Bronchial occluder cuff airtightness detection method and device based on dynamic air pressure sensing

Abstract

The invention discloses a dynamic air pressure sensing bronchial occluder cuff air tightness detection method and device, and relates to the technical field of air tightness detection, and the method comprises the steps: carrying out the inflation test of a target bronchial occluder, carrying out the continuous monitoring of air pressure, and obtaining an air pressure monitoring data sequence; determining the overall air tightness of the final assembly to obtain an overall air tightness determination result of the final assembly; if the overall airtightness of the final assembly is abnormal, performing a dynamic inflation and deflation test on the target bronchial occluder to obtain a springback air pressure monitoring data sequence, performing time sequence iterative analysis and prototype identification on the springback air pressure monitoring data sequence, and determining a target airtightness abnormal prototype; when the abnormal type is air leakage of the cuff body, a closed pressure maintaining test is carried out, and a target cuff air tightness monitoring result is obtained. The technical problem that in the prior art, the abnormal type of the air tightness of the bronchial occluder cannot be accurately recognized in the final assembly state is solved, and the technical effect of automatically recognizing the abnormal type under the final assembly condition is achieved.

Description

Technical Field

[0001] The present invention relates to the technical field of air tightness detection, and in particular to a method and device for detecting air tightness of a bronchial occluder cuff using dynamic air pressure sensing. Background Art

[0002] Factory inspection for bronchial occluders typically relies on airtightness testing of individual components or subassemblies. However, after assembly, due to the continuous flow of gas passages and the airtight structure, it is difficult to distinguish whether the airtightness anomaly originates from the catheter connection, the cuff itself, or elsewhere. Traditional testing methods cannot accurately determine the location of the anomaly while maintaining the final assembly state, resulting in low leak diagnosis efficiency, high rework rates, and difficulty meeting device consistency inspection requirements. Summary of the Invention

[0003] The present application provides a method and device for detecting air tightness of a bronchial occluder cuff using dynamic air pressure sensing, which is used to solve the technical problem in the prior art that the type of air tightness abnormality of a bronchial occluder cannot be accurately identified in the assembled state.

[0004] In view of the above problems, the present application provides a method and device for detecting the air tightness of a bronchial occluder cuff using dynamic air pressure sensing.

[0005] In a first aspect of the present application, a method for detecting air tightness of a bronchial occluder cuff using dynamic air pressure sensing is provided, the method comprising: The target bronchial occluder is introduced into the bronchial imitation clamp for inflation test. When the air pressure reaches the target working pressure, the digital micro-pressure sensor is used to continuously monitor the air pressure in the preset first detection window to obtain an air pressure monitoring data sequence; the overall air tightness of the assembly is judged according to the air pressure monitoring data sequence to obtain the overall air tightness judgment result of the assembly; if the overall air tightness judgment result of the assembly is that the overall air tightness of the assembly is abnormal, the target bronchial occluder in the bronchial imitation clamp is dynamically inflated and deflated, and the digital micro-pressure sensor is used to monitor the air pressure in the preset first detection window to obtain an air pressure monitoring data sequence; The sensor continuously monitors the air pressure within a preset second detection window to obtain a rebound air pressure monitoring data sequence; performs time-series iterative analysis on the rebound air pressure monitoring data sequence, and uses the airtightness abnormality prototype library to perform prototype identification on the analysis results to determine the target airtightness abnormality prototype; when the abnormality type of the target airtightness abnormality prototype is air leakage in the cuff body, the tail of the catheter in the target bronchial occluder is sealed, and directionally pressurized is applied to the cavity of the cuff body to perform a closed pressure retention test to obtain the target cuff airtightness monitoring result.

[0006] The second aspect of the present application provides a dynamic air pressure sensing bronchial occluder cuff air tightness detection device, the device comprising: The inflation test module is used to introduce the target bronchial occluder into the bronchial imitation clamp for inflation test. When the air pressure reaches the target working pressure, the digital micro-pressure sensor is used to continuously monitor the air pressure in the preset first detection window to obtain an air pressure monitoring data sequence; the air tightness judgment module is used to judge the overall air tightness of the assembly according to the air pressure monitoring data sequence to obtain the overall air tightness judgment result of the assembly; the inflation and deflation test module is used to perform dynamic inflation and deflation tests on the target bronchial occluder in the bronchial imitation clamp if the overall air tightness judgment result of the assembly is that the overall air tightness of the assembly is abnormal, and use the The digital micro-pressure sensor continuously monitors the air pressure within a preset second detection window to obtain a rebound air pressure monitoring data sequence; the timing iterative analysis module is used to perform timing iterative analysis on the rebound air pressure monitoring data sequence, and use the airtightness abnormality prototype library to perform prototype identification on the analysis results to determine the target airtightness abnormality prototype; the monitoring result acquisition module is used to seal the tail of the catheter in the target bronchial occluder when the abnormal type of the target airtightness abnormality prototype is air leakage of the cuff body, and directionally pressurize the cavity of the cuff body to perform a closed pressure maintenance test to obtain the target cuff airtightness monitoring result.

[0007] One or more technical solutions provided in this application have at least the following technical effects or advantages: In the present application, a target bronchial occluder is introduced into a simulated bronchial fixture for inflation test. When the air pressure reaches the target working pressure, a digital micro-pressure sensor is used to continuously monitor the air pressure in a preset first detection window to obtain an air pressure monitoring data sequence. The overall air tightness of the assembly is determined based on the air pressure monitoring data sequence to obtain an overall air tightness determination result of the assembly. If the overall air tightness determination result of the assembly is that the overall air tightness of the assembly is abnormal, a dynamic inflation and deflation test is performed on the target bronchial occluder in the simulated bronchial fixture, and the digital micro-pressure sensor is used to continuously monitor the air pressure in a preset first detection window to obtain an air pressure monitoring data sequence. The micro-pressure sensor continuously monitors the air pressure within a preset second detection window to obtain a rebound air pressure monitoring data sequence; the rebound air pressure monitoring data sequence is subjected to time-series iterative analysis, and the analysis results are prototype-identified using the airtightness abnormality prototype library to determine the target airtightness abnormality prototype; when the abnormality type of the target airtightness abnormality prototype is a cuff body leak, the tail of the catheter in the target bronchial occluder is sealed, and the chamber of the cuff body is directionally pressurized to perform a closed pressure retention test to obtain the target cuff airtightness monitoring result. The present invention solves the technical problem in the prior art that the airtightness abnormality type of the bronchial occluder cannot be accurately identified in the final assembly state. By introducing a staged air pressure monitoring and airtightness abnormality prototype identification mechanism based on a digital micro-pressure sensor, the technical effect of automatically identifying the abnormality type under final assembly conditions is achieved. BRIEF DESCRIPTION OF THE DRAWINGS

[0008] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0009] Figure 1 A schematic flow chart of a method for detecting air tightness of a bronchial occluder cuff using dynamic air pressure sensing provided in an embodiment of the present application; Figure 2 A schematic structural diagram of a bronchial occluder cuff air tightness detection device with dynamic air pressure sensing provided in an embodiment of the present application; Figure 3 This is a schematic diagram of the physical reference of the bronchial occluder tested in this application.

[0010] Explanation of the accompanying reference numerals: inflation test module 11 , air tightness determination module 12 , inflation and deflation test module 13 , timing iteration analysis module 14 , monitoring result acquisition module 15 . DETAILED DESCRIPTION

[0011] This application provides a dynamic air pressure sensing bronchial occluder cuff air tightness detection method and device to solve the technical problem in the existing technology that it is impossible to accurately identify the type of air tightness abnormality of the bronchial occluder in the final assembly state. By introducing a staged air pressure monitoring and air tightness abnormality prototype identification mechanism based on a digital micro-pressure sensor, the technical effect of automatically identifying the abnormality type under final assembly conditions is achieved.

[0012] The following will be combined with the accompanying drawings in the embodiments of this application to clearly and completely describe the technical solutions in the embodiments of this application. Obviously, the embodiments described are only some of the embodiments of this application, not all of them. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of this application.

[0013] It should be noted that any variations of the terms "include" and "have" are intended to cover non-exclusive inclusions. For example, a process, method, apparatus, product or server that includes a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or modules that are not explicitly listed or are inherent to these processes, methods, products or devices.

[0014] Example 1, as Figure 1-Figure 3 As shown, the present application provides a method for detecting air tightness of a bronchial occluder cuff using dynamic air pressure sensing, the method comprising: Step S100: The target bronchial occluder is introduced into the bronchial simulation fixture for inflation test. When the air pressure reaches the target working pressure, the air pressure is continuously monitored within a preset first detection window using a digital micro-pressure sensor to obtain an air pressure monitoring data sequence.

[0015] In this embodiment, the target bronchial occluder is first inserted into a bronchial fixture designed to simulate the geometry and support of human bronchial passages, allowing the device to be tested in an environment similar to clinical use. Once the occluder is secured, a gas pressure supply is used to inflate its internal cavity. Gas is continuously injected until the internal pressure reaches the target operating pressure, which is set according to clinical standards, for example, 25 kPa, to simulate the workload of an actual occlusion operation.

[0016] After the air pressure stabilizes and reaches the target operating pressure, the digital micro-pressure sensor is activated to monitor the air pressure status. The digital micro-pressure sensor has high sensitivity (detection resolution can be less than 0.1kPa) and can continuously record small changes in air pressure. Within the preset first detection window, such as 60 seconds, the sensor acquires pressure data at a fixed frequency (e.g., 10 times per second), recording it as a pressure monitoring data sequence.

[0017] Step S200: performing an overall air tightness determination of the final assembly according to the air pressure monitoring data sequence to obtain an overall air tightness determination result of the final assembly.

[0018] In an embodiment of the present application, when determining the overall air tightness of the final assembly based on the air pressure monitoring data sequence, the air pressure monitoring data sequence is first traversed, and the pressure drop rate between adjacent sampling points is calculated using the nearest short-term fluctuation identification function to form a pressure drop rate sequence. The pressure drop rate sequence is then fitted to construct a pressure drop rate curve. Thereafter, based on the pressure drop rate curve, it is identified whether there are nonlinear fluctuation characteristics, such as sudden changes, abnormal inflection points, or continuous declines. If obvious nonlinear fluctuations are identified, it is determined that the target bronchial occluder has an airtightness abnormality in the final assembly state. When no nonlinear fluctuations are identified, it is determined that the target bronchial occluder does not have an airtightness abnormality in the final assembly state. Through this process, the overall air tightness determination result of the final assembly is obtained.

[0019] Furthermore, in the method provided in the embodiment of the application, the overall air tightness determination of the final assembly is performed based on the air pressure monitoring data sequence to obtain the overall air tightness determination result of the final assembly, and further includes: The pressure monitoring data sequence is traversed by using a nearest neighbor short-term fluctuation identification function to identify nearest neighbor short-term fluctuations and determine a pressure drop rate sequence; the pressure drop rate sequence is fitted to construct a pressure drop rate curve; nonlinear fluctuations are identified based on the pressure drop rate curve. If nonlinear fluctuations exist, the overall air tightness determination result of the final assembly is that the overall air tightness of the final assembly is abnormal.

[0020] In the embodiment of the present application, a pre-built nearest neighbor short-term fluctuation identification function is first used to traverse the air pressure monitoring data sequence to perform nearest neighbor short-term fluctuation identification. Through this process, a pressure drop rate sequence is obtained.

[0021] Next, the pressure drop rate series is fitted, using polynomial fitting or spline interpolation to construct a pressure drop rate curve, smoothing local noise in the data and extracting the trend of change. Through this process, the pressure drop rate curve is obtained.

[0022] Nonlinear fluctuation identification is then performed based on the constructed pressure drop rate curve. This process uses a slope segmentation analysis method to segment the pressure drop rate curve, calculate the slope changes between each segment, and identify nonlinear characteristics such as sudden slope changes, dense inflection points, or trend reversals. If significant nonlinear fluctuations in the pressure drop rate are detected in one or more segments, the bronchial occluder is determined to have airtightness abnormalities under the current test conditions.

[0023] The above steps ultimately yield a determination of the overall assembly's airtightness. If nonlinear fluctuations are detected, the output indicates an abnormality in the assembly's overall airtightness. If the pressure drop rate curve fluctuates smoothly and shows no significant abnormalities, the assembly's overall airtightness is considered normal.

[0024] Furthermore, the method provided in the application embodiment also includes: Obtain a nearest neighbor short-term fluctuation identification function, wherein the nearest neighbor short-term fluctuation identification function is: ;in, is the pressure drop rate of the i-th air pressure monitoring data in the air pressure monitoring sequence, is the i-th air pressure monitoring data in the air pressure monitoring sequence, is the i+1th air pressure monitoring data in the air pressure monitoring sequence, is the monitoring time point of the i+1th air pressure monitoring data in the air pressure monitoring sequence, is the monitoring time point of the i-th air pressure monitoring data in the air pressure monitoring sequence.

[0025] In the embodiment of the present application, in order to identify the instantaneous pressure relief trend of the bronchial occluder in the assembled state, the pressure monitoring data sequence is first traversed, and the preset nearest neighbor short-term fluctuation identification function is applied to process the adjacent data points. The nearest neighbor short-term fluctuation identification function is ;in, is the pressure drop rate of the i-th air pressure monitoring data in the air pressure monitoring sequence, is the i-th air pressure monitoring data in the air pressure monitoring sequence, is the i+1th air pressure monitoring data in the air pressure monitoring sequence, is the monitoring time point of the i+1th air pressure monitoring data in the air pressure monitoring sequence, is the monitoring time point of the i-th air pressure monitoring data in the air pressure monitoring sequence. This function is used to measure the drop in air pressure over time and can accurately extract small but continuous pressure release phenomena.

[0026] By traversing the entire data sequence and applying the above function, a pressure drop rate sequence containing all adjacent sampling segments is obtained.

[0027] Step S300: If the result of the overall air tightness determination of the final assembly is that the overall air tightness of the final assembly is abnormal, a dynamic inflation and deflation test is performed on the target bronchial occluder in the bronchial simulation fixture, and the air pressure is continuously monitored within a preset second detection window using the digital micro-pressure sensor to obtain a rebound air pressure monitoring data sequence.

[0028] In this embodiment, when the overall assembly airtightness assessment indicates an abnormality, a dynamic inflation / deflation test is performed on the target bronchial occluder within the simulated bronchial fixture. During this test, gas is first injected into the occluder's internal chamber by controlling the gas pressure supply device until the target operating pressure (e.g., 25 kPa) is reached. Subsequently, the gas is rapidly released to reduce the internal pressure, and the device is inflated again, undergoing a complete inflation-deflation-reinflation cycle. This process simulates the clinical use of the occluder during multiple inflation / deflation adjustments.

[0029] During dynamic inflation and deflation testing, a digital micro-pressure sensor is used to monitor pressure changes in real time within a preset second detection window (e.g., 30 to 60 seconds). The digital micro-pressure sensor has a high resolution (typically 0.1 kPa or higher) and records every instantaneous pressure value at a high sampling frequency (e.g., 10 times per second). This continuous data acquisition generates a rebound pressure monitoring data sequence.

[0030] Step S400: performing time series iterative analysis on the rebound air pressure monitoring data sequence, and performing prototype recognition on the analysis results using the airtightness anomaly prototype library to determine the target airtightness anomaly prototype.

[0031] In an embodiment of the present application, the rebound air pressure monitoring data sequence is first processed according to multiple time granularities, and a multi-time-scale feature extraction method is used to obtain characteristic values ​​such as pressure drop rate, rebound amplitude, and stabilization delay under different observation windows to construct a multi-time-scale rebound air pressure monitoring data feature set. Subsequently, a feature iteration interaction mechanism is used to perform multiple rounds of combination and correlation analysis on the multi-time-scale rebound air pressure monitoring data feature set to generate interactive rebound air pressure monitoring data features. Finally, the interactive rebound air pressure monitoring data features are approximately matched with various types of abnormal data features in a pre-constructed airtightness anomaly prototype library to complete prototype identification, and the prototype type that best matches the current occluder status is identified as the final target airtightness anomaly prototype.

[0032] Furthermore, in the method provided in the embodiment of the application, the rebound air pressure monitoring data sequence is subjected to time series iterative analysis, and the analysis results are subjected to prototype identification using the airtightness anomaly prototype library to determine the target airtightness anomaly prototype, and the method further includes: Feature extraction is performed on the rebound air pressure monitoring data sequence according to multiple time scales to obtain a multi-time scale rebound air pressure monitoring data feature set; the multi-time scale rebound air pressure monitoring data feature set is iteratively interacted to determine the interactive rebound air pressure monitoring data features; the interactive rebound air pressure monitoring data features are approximately matched with the abnormal data features corresponding to each air tightness abnormality prototype in the air tightness abnormality prototype library, and the air tightness prototype with the highest matching degree is used as the target air tightness abnormality prototype.

[0033] In an embodiment of the present application, feature extraction is first performed on the rebound air pressure monitoring data sequence according to multiple time scales. By adopting a multi-time scale feature extraction method, key statistics such as the pressure drop amplitude characteristics, air pressure recovery rate characteristics, and stable time characteristics during the rebound process are extracted in multiple time windows (such as 5 seconds, 10 seconds, and 30 seconds) to form a structured multi-time scale rebound air pressure monitoring data feature set.

[0034] Next, based on the constructed multi-time-scale rebound air pressure monitoring data feature set, feature iterative interaction analysis is performed. In this process, the feature vectors in the multi-time-scale rebound air pressure monitoring data feature set are first randomly combined pairwise to generate a preliminary data feature combination set. The correlation between the subfeatures in any data feature combination is then calculated using a subfeature similarity calculation method. The calculated results are normalized to obtain an interaction matrix set. Next, data enhancement is performed on the corresponding data feature combination set based on the interaction matrix set to obtain an enhanced data feature combination set. Finally, the enhanced data feature combination set is subjected to an overall mean processing to form an interactive rebound air pressure monitoring data feature. This interactive rebound air pressure monitoring data feature is the result of the time-series iterative analysis of the rebound air pressure monitoring data sequence. Finally, in the anomaly identification stage, the obtained interactive rebound air pressure monitoring data feature is approximately matched one-to-one with the anomaly data features corresponding to each air tightness anomaly prototype in the preset air tightness anomaly prototype library. The matching method uses the Euclidean distance matching method to calculate the minimum distance between the interactive features and the prototype features in the high-dimensional feature space and judge the similarity based on this. Specifically, for the interactive rebound air pressure monitoring data features [ , , , ] and prototype features in the airtight anomaly prototype library[ , , , ], calculate its Euclidean distance D = By traversing the entire airtightness anomaly prototype library, comparing and sorting all distance values, and selecting the anomaly prototype with the smallest distance, that is, the highest similarity, as the target airtightness anomaly prototype for the current detection object. By extracting key statistical features at multiple time scales and performing feature interaction and matching recognition, the accuracy of air pressure monitoring anomaly type recognition is improved.

[0035] Furthermore, the method provided in the application embodiment also includes: Obtain K airtightness anomaly types and corresponding K historical rebound air pressure monitoring data feature sets, where K is a positive integer; perform representative feature extraction on the K historical rebound air pressure monitoring data feature sets to determine K abnormal data features; associate the K airtightness anomaly types with the K abnormal data features to construct the airtightness anomaly prototype library.

[0036] In this embodiment, historical test data is first aggregated to obtain K airtightness anomaly types and their corresponding K historical rebound pressure monitoring data feature sets, where K is a positive integer representing the number of identifiable anomaly types. Each airtightness anomaly type corresponds to multiple rebound test samples. Through standardization and multi-time series feature extraction, a feature set with consistent structure and unified parameters is formed to express the pressure change behavior under various abnormal conditions.

[0037] During the feature extraction phase, principal component analysis (PCA) was used to reduce the dimensionality of each set of historical rebound pressure monitoring data. This method calculates the covariance matrix between dimensions and extracts the principal components with the largest variance contribution, thereby retaining the most discriminative characteristic patterns in the original data. For each anomaly type, only the most representative eigenvector is retained as the core representation of that anomaly state, ultimately forming K anomaly data features.

[0038] Finally, the K extracted abnormal data features are associated one-to-one with the corresponding K airtightness anomaly types, and an airtightness anomaly prototype library is constructed through structured mapping relationships. The airtightness anomaly prototype library is shown in Table 1.

[0039] Table 1: Examples of airtightness anomaly types Furthermore, the method provided in the application embodiment also includes: The multi-time-scale rebound air pressure monitoring data feature sets are randomly combined in pairs to obtain a data feature combination set; sub-feature similarity calculation is performed on any data feature combination in the data feature combination set, and the calculation results are normalized to construct an interaction matrix set; data enhancement is performed on the data feature combination sets according to the interaction matrix set to obtain an enhanced data feature combination set; and overall mean processing is performed on the enhanced data feature combination set to obtain the interactive rebound air pressure monitoring data features.

[0040] In an embodiment of the present application, each feature vector in a multi-time-scale rebound air pressure monitoring data feature set is first randomly combined in pairs. This combination method does not have a fixed order. By setting the number of combinations, any two features in the feature set are randomly paired to generate multiple feature pairs, forming a data feature combination set. For example, if the multi-time-scale rebound air pressure monitoring data feature set contains 10 feature vectors and the number of combinations is set to 15, then 15 feature combinations are formed through random sampling (perhaps using a pairwise sampling algorithm without replacement), each containing two feature vectors of identical length, forming the data feature combination set.

[0041] After forming a set of data feature combinations, a sub-feature similarity calculation is performed on each feature combination. Sub-feature similarity calculation involves aligning two feature vectors in a set of feature combinations by dimension and comparing the numerical differences along each corresponding dimension. Specifically, the absolute difference between the two eigenvalues ​​is calculated for each dimension in turn to measure the degree of local difference between the two features along that dimension. The resulting set of differences constitutes a similarity vector, which is used to represent the difference structure of the feature combination across each dimension. For example, if two feature vectors respectively contain indicators such as pressure drop amplitude, rebound rate, and stabilization time, the similarity vector also contains five elements, corresponding to the degree of difference in these five indicators. In this process, if the pressure drop amplitude of feature 1 is 5.2kPa and that of feature 2 is 4.8kPa, the difference in this dimension is 0.4kPa; if the rebound rates are 0.8kPa / s and 0.5kPa / s respectively, the difference is 0.3kPa / s. Similarly, the difference in each dimension is calculated to form a sub-feature similarity vector.

[0042] In order to eliminate the impact of differences in the dimensions and numerical ranges of different indicators on subsequent processing, the similarity vector is standardized using a normalization method. The commonly used methods are minimum-maximum normalization or Z-score normalization. The normalized results are constructed as an interaction matrix set. For example, in the minimum-maximum normalization process, each value in the sub-feature similarity vector is linearly transformed according to the minimum and maximum values ​​of the dimension so that it falls into the interval [0, 1]. If the value range of a certain dimension is [0.0, 2.0], the normalized result of the above 0.4 is 0.2. The interaction matrix set formed after normalization is a standardized similarity data set that corresponds one-to-one to the data feature combination set, which is used for numerical adjustment and structural mapping in the subsequent enhancement process.

[0043] After constructing the interaction matrix, a neural network-based data augmentation mechanism is introduced. A feedforward neural network model is constructed to learn the nonlinear relationship between the original feature combinations and their interaction matrices. This neural network consists of an input layer, a hidden layer, and an output layer. The input layer receives the concatenated feature pairs and the interaction matrix, the hidden layer extracts deep nonlinear features, and the output layer provides an enhanced feature representation. The network is trained on known anomaly samples. By repeatedly inputting different combinations of samples, it outputs an enhanced feature representation, forming a set of enhanced data feature combinations. Each set of enhanced feature vectors retains the original temporal information while incorporating the structural differences between the two features.

[0044] Finally, the overall mean value of all enhanced data feature combinations is taken. Multiple enhanced features generated from the same set of original feature combinations are averaged dimension by dimension to remove local noise caused by model perturbations and obtain stable and representative interactive rebound air pressure monitoring data features.

[0045] Furthermore, the method provided in the application embodiment also includes: Traversing the K historical rebound air pressure monitoring data feature sets to extract abnormal time windows, and obtaining K abnormal time window sets; extracting the maximum value and the minimum value in the K abnormal time window sets respectively, and summarizing them to obtain the multi-time series scale.

[0046] In an embodiment of the present application, K historical rebound air pressure monitoring data feature sets are first traversed and processed, and an abnormal time window extraction operation is performed. This operation adopts a sliding threshold detection method, sets a sliding window of fixed width on each rebound air pressure monitoring data sequence, and calculates the pressure drop rate feature and the rebound slope feature in each window. When the pressure drop rate in a continuous window exceeds a set threshold (such as -2kPa / s) or the rebound slope is lower than a set threshold (such as 0.5kPa / s), the time interval is marked as an abnormal response interval, and its start time and end time are recorded to complete the positioning of the abnormal time window. In this way, all K types of historical abnormal samples are processed, and finally K abnormal time window sets are extracted.

[0047] For a set of K abnormal time windows, we use a time interval boundary extraction method to calculate the time span of all abnormal time windows. This is done by subtracting the start time from the end time of the window to obtain a number of response durations. Then, within each abnormal time window set, we extract the maximum and minimum time span values ​​within that set. Finally, we aggregate the K maximum and K minimum values ​​extracted from the K abnormal time window sets to form a dataset containing 2K values. All the maximum and minimum values ​​in this dataset together constitute a multi-time series scale.

[0048] Step S500: When the abnormality type of the target airtightness abnormal prototype is cuff body leakage, the tail of the catheter in the target bronchial occluder is sealed, and the cavity of the cuff body is directionally pressurized to perform a closed pressure maintenance test to obtain the target cuff airtightness monitoring result.

[0049] In the embodiment of the present application, when the abnormal type of the target air tightness abnormal prototype is a leakage of the cuff body, a closed pressure maintenance test method is used to verify the target bronchial occluder. Specifically, the cuff directional pressurization clamp is first used to seal the tail of the catheter of the target bronchial occluder to prevent gas leakage from the tail from interfering with the test; then, gas is applied to the chamber of the cuff body with the help of a gas pressure supply, and a directional pressurization operation is performed. The gas is continuously injected until the air pressure in the chamber reaches the preset directional pressurization threshold. At this time, the gas pressure supply is immediately closed to form a closed environment. In this closed state, a closed pressure maintenance test is performed, that is, a digital micro-pressure sensor is used to continuously record the pressure change process inside the cavity to obtain a continuous closed pressure data sequence; finally, a multi-dimensional analysis is performed on the data sequence to determine whether the pressure has abnormally decayed, thereby obtaining the target cuff air tightness monitoring result.

[0050] Furthermore, in the method provided in the embodiment of the application, when the abnormal type of the target airtightness abnormal prototype is a cuff body leak, the tail of the catheter in the target bronchial occluder is sealed, and the cavity of the cuff body is directionally pressurized to perform a closed pressure retention test to obtain the target cuff airtightness monitoring result, which also includes: Obtain a cuff directional pressurization fixture and use the cuff directional pressurization fixture to seal the tail of the catheter; obtain a gas pressure supply to directionally pressurize the chamber of the cuff body until a preset directional pressurization threshold is reached, turn off the gas pressure supply and use a digital micro-pressure sensor to perform closed pressure changes to obtain a closed pressure data sequence; perform multi-dimensional analysis on the closed pressure data sequence to obtain the target cuff airtightness monitoring result.

[0051] In this embodiment, when verifying the cuff of a target bronchial occluder for leaks, a closed pressure hold test is performed to assess its airtightness. First, a cuff directional pressurization fixture is obtained and installed to seal the occluder's catheter tail, ensuring that no gas leaks through the tail during the test, thereby establishing a complete closed testing environment.

[0052] The cuff's chamber is then pressurized using a gas supply, continuing until the pressure inside the chamber reaches a preset threshold (e.g., 20 cmH2O). Upon reaching this threshold, the supply is immediately shut off, sealing the chamber. A digital micro-pressure sensor continuously monitors the chamber's pressure in this sealed state, recording the entire process of pressure changes over time to form a closed-chamber pressure data sequence.

[0053] After data collection is complete, a multi-dimensional analysis of the pressure data sequence is performed, including calculation of the pressure change rate per unit time (ΔP / Δt) and analysis of the pressure curve trend. If the pressure drop rate is less than or equal to 0.5 cmH2O / min and the overall change curve is stable without significant fluctuations, the occluder is considered to be airtight. If the pressure drop rate is greater than 0.5 cmH2O / min, or if the pressure change exhibits abnormal trends such as "pressure drop followed by rebound" or "continuous fluctuations," the occluder is considered to be airtight, indicating a cuff leak.

[0054] Finally, based on the four steps of sealing establishment, directional pressurization control, pressure monitoring and trend analysis, the target cuff airtightness monitoring results are obtained, and it is clearly judged whether the airtightness is normal or abnormal.

[0055] Furthermore, the method provided in the application embodiment also includes: If the target cuff air tightness monitoring result shows that there is an air tightness abnormality, the gas pressure supply device is used again to directionally pressurize the chamber of the cuff body until the preset directional pressurization threshold is reached, and the pressurized cuff body is immersed in water, and the position where the bubbles overflow is used as the air tightness abnormality position.

[0056] In this embodiment of the present application, if the closed-type pressure hold test determines that the target cuff's airtightness monitoring result indicates an airtightness abnormality, a bubble overflow location detection procedure is performed. During this procedure, the gas pressure supply is used to re-direct pressurize the cuff's chamber. Using a constant pressure control mode, the internal pressure of the chamber is steadily raised to a preset directed pressurization threshold (e.g., 20 cmH2O), ensuring consistent and repeatable pressurization conditions. While maintaining the pressurized state, the inflated cuff is slowly immersed in water, preferably in a transparent container for easy observation.

[0057] While submerged in water, observe the cuff surface for bubbles. Based on the principle of leak detection, under pressurized conditions, gas within the chamber will escape through the leak path, forming continuous or intermittent bubbles in the water. Visually observe the location of the bubbles with the naked eye or a high-resolution camera; this location indicates an airtightness anomaly.

[0058] In the embodiments of the present application, in summary, the embodiments of the present application have at least the following technical effects: In the present application, a target bronchial occluder is introduced into a simulated bronchial fixture for inflation test. When the air pressure reaches the target working pressure, a digital micro-pressure sensor is used to continuously monitor the air pressure in a preset first detection window to obtain an air pressure monitoring data sequence. The overall air tightness of the assembly is determined based on the air pressure monitoring data sequence to obtain an overall air tightness determination result of the assembly. If the overall air tightness determination result of the assembly is that the overall air tightness of the assembly is abnormal, a dynamic inflation and deflation test is performed on the target bronchial occluder in the simulated bronchial fixture, and the digital micro-pressure sensor is used to continuously monitor the air pressure in a preset first detection window to obtain an air pressure monitoring data sequence. The micro-pressure sensor continuously monitors the air pressure within a preset second detection window to obtain a rebound air pressure monitoring data sequence; the rebound air pressure monitoring data sequence is subjected to time-series iterative analysis, and the analysis results are prototype-identified using the airtightness abnormality prototype library to determine the target airtightness abnormality prototype; when the abnormality type of the target airtightness abnormality prototype is a cuff body leak, the tail of the catheter in the target bronchial occluder is sealed, and the chamber of the cuff body is directionally pressurized to perform a closed pressure retention test to obtain the target cuff airtightness monitoring result. The present invention solves the technical problem in the prior art that the airtightness abnormality type of the bronchial occluder cannot be accurately identified in the final assembly state. By introducing a staged air pressure monitoring and airtightness abnormality prototype identification mechanism based on a digital micro-pressure sensor, the technical effect of automatically identifying the abnormality type under final assembly conditions is achieved.

[0059] The second embodiment is based on the same inventive concept as the method for detecting air tightness of the bronchial occluder cuff using dynamic air pressure sensing in the above embodiment. Figure 2 As shown, the present application provides a dynamic air pressure sensing bronchial occluder cuff air tightness detection device. The device and method embodiments in the present application are based on the same inventive concept. The device includes: The inflation test module 11 is used to introduce the target bronchial occluder into the bronchial imitation clamp for inflation test. When the air pressure reaches the target working pressure, the digital micro-pressure sensor is used to continuously monitor the air pressure in the preset first detection window to obtain an air pressure monitoring data sequence; the air tightness judgment module 12 is used to judge the overall air tightness of the assembly according to the air pressure monitoring data sequence to obtain the overall air tightness judgment result of the assembly; the inflation and deflation test module 13 is used to perform dynamic inflation and deflation tests on the target bronchial occluder in the bronchial imitation clamp if the overall air tightness judgment result of the assembly is that the overall air tightness of the assembly is abnormal, and use the The digital micro-pressure sensor continuously monitors the air pressure within a preset second detection window to obtain a rebound air pressure monitoring data sequence; the time series iterative analysis module 14 is used to perform time series iterative analysis on the rebound air pressure monitoring data sequence, and use the air tightness abnormality prototype library to perform prototype identification on the analysis results to determine the target air tightness abnormality prototype; the monitoring result acquisition module 15 is used to seal the tail of the catheter in the target bronchial occluder when the abnormal type of the target air tightness abnormality prototype is air leakage of the cuff body, and directionally pressurize the cavity of the cuff body to perform a closed pressure maintenance test to obtain the target cuff air tightness monitoring result.

[0060] Furthermore, the device is also used to implement the following functions: Feature extraction is performed on the rebound air pressure monitoring data sequence according to multiple time scales to obtain a multi-time scale rebound air pressure monitoring data feature set; the multi-time scale rebound air pressure monitoring data feature set is iteratively interacted to determine the interactive rebound air pressure monitoring data features; the interactive rebound air pressure monitoring data features are approximately matched with the abnormal data features corresponding to each air tightness abnormality prototype in the air tightness abnormality prototype library, and the air tightness prototype with the highest matching degree is used as the target air tightness abnormality prototype.

[0061] Furthermore, the device is also used to implement the following functions: Obtain K airtightness anomaly types and corresponding K historical rebound air pressure monitoring data feature sets, where K is a positive integer; perform representative feature extraction on the K historical rebound air pressure monitoring data feature sets to determine K abnormal data features; associate the K airtightness anomaly types with the K abnormal data features to construct the airtightness anomaly prototype library.

[0062] Furthermore, the device is also used to implement the following functions: The multi-time-scale rebound air pressure monitoring data feature sets are randomly combined in pairs to obtain a data feature combination set; sub-feature similarity calculation is performed on any data feature combination in the data feature combination set, and the calculation results are normalized to construct an interaction matrix set; data enhancement is performed on the data feature combination sets according to the interaction matrix set to obtain an enhanced data feature combination set; and overall mean processing is performed on the enhanced data feature combination set to obtain the interactive rebound air pressure monitoring data features.

[0063] Furthermore, the device is also used to implement the following functions: Traversing the K historical rebound air pressure monitoring data feature sets to extract abnormal time windows, and obtaining K abnormal time window sets; extracting the maximum value and the minimum value in the K abnormal time window sets respectively, and summarizing them to obtain the multi-time series scale.

[0064] Furthermore, the device is also used to implement the following functions: Obtain a cuff directional pressurization fixture and use the cuff directional pressurization fixture to seal the tail of the catheter; obtain a gas pressure supply to directionally pressurize the chamber of the cuff body until a preset directional pressurization threshold is reached, turn off the gas pressure supply and use a digital micro-pressure sensor to perform closed pressure changes to obtain a closed pressure data sequence; perform multi-dimensional analysis on the closed pressure data sequence to obtain the target cuff airtightness monitoring result.

[0065] Furthermore, the device is also used to implement the following functions: If the target cuff air tightness monitoring result shows that there is an air tightness abnormality, the gas pressure supply device is used again to directionally pressurize the chamber of the cuff body until the preset directional pressurization threshold is reached, and the pressurized cuff body is immersed in water, and the position where the bubbles overflow is used as the air tightness abnormality position.

[0066] Furthermore, the device is also used to implement the following functions: The pressure monitoring data sequence is traversed by using a nearest neighbor short-term fluctuation identification function to identify nearest neighbor short-term fluctuations and determine a pressure drop rate sequence; the pressure drop rate sequence is fitted to construct a pressure drop rate curve; nonlinear fluctuations are identified based on the pressure drop rate curve. If nonlinear fluctuations exist, the overall air tightness determination result of the final assembly is that the overall air tightness of the final assembly is abnormal.

[0067] Furthermore, the device is also used to implement the following functions: Obtain a nearest neighbor short-term fluctuation identification function, wherein the nearest neighbor short-term fluctuation identification function is: ;in, is the pressure drop rate of the i-th air pressure monitoring data in the air pressure monitoring sequence, is the i-th air pressure monitoring data in the air pressure monitoring sequence, is the i+1th air pressure monitoring data in the air pressure monitoring sequence, is the monitoring time point of the i+1th air pressure monitoring data in the air pressure monitoring sequence, is the monitoring time point of the i-th air pressure monitoring data in the air pressure monitoring sequence.

[0068] It should be noted that the order in which the embodiments of the present application are presented is for illustrative purposes only and does not necessarily represent the superiority or inferiority of the embodiments. Furthermore, the foregoing descriptions of specific embodiments of this specification are provided. The processes depicted in the accompanying drawings do not necessarily require the specific order or sequential sequence shown to achieve the desired results. In certain embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0069] The above description is only a preferred embodiment of the present application and is not intended to limit the present application. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of the present application shall be included in the scope of protection of the present application.

[0070] This specification and drawings are merely illustrative of the present application and are intended to cover any and all modifications, variations, combinations, or equivalents within the scope of this application. Obviously, those skilled in the art may make various modifications and variations to this application without departing from the scope of this application. Thus, this application is intended to include such modifications and variations as fall within the scope of this application and its equivalents.

Claims

1. A method for detecting air tightness of a bronchial occluder cuff using dynamic air pressure sensing, characterized in that: The method comprises: The target bronchial occluder is introduced into the bronchial simulation fixture for inflation testing. When the air pressure reaches the target working pressure, the digital micro-pressure sensor is used to continuously monitor the air pressure within the preset first detection window to obtain an air pressure monitoring data sequence; Performing an overall air tightness determination of the final assembly according to the air pressure monitoring data sequence to obtain an overall air tightness determination result of the final assembly; If the result of the overall air tightness determination of the final assembly is that the overall air tightness of the final assembly is abnormal, a dynamic inflation and deflation test is performed on the target bronchial occluder in the bronchial simulation fixture, and the air pressure is continuously monitored within a preset second detection window using the digital micro-pressure sensor to obtain a rebound air pressure monitoring data sequence; Performing time series iterative analysis on the rebound air pressure monitoring data sequence, and performing prototype identification on the analysis results using the air tightness anomaly prototype library to determine the target air tightness anomaly prototype; When the abnormal type of the target airtightness abnormal prototype is cuff body leakage, the tail of the catheter in the target bronchial occluder is sealed, and the cavity of the cuff body is directionally pressurized to perform a closed pressure maintenance test to obtain the target cuff airtightness monitoring result.

2. The method for detecting air tightness of a bronchial occluder cuff using dynamic air pressure sensing according to claim 1, wherein: Performing time series iterative analysis on the rebound air pressure monitoring data sequence, and using the airtightness anomaly prototype library to perform prototype identification on the analysis results to determine the target airtightness anomaly prototype, including: Extract features from the rebound air pressure monitoring data sequence according to multiple time series scales to obtain a multi-time series scale rebound air pressure monitoring data feature set; Iteratively interacting the multi-time-scale rebound air pressure monitoring data feature set to determine interactive rebound air pressure monitoring data features; Approximately match the interactive rebound air pressure monitoring data features with the abnormal data features corresponding to each airtightness abnormality prototype in the airtightness abnormality prototype library, and use the airtightness prototype with the highest matching degree as the target airtightness abnormality prototype.

3. The method for detecting air tightness of a bronchial occluder cuff using dynamic air pressure sensing according to claim 2, wherein: include: Obtain K airtightness anomaly types and corresponding K historical rebound air pressure monitoring data feature sets, where K is a positive integer; Extract representative features from the K historical rebound air pressure monitoring data feature sets to determine K abnormal data features; The K airtightness anomaly types and the K abnormal data features are associated to construct the airtightness anomaly prototype library.

4. The method for detecting air tightness of a bronchial occluder cuff using dynamic air pressure sensing according to claim 3, wherein: include: Randomly combining the multi-time-scale rebound air pressure monitoring data feature sets in pairs to obtain a data feature combination set; Calculate the sub-feature similarity of any data feature combination in the data feature combination set, normalize the calculation results, and construct an interaction matrix set; Performing data enhancement on the data feature combination sets according to the interaction matrix set to obtain an enhanced data feature combination set; Performing overall mean processing on the enhanced data feature combination set to obtain the interactive rebound air pressure monitoring data feature.

5. The method for detecting air tightness of a bronchial occluder cuff using dynamic air pressure sensing according to claim 3, wherein: include: Traversing the K historical rebound air pressure monitoring data feature sets to extract abnormal time windows and obtain K abnormal time window sets; The maximum value and the minimum value in the K abnormal time window sets are extracted respectively, and the multi-time series scale is obtained by summarizing them.

6. The method for detecting air tightness of a bronchial occluder cuff using dynamic air pressure sensing according to claim 1, wherein: When the abnormality type of the target airtightness abnormal prototype is cuff body leakage, the tail of the catheter in the target bronchial occluder is sealed, and the cavity of the cuff body is directionally pressurized to perform a closed pressure retention test to obtain the target cuff airtightness monitoring results, including: Obtaining a cuff directional pressurizing fixture, and using the cuff directional pressurizing fixture to seal the tail of the catheter; The gas pressure supply device is used to directionally pressurize the chamber of the cuff body until a preset directional pressurization threshold is reached, the gas pressure supply device is turned off, and a closed pressure change is performed using a digital micro-pressure sensor to obtain a closed pressure data sequence; A multi-dimensional analysis is performed on the closed pressure data sequence to obtain the target cuff airtightness monitoring result.

7. The method for detecting air tightness of a bronchial occluder cuff using dynamic air pressure sensing according to claim 6, wherein: If the target cuff air tightness monitoring result shows that there is an air tightness abnormality, the gas pressure supply device is used again to directionally pressurize the chamber of the cuff body until the preset directional pressurization threshold is reached, and the pressurized cuff body is immersed in water, and the position where the bubbles overflow is used as the air tightness abnormality position.

8. The method for detecting air tightness of a bronchial occluder cuff using dynamic air pressure sensing according to claim 1, wherein: Performing an overall air tightness determination of the final assembly according to the air pressure monitoring data sequence to obtain an overall air tightness determination result of the final assembly, including: Using a nearest neighbor short-term fluctuation identification function to traverse the air pressure monitoring data sequence to identify nearest neighbor short-term fluctuations and determine a pressure drop rate sequence; Fitting the pressure drop rate sequence to construct a pressure drop rate curve; Nonlinear fluctuation identification is performed based on the pressure drop rate curve. If nonlinear fluctuation exists, the overall air tightness determination result of the final assembly is that the overall air tightness of the final assembly is abnormal.

9. The method for detecting air tightness of a bronchial occluder cuff using dynamic air pressure sensing according to claim 8, wherein: include: Obtain a nearest neighbor short-term fluctuation identification function, wherein the nearest neighbor short-term fluctuation identification function is: ； in, is the pressure drop rate of the i-th air pressure monitoring data in the air pressure monitoring sequence, is the i-th air pressure monitoring data in the air pressure monitoring sequence, is the i+1th air pressure monitoring data in the air pressure monitoring sequence, is the monitoring time point of the i+1th air pressure monitoring data in the air pressure monitoring sequence, is the monitoring time point of the i-th air pressure monitoring data in the air pressure monitoring sequence.

10. A dynamic air pressure sensing bronchial occluder cuff air tightness detection device, characterized in that: The device is used to perform the method for detecting air tightness of a bronchial occluder cuff using dynamic air pressure sensing as described in any one of claims 1 to 9, and the device comprises: The inflation test module is used to introduce the target bronchial occluder into the bronchial simulation fixture for inflation testing. When the air pressure reaches the target working pressure, the digital micro-pressure sensor is used to continuously monitor the air pressure within the preset first detection window to obtain an air pressure monitoring data sequence; An air tightness determination module is used to determine the overall air tightness of the final assembly according to the air pressure monitoring data sequence to obtain an overall air tightness determination result of the final assembly; an inflation and deflation test module, configured to perform a dynamic inflation and deflation test on the target bronchial occluder in the bronchial simulation fixture if the result of the overall air tightness determination of the assembly indicates that the overall air tightness of the assembly is abnormal, and to continuously monitor the air pressure within a preset second detection window using the digital micro-pressure sensor to obtain a rebound air pressure monitoring data sequence; A time series iterative analysis module is used to perform time series iterative analysis on the rebound air pressure monitoring data sequence, and use the air tightness anomaly prototype library to perform prototype identification on the analysis results to determine the target air tightness anomaly prototype; The monitoring result acquisition module is used to seal the tail of the catheter in the target bronchial occluder when the abnormal type of the target airtightness abnormal prototype is air leakage of the cuff body, and to directionally pressurize the cavity of the cuff body to perform a closed pressure maintenance test to obtain the target cuff airtightness monitoring result.

Images