Intelligent monitoring and control system for drainage of ascites in severe pancreatitis

Abstract

This invention relates to the field of medical drainage equipment technology, specifically to an intelligent monitoring and control system for draining ascites in severe pancreatitis. The system acquires pressure values ​​at each sampling moment during both forward injection and reverse aspiration; determines the forward drainage time based on the temporal distribution of the forward pressure values, and the reverse drainage time based on the temporal distribution of the negative pressure values, thereby determining the time difference between forward and reverse drainage; determines the attenuation trajectory smoothness based on the neighborhood reference offset of the negative pressure value at each sampling moment; determines a unidirectional blockage characteristic value based on the time difference between forward and reverse drainage and the attenuation trajectory smoothness; and controls the drainage tube blockage based on the unidirectional blockage characteristic value. This avoids the risk of intracavitary accumulation caused by repeated injection of flushing fluid in case of misjudgment in existing technologies, improving the system's anti-blockage safety and control accuracy.

Description

Technical Field

[0001] This invention relates to the field of medical drainage equipment technology, specifically to an intelligent monitoring and control system for drainage of ascites in severe pancreatitis. Background Technology

[0002] During the drainage of ascites in severe pancreatitis, continuous aspiration of the fluid and flushing of the drainage tubing are usually required. However, the fluid in the drainage tubing is typically a multiphase mixture of high viscosity and semi-solid necrotic debris. This multiphase mixture flows slowly within the catheter, easily causing blockage of the side holes of the drainage catheter.

[0003] Existing technologies typically rely on a single upper limit threshold for positive injection pressure for pipeline flushing and shutdown control. Specifically, when the injection pressure reaches the threshold, it is considered a blockage and the system shuts down; otherwise, injection or drainage continues. However, in practical applications, the external flexible coverings (such as free soft tissue) inside the chamber deform and adhere tightly to the side holes of the conduit under negative pressure suction, blocking the outflow of fluid. Conversely, under positive pressure injection of flushing fluid, these flexible coverings can easily push open the gaps, allowing flushing fluid to drain into the chamber without triggering the set upper limit alarm for injection pressure. This leads to the equipment determining that the pipeline is not clear when there is no pressure alarm during positive injection and no liquid is discharged during reverse suction. Consequently, it executes the operating logic of repeatedly injecting flushing fluid into the chamber, causing dangerous accumulation of flushing fluid and increasing local pressure. Summary of the Invention

[0004] To address the technical problem that existing technologies typically rely on a single upper limit threshold for positive injection pressure for pipeline flushing and shutdown control, failing to distinguish between unidirectional flow obstruction caused by flexible covering adhesion and mechanical disturbances generated by multiphase drainage material passages, leading to misjudgments and improper flushing operations, this invention aims to provide an intelligent monitoring and control system for ascites drainage in severe pancreatitis. The specific technical solution adopted is as follows: This invention proposes an intelligent monitoring and control system for drainage of ascites in severe pancreatitis, the system comprising: The data acquisition module is used to acquire the positive pressure value at each sampling moment during the positive injection process and the negative pressure value at each sampling moment during the negative aspiration process in the drainage of ascites. The feature analysis module is used to determine the positive drainage time based on the temporal distribution of the positive pressure values ​​at each sampling time; to determine the negative drainage time based on the temporal distribution of the negative pressure values ​​at each sampling time; and to determine the difference between the positive and negative drainage time based on the positive and negative drainage time. The blockage analysis module is used to determine the attenuation trajectory smoothness based on the neighborhood reference offset of the negative pressure value at each sampling time; and to determine the unidirectional blockage characteristic value based on the difference in drainage time between the forward and reverse directions and the attenuation trajectory smoothness. The status control module is used to control the blockage and unblocking of the drainage pipeline based on the unidirectional blockage characteristic value.

[0005] Furthermore, the method for obtaining the time consumed in the forward drainage includes: The sampling times corresponding to all positive pressure values ​​within the preset pressure drop range are taken as the target pressure drop times; the positive drainage time is determined based on the number of target pressure drop times.

[0006] Furthermore, the method for obtaining the time difference between forward and reverse drainage includes: The time difference between forward and reverse drainage is determined based on the deviation between the forward drainage time and the negative drainage time.

[0007] Furthermore, the method for obtaining the smoothness of the attenuation trajectory includes: Any two adjacent sampling times during the negative suction process are considered as a matching pair; The number of effective smoothing points is determined based on the difference between the negative pressure value and the neighborhood trajectory baseline of each matching pair. The smoothness of the decay trajectory is determined based on the proportion of the effective smoothing points in the total number of all matching pairs.

[0008] Furthermore, the method for obtaining the effective number of smoothing points includes: The state vector for each sampling moment is determined based on the negative pressure value within the preset time neighborhood at each sampling moment; Based on the degree of difference between the state vectors at two sampling times within each matching pair, the feature distance of each matching pair is determined; the number of matching pairs whose feature distance is less than or equal to the preset feature distance threshold is counted to determine the number of effective smoothing points.

[0009] Furthermore, the method for obtaining the unidirectional blocking feature value includes: The deviation between the attenuation trajectory smoothness and the preset smoothness threshold is nonlinearly mapped to obtain a nonlinear gate value; A unidirectional blockage characteristic value is determined based on the nonlinear gating value and the time difference between forward and reverse drainage; both the nonlinear gating value and the time difference between forward and reverse drainage are positively correlated with the unidirectional blockage characteristic value.

[0010] Furthermore, the sigmoid function is used to perform a nonlinear mapping between the deviation between the smoothness of the decay trajectory and a preset smoothness threshold.

[0011] Furthermore, the negative pressure value is the absolute value of the instantaneous pressure value obtained at each sampling moment during the negative suction process.

[0012] Furthermore, the state vector is a one-dimensional array composed of the negative pressure values ​​of all neighboring sampling times within a preset time neighborhood for each sampling time.

[0013] Furthermore, Euclidean distance is used to calculate the degree of difference between the state vectors at two sampling times within each matching pair.

[0014] The present invention has the following beneficial effects: This invention overcomes the limitation of existing technologies that rely solely on positive injection pressure for judgment by using both positive and negative pressure values, providing a bidirectionally comparable data foundation for distinguishing different blockage types. By analyzing the temporal distribution of positive and negative pressure values ​​separately, the time consumption for positive and negative drainage is determined, along with the difference in drainage time between the two directions. This quantifies the degree of resistance asymmetry in the pipeline during the forward and reverse fluid transport processes, providing a macroscopic basis for identifying bidirectional flow obstruction caused by flexible covering adhesion. Based on the neighborhood reference offset of the negative pressure value at each sampling time, the smoothness of the attenuation trajectory is determined, transforming a simple hydrodynamic obstruction problem into a comprehensive analysis. By transforming the signal feature filtering problem into a data dimension, the system effectively identifies and isolates disturbances in the multiphase drainage material through-holes, thereby quantifying the local morphological self-similarity of the pressure decay trajectory at the microscopic level. By determining unidirectional blockage characteristic values ​​based on the time difference between forward and reverse drainage and the smoothness of the decay trajectory, the system can effectively distinguish between unidirectional blockage caused by flexible coverings and mechanical disturbances generated during multiphase mixed drainage material through-holes, achieving a fusion of time-consuming and smoothness characteristics. Finally, based on the unidirectional blockage characteristic values, the system regulates and unblocks the drainage pipeline, avoiding the risk of cavity accumulation caused by repeated injection of flushing fluid in case of misjudgment, thus improving the system's anti-blockage safety and control accuracy. Attached Figure Description

[0015] Figure 1 This is a structural diagram of an intelligent monitoring and control system for drainage of ascites in severe pancreatitis, provided in one embodiment of the present invention. Figure 2 This is a flowchart of a method for obtaining the smoothness of an attenuation trajectory, provided in one embodiment of the present invention. Detailed Implementation

[0016] The following description, in conjunction with the accompanying drawings, details a specific scheme for an intelligent monitoring and control system for draining ascites in severe pancreatitis provided by the present invention.

[0017] Please see Figure 1The diagram illustrates a structural diagram of an intelligent monitoring and control system for draining ascites in severe pancreatitis, according to an embodiment of the present invention. The system includes: The data acquisition module 101 is used to acquire the positive pressure value at each sampling moment during the positive injection process and the negative pressure value at each sampling moment during the negative aspiration process in the drainage of peritoneal effusion.

[0018] During the drainage of ascites in severe pancreatitis, the fluid flowing through the drainage tube is not a single, homogeneous liquid, but a multiphase mixture of high-viscosity grease and semi-solid necrotic debris. Simultaneously, free soft tissue exists within the abdominal cavity, which is prone to deformation and adhesion to the catheter side holes under negative pressure. These two factors—the former being mechanical disturbance within the fluid, and the latter being physical coverage by external tissue—can both potentially lead to drainage obstruction. However, current technology only monitors pressure changes during forward injection, identifying blockage and stopping the procedure when the pressure reaches a preset threshold. However, in scenarios where flexible coverings adhere, the covering is dislodged during forward injection, preventing the pressure from reaching the threshold; conversely, during negative pressure aspiration, the covering is sucked tightly, resulting in no fluid drainage. This unidirectional coverage phenomenon—open in the forward direction but blocked in the reverse—cannot be identified using forward pressure data alone. Therefore, in the embodiments of the present invention, the positive pressure value at each sampling moment during the positive injection process and the negative pressure value at each sampling moment during the negative aspiration process are obtained in the drainage of peritoneal effusion. By obtaining complete pressure information under bidirectional flow conditions, the unique flow asymmetry characteristics of flexible covering adhesion are captured.

[0019] In one specific implementation of this invention, the system inserts a drainage catheter into the abdominal cavity of the target individual. A controllable pump, a negative pressure source, a controllable connecting valve, and a pressure sensor are all deployed outside the target individual, forming a closed fluid pathway through a drainage tube connected to the internal drainage catheter. When the data acquisition module 101 performs historical data recording, it first controls the controllable pump to inject a preset volume of physiological saline into the tube to pressurize it. After the injection stops, the pressure begins to drop. The pressure sensor continuously records the instantaneous pressure value at each sampling moment during the natural pressure decay process within the closed tube at a preset sampling period (0.1 seconds in this embodiment), storing this as the positive pressure value at each sampling moment in the memory. Subsequently, the negative pressure source is controlled to aspirate and depressurize the tube. After aspiration stops, the pressure begins to rise again. The pressure sensor continuously records the instantaneous pressure value during the natural rise of the negative pressure within the tube at the same sampling period, storing the absolute value of this instantaneous pressure value as the negative pressure value at each sampling moment in the memory. The positive and negative pressure values ​​are both offline data recorded after the diagnosis was completed, which are then called and processed by the subsequent feature analysis module 102.

[0020] The feature analysis module 102 is used to determine the positive drainage time based on the temporal distribution of the positive pressure values ​​at each sampling time; to determine the negative drainage time based on the temporal distribution of the negative pressure values ​​at each sampling time; and to determine the difference between the positive and negative drainage time based on the positive drainage time and the negative drainage time.

[0021] Within the same pressure drop observation range, the time required for fluid discharge directly reflects the resistance of the pipeline to fluid flow. Therefore, this embodiment of the invention determines the positive drainage time based on the temporal distribution of the positive pressure values ​​at each sampling time, and the negative drainage time based on the temporal distribution of the negative pressure values ​​at each sampling time. The positive drainage time represents the smoothness of the flushing fluid flowing from the conduit into the chamber; the negative drainage time represents the smoothness of the accumulated fluid being drawn from the chamber into the conduit. Under normal circumstances, the two should be roughly equivalent. However, when the flexible cover is attached to the side hole of the conduit, negative pressure suction will cause the cover to be tightly adhered, resulting in a significant increase in the negative drainage time; while positive injection will cause the cover to be flushed open, and the positive drainage time is relatively normal. This asymmetry in bidirectional time consumption reflects the impact of unidirectional coverage by the flexible cover. Therefore, this embodiment of the invention determines the difference between the positive and negative drainage time based on the positive and negative drainage times, thereby providing a basis for identifying unidirectional coverage by the flexible cover.

[0022] The blockage analysis module 103 is used to determine the attenuation trajectory smoothness based on the neighborhood reference offset of the negative pressure value at each sampling time; and to determine the unidirectional blockage characteristic value based on the difference in drainage time between the forward and reverse directions and the attenuation trajectory smoothness.

[0023] While the difference in drainage time between the forward and reverse directions can reflect the asymmetry of bidirectional flow caused by unidirectional covering, the passage of multiphase mixed drainage material through the orifice also significantly prolongs drainage time. When high-viscosity grease or semi-solid necrotic debris passes through the side orifice of the conduit, alternating blockage and unblocking occur, causing intermittent fluid flow, which macroscopically manifests as prolonged drainage time. In this case, the difference in drainage time between the forward and reverse directions alone cannot distinguish between unidirectional covering caused by flexible covering and prolonged drainage time caused by multiphase mixed drainage material. To overcome this dilemma, this invention analyzes the microscopic morphology of the pressure decay trajectory. In the absence of fluid exchange (such as when soft tissue completely covers the side orifice), the pressure recovery inside the tube is driven only by the elastic reset of the conduit wall, and its pressure decay trajectory exhibits high smoothness and self-similarity. However, when multiphase drainage material passes through the orifice, debris passing through the side orifice generates mechanical disturbance, resulting in discontinuous features such as abrupt jumps and jitters in the pressure decay trajectory. Therefore, this embodiment of the invention determines the smoothness of the decay trajectory based on the neighborhood reference offset of the negative pressure value at each sampling time. When the local morphology is highly similar across different times, i.e., when the smoothness of the decay trajectory is large, it indicates that the pressure decay trajectory is smooth and undisturbed; when there are significant differences between the local morphologies, i.e., when the smoothness of the decay trajectory is small, it indicates the presence of hydromechanical disturbance.

[0024] After obtaining the attenuation trajectory smoothness, this embodiment of the invention determines a unidirectional blockage characteristic value based on the difference in drainage time between the forward and reverse directions and the attenuation trajectory smoothness. This ensures that when the attenuation trajectory smoothness is low, indicating the presence of fluid disturbance, it suggests that the macroscopic time extension is caused by multiphase slow flow, and in this case, the difference in drainage time between the forward and reverse directions is constrained; when the attenuation trajectory smoothness is high, indicating the absence of fluid disturbance, it suggests that the macroscopic time extension is caused by soft tissue adhesion, and in this case, the original value of the difference in drainage time between the forward and reverse directions is retained. Therefore, the unidirectional blockage characteristic value can accurately reflect the degree of pipeline blockage after eliminating fluid interference.

[0025] The status control module 104 is used to control the blockage and unblocking of the drainage pipeline based on the unidirectional blockage characteristic value.

[0026] Considering that the time taken for positive drainage directly reflects whether there is extreme physical blockage in the pipeline, when the time exceeds the limit, it indicates that the fluid cannot flow at all. At this time, the unidirectional blockage characteristic value may not be able to be calculated normally due to data acquisition timeout in extreme physical blockage scenarios, or the calculation result may be unreliable. However, the unidirectional blockage characteristic value eliminates the fluid disturbance interference caused by the multiphase drainage material through-hole, and can accurately identify the real unidirectional blockage caused by soft tissue adhesion. Therefore, the embodiments of the present invention use the unidirectional blockage characteristic value to regulate and unblock the drainage pipeline.

[0027] In one specific implementation of this invention, the system inputs the unidirectional blockage characteristic value output by the blockage analysis module 103 into the peritoneal effusion drainage system as a reference for judging the degree of blockage in the pipeline, thereby improving the system's anti-blockage safety and control accuracy.

[0028] In summary, this invention overcomes the limitation of existing technologies that rely solely on positive injection pressure for judgment by using both positive and negative pressure values, providing a bidirectionally comparable data foundation for distinguishing different blockage types. By separately analyzing the temporal distribution of positive and negative pressure values, the invention determines the time consumption for positive and negative drainage, and the difference between the two drainage times, thereby quantifying the degree of resistance asymmetry in the pipeline during the forward and reverse fluid transport processes. This provides a macroscopic basis for identifying bidirectional flow obstruction caused by flexible covering adhesion. Based on the neighborhood reference offset of the negative pressure value at each sampling time, the invention determines the smoothness of the attenuation trajectory, transforming simple hydrodynamic stagnation... The problem was transformed into a signal feature filtering problem at the data dimension, which enabled the effective identification and removal of disturbances in the multiphase drainage material through-holes, thereby quantifying the local morphological self-similarity of the pressure decay trajectory at the microscopic level. By determining the unidirectional blockage characteristic value based on the time difference between forward and reverse drainage and the smoothness of the decay trajectory, the unidirectional blockage characteristic value can be effectively distinguished from the mechanical disturbances generated when the flexible covering material passes through the multiphase mixed drainage material through-holes, achieving the fusion of time-consuming characteristics and smoothness characteristics. Finally, the drainage pipeline blockage control is performed based on the unidirectional blockage characteristic value, avoiding the risk of chamber accumulation caused by repeated injection of flushing fluid in case of misjudgment in the existing technology, and improving the anti-blockage safety and control accuracy of the system.

[0029] Preferably, in some possible implementations of the embodiments of the present invention, the method for obtaining the time consumed by forward drainage includes: The sampling times corresponding to all positive pressure values ​​within the preset pressure drop range are taken as the target pressure drop times; the positive drainage time is determined based on the number of target pressure drop times.

[0030] Considering that not all pressure decay data during the forward injection process have equal analytical value—the initial pressure may be too high, significantly affected by initial disturbances, while the final pressure may be too low, approaching ambient noise levels—a unified observation interval needs to be established to ensure comparability between different diagnostic cycles and between different individuals. Therefore, this embodiment of the invention uses the sampling times corresponding to all forward pressure values ​​within a preset pressure drop range as the target pressure drop times. Since this embodiment uses a fixed sampling interval, the forward drainage time is determined based on the number of target pressure drop times.

[0031] In one specific embodiment of the present invention, the preset pressure drop range is set to [8kPa, 12kPa], which can be adjusted according to the specific implementation scenario. Specifically, since the natural pressure decay during the forward injection process is a gradual decrease from the target injection pressure limit, the system continuously monitors the forward pressure value at each sampling moment during the forward injection process. When the forward pressure value drops to the upper limit of the preset pressure drop range (i.e., 12kPa), the sampling moment is marked as the start time of forward drainage. The upper limit of the preset pressure drop range is strictly less than the target injection pressure limit, which is a fixed value preset and stored in the microprocessor according to clinical safety requirements. Subsequently, continuous monitoring and timing are performed. When the forward pressure value drops to the lower limit of the preset pressure drop range (i.e., 8kPa), the sampling moment is marked as the end time of forward drainage. The system uses the start time of forward drainage, the end time of forward drainage, and all sampling moments in between as the target pressure drop moment. Subtract 1 from the total number of target pressure drop moments to obtain the total number of target pressure drop time intervals; multiply the total number of target pressure drop time intervals by the sampling interval to obtain the forward drainage time.

[0032] It should be noted that, to avoid endless waiting when acquiring the target pressure drop moment due to severe mechanical lock-up in the pipeline, this embodiment of the invention sets a timeout limit, preferably within the range of [80s, 100s]. This embodiment sets it to 90s, which can be adjusted according to the specific implementation scenario. If the forward pressure value has dropped to the lower limit of the preset pressure drop range before reaching the timeout limit, the system immediately stops recording. If the timing duration reaches the timeout limit and the forward pressure value has not yet dropped to the lower limit of the preset pressure drop range, the system directly uses the current sampling moment as the forward drainage termination moment and truncates it.

[0033] Considering that the target limit for negative pressure suction is usually lower than the target injection pressure limit, and that the negative pressure recovery process is affected by pipeline tightness and the viscosity of the accumulated liquid, its effective attenuation range differs from that of positive injection. Therefore, in this embodiment, the negative pressure drop range is set to [6kPa, 10kPa]. Based on the same principle as the above-mentioned positive drainage time, simply replace the positive pressure value with the negative pressure value and the preset pressure drop range with the negative pressure drop range to obtain the negative drainage time.

[0034] Preferably, in some possible implementations of the embodiments of the present invention, the method for obtaining the time difference between forward and reverse drainage includes: Considering that when the flexible cover is attached to the catheter side hole, negative pressure suction will cause the cover to adhere tightly, resulting in a significant increase in the time required for negative drainage; while positive injection will cause the cover to be flushed away, and the time required for positive drainage will be relatively normal. In order to analyze the asymmetry of the time consumption in the two directions, this embodiment of the invention determines the difference in drainage time between the positive and negative directions based on the deviation between the drainage time consumption in the positive and negative directions. The larger the difference in drainage time consumption in the positive and negative directions, the higher the degree of adhesion of the flexible cover to the catheter side hole, and the greater the probability of blockage.

[0035] In one specific implementation of this invention, the system uses the negative drainage time minus the positive drainage time to obtain the difference between the positive and negative drainage time.

[0036] Preferably, in some possible implementations of the embodiments of the present invention, the specific process of obtaining the attenuation trajectory smoothness in the blocking analysis module 103 includes: Please refer to Figure 2 The diagram illustrates a flowchart of a method for obtaining the smoothness of an attenuation trajectory according to an embodiment of the present invention. The method includes: S201: Take any two adjacent sampling times during the negative suction process as a matching pair.

[0037] Considering that the pressure decay trajectory is a time series composed of discrete sampling points, evaluating the smoothness of the entire trajectory cannot be done by simply observing the relationship between a single point or adjacent points. This is because local disturbances may only affect a few points, while global smoothness needs to be reflected through the self-similarity between a large number of point pairs. Therefore, in this embodiment of the invention, any two adjacent sampling times during the negative suction process are treated as a matching pair.

[0038] S202: Determine the number of effective smoothing points based on the difference in the neighborhood trajectory benchmark of each matching pair according to the negative pressure value; Considering that true smooth reset in a pressure decay trajectory should be a recursive similarity between multiple consecutive local trajectories, rather than isolated, discrete matching pairs, for example, if multiple consecutive matching pairs have similar trajectories, it indicates that the pressure decay trajectory has maintained a consistent smooth shape over a long period of time. Therefore, this embodiment of the invention determines the number of effective smoothing points based on the difference in the neighborhood trajectory benchmark of the negative pressure value for each matching pair. The number of effective smoothing points measures the length of the pressure decay trajectory where continuous smoothing occurs.

[0039] Specifically, considering that the pressure value at a single sampling moment is just an isolated value and cannot reflect the trend or trajectory of pressure changes, in order to capture the local shape of pressure changes, this embodiment of the invention determines the state vector of each sampling moment based on the negative pressure value in the preset time neighborhood of each sampling moment; wherein the state vector represents the geometric shape of the pressure change trajectory within the neighborhood time window of the sampling moment.

[0040] In one specific implementation of this invention, the system constructs a preset time neighborhood by taking each sampling moment as the center and extracting one sampling moment forward and one sampling moment backward. The negative pressure values ​​of all sampling moments within the preset time neighborhood are arranged in chronological order to construct the state vector for each sampling moment. For example, the first... The state vector at each sampling time It can be represented as ,in Indicates the first The negative pressure value at each sampling time.

[0041] It should be noted that, since the preset time neighborhoods of the first and last sampling times are incomplete, this invention does not analyze them.

[0042] Considering that if the pressure decay trajectory is smooth, the local trajectory shapes at different sampling times should be highly similar; if the pressure decay trajectory is perturbed, the local trajectory shapes at different sampling times will show significant differences. To quantify the differences between different pairs, this embodiment of the invention determines the feature distance of each pair based on the degree of difference between the state vectors at two sampling times within each pair, thereby quantifying the difference in pressure change trajectories at two sampling times; by counting the number of all pairs with feature distances less than or equal to a preset feature distance threshold, the number of effective smoothing points is determined; by introducing the preset feature distance threshold as a reference benchmark, pairs with large differences in pressure change trajectories are filtered out, thus retaining only pairs with similar neighborhood trajectory shapes for subsequent smoothness evaluation.

[0043] In one specific implementation of this invention, the system uses the Euclidean distance between the state vectors at two sampling times within each matching pair as the feature distance of each matching pair. The preferred range for the preset feature distance threshold is [0.3 kPa, 1 kPa], and this embodiment sets it to 0.7 kPa, which can be adjusted within this preferred range according to the specific implementation scenario. Specifically, for scenarios with low drainage material viscosity, relatively uniform pipe wall elasticity, and high sensitivity to disturbances, a lower preset feature distance threshold can be set; for scenarios with high drainage material viscosity and significant background noise within the pipeline, a higher preset feature distance threshold can be set.

[0044] It should be noted that the calculation of Euclidean distance is a technique well known to those skilled in the art, and its implementation process will not be elaborated here.

[0045] S203: Determine the attenuation trajectory smoothness based on the proportion of the effective smoothing points in the total number of all matching pairs.

[0046] Considering that not all regions in the pressure decay trajectory have high smoothness, and the number of effective smooth points represents the number of matching pairs with high smoothness, this embodiment of the invention determines the smoothness of the decay trajectory based on the proportion of the number of effective smooth points in the total number of matching pairs.

[0047] In one specific implementation of this invention, the number of effective smoothing points is used as the numerator, the total number of all matching pairs is used as the denominator, and the ratio of the two is used as the smoothness of the decay trajectory. Since the negative suction process is a continuous stage containing multiple sampling moments, when the total number of all matching pairs is 0, it indicates that the data is erroneous and needs to be re-acquired.

[0048] Preferably, in some possible implementations of the embodiments of the present invention, the method for obtaining the unidirectional blocking feature value includes: Because catheter materials and individual tissue environments differ, the values ​​of attenuation trajectory smoothness may vary. Therefore, to ensure a unified interpretation of attenuation trajectory smoothness, this embodiment of the invention performs a nonlinear mapping between the attenuation trajectory smoothness and a preset smoothness threshold to obtain a nonlinear gate value. When the deviation is negative, it indicates the presence of fluid mechanical disturbance; when the deviation is positive, it indicates the absence of fluid disturbance; and when the deviation is zero, it is in a critical state. By performing a nonlinear mapping on the deviation, the range of the nonlinear gate value is made (0,1), thereby limiting the time difference between forward and reverse drainage.

[0049] In one specific implementation of this invention, the smoothness deviation is obtained by subtracting a preset smoothness threshold from the smoothness of the decay trajectory. This invention uses a sigmoid function for nonlinear mapping; the nonlinear gating value is obtained by inputting the smoothness deviation into the sigmoid function. The preferred range of the preset smoothness threshold is [0.7, 0.9], and this invention sets it to 0.8, which can be adjusted according to the specific implementation scenario. Specifically: the smaller the preset smoothness threshold, the easier it is for the smoothness of the decay trajectory to exceed the preset smoothness threshold, thereby preserving the time difference between forward and reverse drainage and improving the sensitivity of unidirectional coverage recognition; the larger the preset smoothness threshold, the easier it is for the smoothness of the decay trajectory to fall below the preset smoothness threshold, thereby suppressing the time difference between forward and reverse drainage and reducing the false positive rate.

[0050] As an example, the formula for nonlinear mapping using the sigmoid function can be expressed as: in, Indicates the nonlinear gate value; Indicates smoothness deviation; To amplify the gain coefficient, this embodiment of the invention sets it to 50. The purpose is to amplify the original input range of smoothness deviation [-0.8, 0.2] by 50 times and map it to the interval [-40, 10], so that the range of the nonlinear gate value is (0, 1).

[0051] It should be noted that the purpose of using the sigmoid function for nonlinear mapping is to ensure that when the smoothness of the decay trajectory is slightly higher than the preset smoothness threshold, the nonlinear gate value jumps directly to close to 1, fully preserving the time difference to trigger unblocking; when the smoothness of the decay trajectory is slightly lower than the preset smoothness threshold, the nonlinear gate value jumps directly to close to 0, completely suppressing the time difference to avoid misjudgment. The sigmoid function is a well-known technique in the art, and its implementation process will not be described in detail here.

[0052] When the attenuation trajectory smoothness is high, the nonlinear gate value approaches 1, indicating that the time difference between forward and reverse drainage can truly reflect the unidirectional blockage caused by soft tissue coverage. In this case, a larger time difference between forward and reverse drainage should be retained. When the attenuation trajectory smoothness is low, the nonlinear gate value approaches 0, indicating that the time difference between forward and reverse drainage is a pseudo-signal caused by fluid disturbance. In this case, the time difference between forward and reverse drainage should be suppressed. Based on this design principle, this embodiment of the invention determines the unidirectional blockage characteristic value according to the nonlinear gate value and the time difference between forward and reverse drainage; both the nonlinear gate value and the time difference between forward and reverse drainage are positively correlated with the unidirectional blockage characteristic value.

[0053] In one specific implementation of this invention, the product of the nonlinear gate value and the time difference between forward and reverse drainage is used as the unidirectional blockage characteristic value.

Claims

1. An intelligent monitoring and regulating system for severe pancreatitis ascites drainage, characterized in that, The system includes: The data acquisition module is used to acquire the positive pressure value at each sampling moment during the positive injection process and the negative pressure value at each sampling moment during the negative aspiration process in the drainage of ascites. The feature analysis module is used to determine the positive drainage time based on the temporal distribution of the positive pressure values ​​at each sampling time; to determine the negative drainage time based on the temporal distribution of the negative pressure values ​​at each sampling time; and to determine the difference between the positive and negative drainage time based on the positive and negative drainage time. The blockage analysis module is used to determine the attenuation trajectory smoothness based on the neighborhood reference offset of the negative pressure value at each sampling time; and to determine the unidirectional blockage characteristic value based on the difference in drainage time between the forward and reverse directions and the attenuation trajectory smoothness. The status control module is used to control the blockage and unblocking of the drainage pipeline based on the unidirectional blockage characteristic value.

2. The intelligent monitoring and regulating system for severe pancreatitis ascites drainage according to claim 1, characterized in that, The method for obtaining the time consumed in the positive drainage includes: The sampling times corresponding to all positive pressure values ​​within the preset pressure drop range are taken as the target pressure drop times; the positive drainage time is determined based on the number of target pressure drop times.

3. The intelligent monitoring and regulating system for severe pancreatitis ascites drainage according to claim 1, characterized in that, The method for obtaining the time difference between forward and reverse drainage includes: The time difference between forward and reverse drainage is determined based on the deviation between the forward drainage time and the negative drainage time.

4. The intelligent monitoring and regulating system for severe pancreatitis ascites drainage according to claim 1, characterized in that, The method for obtaining the smoothness of the attenuation trajectory includes: Any two adjacent sampling times during the negative suction process are considered as a matching pair; The number of effective smoothing points is determined based on the difference between the negative pressure value and the neighborhood trajectory baseline of each matching pair. The smoothness of the decay trajectory is determined based on the proportion of the effective smoothing points in the total number of all matching pairs.

5. The intelligent monitoring and regulating system for severe pancreatitis ascites drainage according to claim 4, characterized in that, The method for obtaining the effective number of smoothing points includes: The state vector for each sampling moment is determined based on the negative pressure value within the preset time neighborhood at each sampling moment; Based on the degree of difference between the state vectors at two sampling times within each matching pair, the feature distance of each matching pair is determined; the number of matching pairs whose feature distance is less than or equal to the preset feature distance threshold is counted to determine the number of effective smoothing points.

6. The intelligent monitoring and regulating system for severe pancreatitis ascites drainage according to claim 1, characterized in that, The method for obtaining the unidirectional blocking feature value includes: The deviation between the attenuation trajectory smoothness and the preset smoothness threshold is nonlinearly mapped to obtain a nonlinear gate value; A unidirectional blockage characteristic value is determined based on the nonlinear gating value and the time difference between forward and reverse drainage; both the nonlinear gating value and the time difference between forward and reverse drainage are positively correlated with the unidirectional blockage characteristic value.

7. The intelligent monitoring and regulating system for severe pancreatitis ascites drainage according to claim 6, characterized in that, The sigmoid function is used to perform a nonlinear mapping between the deviation between the smoothness of the decay trajectory and a preset smoothness threshold.

8. The intelligent monitoring and control system for drainage of ascites in severe pancreatitis according to claim 1, characterized in that, The negative pressure value is the absolute value of the instantaneous pressure value obtained at each sampling moment during the negative suction process.

9. The intelligent monitoring and control system for drainage of ascites in severe pancreatitis according to claim 5, characterized in that, The state vector is a one-dimensional array consisting of the negative pressure values ​​of all neighboring sampling times within a preset time neighborhood for each sampling time.

10. The intelligent monitoring and control system for drainage of ascites in severe pancreatitis according to claim 5, characterized in that, The Euclidean distance is used to calculate the degree of difference between the state vectors at two sampling times within each matching pair.

Images