systems. The multi-evidence fusion decision mechanism greatly enhances the system's anti-interference ability and judgment confidence in complex clinical environments, making the final warning output both timely and accurate, providing a reliable basis for initiating active braking (such as airbag inflator), thus solving the ultimate problems of limited braking capacity and poor warning reliability in a closed loop.

[0137] In summary, one embodiment of the present invention provides a fixation system for an anti-slip chest drainage tube in cardiothoracic surgery. This system synchronously acquires the dynamic pressure distribution at the fixation interface using a multi-sensor array, extracts the centroid and dispersion of the pressure distribution as dual global features, and simultaneously senses the overall positional movement and contact uniformity disruption of the drainage tube. It introduces consistency analysis between the local pressure gradient and the direction of centroid movement, distinguishing between directional traction and non-directional interference from a physical mechanism perspective, reducing misjudgments caused by patient movements such as turning over. Simultaneously, the system establishes an individual adaptive physical baseline and uses a safety tolerance directly derived from mechanical design parameters as a judgment threshold to ensure accuracy. By fusing multi-dimensional features for comprehensive risk assessment, and triggering immediate alarms and mechanical locking upon confirmation of high risk, the system improves the timeliness and reliability of early warnings.

[0138] The above-described embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention, and should all be included within the protection scope of the present invention.

Claims

1. A system for fixing a chest drainage tube to prevent slippage in cardiothoracic surgery, characterized in that, Includes a memory and a processor, the processor being used to process instructions stored in the memory to implement the following detection process: Real-time acquisition of pressure data time series and two-dimensional parameterized coordinates from several flexible thin-film pressure sensors (3); the two-dimensional parameterized coordinates include the axial coordinates and circumferential coordinates of the sleeve; Determine the centroid coordinates of the sleeve; based on the inherent dimensions of the sensor array in the axial and circumferential directions, perform a fusion analysis on the dispersion of the sensor coordinates relative to the centroid coordinates, and combine the pressure data of each sensor to determine the pressure distribution state factor at each moment; Construct a gradient vector representing the direction of local pressure change and a centroid displacement vector representing the direction of motion of the overall pressure center, and calculate the degree of matching between the gradient vector and the centroid displacement vector; The statistical characteristics of the pressure distribution state factor and the axial position of the center of mass are obtained by initializing the learning period, and the mechanical design parameters of the sleeve are integrated to adaptively establish a personalized set of reference parameters including state reference, fluctuation range and physical tolerance. Based on the benchmark parameter set, the pressure dispersion deviation and centroid deviation are calculated in real time. Combined with the matching degree, a fusion analysis is performed to determine the comprehensive risk assessment coefficient at each moment, and then the output result of the signal analysis and transmission module (9) is obtained.

2. The anti-slip chest drainage tube fixation system for cardiothoracic surgery according to claim 1, characterized in that, The steps for determining the centroid coordinates include: At the same moment, the pressure data of each sensor is used as the weight of the corresponding sensor's two-dimensional parameterized coordinates. The weighted average method is used to calculate the weighted average of the axial coordinates and the circumferential coordinates, which are used as the centroid coordinates at the corresponding moment. The centroid coordinates include the centroid axial coordinates and the centroid circumferential coordinates.

3. The anti-slippage chest drainage tube fixation system for cardiothoracic surgery according to claim 1, characterized in that, The determination of the pressure distribution state factor at each moment includes: Obtain the total axial span and total circumferential width of the sensor array; The axial coordinates and centroid axial coordinates of each sensor are normalized using the total axial span, and the circumferential coordinates and centroid circumferential coordinates of each sensor are normalized using the total circumferential width. For any given moment, based on the normalized two-dimensional parameterized coordinates and centroid coordinates of each sensor, calculate the squared Euclidean distance from each sensor to the centroid. The contribution value of each sensor is obtained based on the squared Euclidean distance and pressure data of each sensor; The pressure distribution state factor at the corresponding moment is determined based on the contribution value of each sensor.

4. The anti-slip chest drainage tube fixation system for cardiothoracic surgery according to claim 3, characterized in that, The step of determining the pressure distribution state factor at the corresponding time based on the contribution value of each sensor includes: Calculate the sum of the contribution values ​​of all sensors at the same time, as the first sum; and calculate the sum of the pressure data of all sensors at the same time, as the second sum. The pressure distribution state factor at the corresponding time is determined based on the ratio of the first sum to the second sum.

5. The anti-slippage chest drainage tube fixation system for cardiothoracic surgery according to claim 1, characterized in that, The construction of the gradient vector characterizing the direction of local pressure change and the centroid displacement vector characterizing the direction of motion of the overall pressure center includes: For any given moment, based on the distribution of the sensor array in the two-dimensional parameter space, the first and second coefficients of the two-dimensional pressure plane model are obtained by fitting the two-dimensional parameterized coordinates of each sensor and the pressure data using the least squares method. The first coefficient and the second coefficient are used as the partial derivatives of the pressure field in the axial and circumferential directions, respectively, and the gradient vector is constructed based on the partial derivatives of the pressure field in the axial and circumferential directions. The axial centroid coordinate difference and circumferential centroid coordinate difference at the first and second moments are obtained to form a centroid displacement vector; the first moment is any of the moments mentioned above, and the second moment is a moment that differs from the first moment by a preset time window.

6. The anti-slip chest drainage tube fixation system for cardiothoracic surgery according to claim 1, characterized in that, The calculation of the matching degree between the gradient vector and the centroid displacement vector includes: Calculate the product of the gradient vector and the centroid displacement vector, and use it as the first product; Obtain the magnitudes of the gradient vector and the centroid displacement vector, and determine the second product based on the product of the two magnitudes; The degree of matching is determined based on the first product and the second product.

7. The anti-slip chest drainage tube fixation system for cardiothoracic surgery according to claim 1, characterized in that, The adaptive establishment of a personalized set of benchmark parameters, including state benchmarks, fluctuation ranges, and physical tolerances, includes: During the initialization learning period, the centroid axial position and pressure distribution state factor at each moment are acquired; the initialization learning period is at least used to characterize the time period during which the patient remains at rest and the drainage tube is in a stable and normal clamping state. The average values ​​of the centroid axial position and pressure distribution state factor at all times during the initial learning period are calculated to determine the state reference; the state reference includes the position reference and the pressure state reference. Based on the pressure distribution state factor at all times during the initial learning period, the fluctuation range of the pressure distribution state factor is determined using the standard deviation function; Obtain the mechanical design parameters of the sleeve and the sensor axis vector range to determine the physical tolerance of the axial displacement; the mechanical design parameters include at least the mechanical design length and the safety sliding margin.

8. The anti-slip chest drainage tube fixation system for cardiothoracic surgery according to claim 7, characterized in that, The real-time calculation of pressure dispersion and centroid deviation based on the benchmark parameter set includes: For any given moment, calculate the first difference value between the pressure distribution state factor and the pressure state benchmark, and use the ratio of the first difference value to the fluctuation range as the degree of pressure dispersion deviation. Calculate a second difference value between the axial position of the centroid and the position reference, and use the ratio of the second difference value to the physical tolerance as the degree of centroid deviation.

9. The anti-slip chest drainage tube fixation system for cardiothoracic surgery according to claim 1, characterized in that, The output of the signal analysis and transmission module (9) is obtained, including: For each moment, a risk contribution coefficient is determined, and the product of the risk contribution coefficient and the degree of matching is taken as the degree of directional matching; the risk contribution coefficient is determined by comparing the degree of matching with a preset matching threshold. Based on the degree of directional matching, the degree of pressure dispersion deviation, and the degree of centroid deviation, a comprehensive risk assessment coefficient is determined for each moment. If the comprehensive risk assessment coefficient is greater than the assessment threshold for a consecutive preset number of time periods, it is determined to be fixed abnormal; otherwise, it is determined to be fixed normal. The determination result is used as the output result of the signal analysis and transmission module (9).

10. A device for fixing a chest drainage tube to prevent slippage in cardiothoracic surgery, characterized in that, include: Wearing mechanism (5) for securing the device to the patient's body via end Velcro (8); An airbag clamping mechanism (1) is fixedly installed on the wearable mechanism (5), and a fixing groove (6) is provided on one or both sides of it. A detachable unlocking block (2) is provided with a silicone clamping mechanism (4). The unlocking block (2) is configured to be able to be inserted into the fixing groove (6), so that the silicone clamping mechanism (4) fits with the airbag clamping mechanism (1) to form a sleeve for clamping the drainage tube. A press-type elastic self-locking mechanism (7) is provided between the unlocking block (2) and the fixing groove (6) to lock the two when the unlocking block (2) is inserted into the position, and to release the lock when it is pressed again so that the unlocking block (2) pops out. Multiple flexible thin-film pressure sensors (3) are arrayed on the inner surface of the sleeve formed by the airbag clamping mechanism (1) and the silicone clamping mechanism (4) to monitor the contact pressure between the drainage tube and the sleeve. The signal analysis and transmission module (9) is electrically connected to the plurality of flexible thin-film pressure sensors (3) and is used to implement the execution steps of the anti-slip chest drainage tube fixation system for cardiothoracic surgery as described in any one of claims 1-9; An inflation module (11) is connected to the airbag clamping mechanism (1) via an air tube (10); and a control unit is configured to: when the signal analysis and transmission module (9) analyzes and determines that the contact pressure signal is abnormal, control the inflation module (11) to inflate the airbag clamping mechanism (1) with gas to increase the volume of the airbag clamping mechanism (1) and thereby enhance the clamping force on the drainage tube.

Images