Transluminal ventricular assist device and aortic pressure estimation method, apparatus and storage medium thereof

Abstract

This application provides a percutaneous ventricular assist device and its aortic pressure estimation method, apparatus, and storage medium. The method includes: acquiring a gas-side pressure signal collected by a gas-side pressure sensor and determining whether a pneumatic diaphragm is in an effective pressure transmission state; when the pneumatic diaphragm is in an effective pressure transmission state, determining an effective identification window corresponding to the gas-side pressure signal meeting a preset pressure signal quality condition and the pneumatic operation state meeting a preset operation state condition based on the gas-side pressure signal and the pneumatic operation state of the percutaneous ventricular assist device; acquiring a gas-side dynamic pressure sequence within the effective identification window; and determining the mapping relationship between the gas-side dynamic pressure sequence and aortic pressure based on a pre-established multi-condition aortic pressure back-calculation model to obtain aortic pressure estimation results. This method enables stable and accurate aortic pressure estimation using existing gas-side pressure signals without adding a dedicated blood-side pressure sensor.

Description

Technical Field

[0001] This invention relates to the field of medical device technology, specifically to a percutaneous ventricular assist device and its aortic pressure estimation method, apparatus, and storage medium. Background Technology

[0002] A percutaneous ventricular assist device (pVAD) is a ventricular assist device that provides circulatory support to patients through interventional procedures. One type of pVAD employs a pneumatic diaphragm structure, comprising a blood-side pump chamber, a gas-side diaphragm chamber, and a pneumatic diaphragm separating the two. Driven by gas-side pressure within the gas-side diaphragm chamber, the pneumatic diaphragm periodically changes the volume of the blood-side pump chamber, delivering blood from the inlet side to the aortic side. Aortic pressure constitutes the output load during the pumping process, reflecting the perfusion status of the patient's circulatory system; its changes affect the pumping process, single-pump volume, and actual assist flow rate. Therefore, obtaining an estimated aortic pressure during the operation of a pVAD is crucial for evaluating the assist effect, determining the output load, estimating flow rate, adjusting operating parameters, and displaying pressure trends.

[0003] Currently, aortic pressure is typically measured using invasive arterial pressure catheters or pressure sensors placed in the bloodstream. While these methods provide a reference pressure, they require the addition of dedicated blood-side pressure sensors or pressure tapping structures in the bloodstream. This increases the cost of blood-contact components, sealing interfaces, electrical connections, sterilization requirements, and disposable consumables, and may introduce issues such as thrombosis, zero-point drift, installation location, and long-term reliability.

[0004] For pneumatic diaphragm-type percutaneous ventricular assist devices, although a pressure sensor for gas path control is provided on the gas side, the gas side pressure signal is a composite signal, which is affected by a variety of factors such as gas source pressure, valve status, gas flow, gas compression, diaphragm position, and blood side load. Specifically, the amplitude and phase of the blood-side aortic pressure response in the gas-side pressure signal change with valve status, diaphragm position, and gas flow state. The diaphragm has different effective areas, equivalent stiffness, and damping under relaxed, wrinkled, wall-adhered, large-stroke motion, and stable tension states, resulting in a lack of stable boundary conditions for pressure transmission. Pressure regulating valve operation, gas source fluctuations, motor pressure regulation, gas mass changes, and temperature changes introduce additional disturbance components into the gas-side pressure signal. Furthermore, the transmission of aortic pressure to the gas-side pressure signal exhibits inertial, damped, time-delay, and nonlinear characteristics; the same gas-side pressure level may correspond to different aortic pressures at different waveform stages. In addition, various factors such as patient aortic pressure, heart rate, pulse pressure, system resistance, compliance, diaphragm pump parameters, gas-side volume, and tubing parameters all affect the pressure transmission relationship, making it difficult for calibration coefficients obtained under a single operating condition to cover the diverse physiological and equipment states encountered in actual use. Therefore, existing technologies cannot directly and stably estimate aortic pressure using gas-side pressure signals. Summary of the Invention

[0005] This application provides a percutaneous ventricular assist device and its aortic pressure estimation method, apparatus and storage medium, aiming to solve the technical problem that in existing pneumatic diaphragm percutaneous ventricular assist devices, the gas-side pressure signal is a composite signal affected by multiple factors such as gas source pressure, valve status, gas flow, diaphragm position and blood-side load, making it difficult to directly use the gas-side pressure signal to stably and accurately estimate aortic pressure without adding a blood-side pressure sensor.

[0006] In a first aspect, embodiments of this application provide an aortic pressure estimation method for a percutaneous ventricular assist device. The percutaneous ventricular assist device includes a blood-side pump chamber, a gas-side membrane chamber, a pneumatic diaphragm separating the blood-side pump chamber and the gas-side membrane chamber, a gas path and pressure regulating component communicating with the gas-side membrane chamber, and a gas-side pressure sensor communicating with the gas-side membrane chamber. The pneumatic diaphragm, driven by the gas-side pressure within the gas-side membrane chamber, periodically changes the volume of the blood-side pump chamber, delivering blood from the inlet side to the aortic side. The aortic pressure estimation method includes:

[0007] Acquire the gas-side pressure signal collected by the gas-side pressure sensor;

[0008] Based on the gas-side pressure signal, it is determined whether the pneumatic diaphragm is in an effective pressure transmission state; wherein, when the pneumatic diaphragm is in the effective pressure transmission state, its response to pressure changes satisfies a preset stability condition.

[0009] When the pneumatic diaphragm is in the effective pressure transmission state, an effective identification window for aortic pressure estimation is determined based on the gas-side pressure signal and the pneumatic operation state of the percutaneous ventricular assist device; wherein, the effective identification window is the time interval corresponding to the gas-side pressure signal meeting the preset pressure signal quality condition and the pneumatic operation state of the percutaneous ventricular assist device meeting the preset operation state condition.

[0010] Obtain the gas-side dynamic pressure sequence within the effective identification window;

[0011] Based on a pre-established multi-condition aortic pressure back-calculation model, the dynamic mapping relationship between the gas-side dynamic pressure sequence and the aortic pressure is determined, and the aortic pressure estimation result is obtained.

[0012] Secondly, embodiments of this application provide an aortic pressure estimation device for percutaneous ventricular assist devices, comprising:

[0013] The gas-side pressure acquisition module is used to acquire the gas-side pressure signal collected by the gas-side pressure sensor.

[0014] The diaphragm state determination module is used to determine whether the pneumatic diaphragm is in an effective pressure transmission state based on the gas-side pressure signal; wherein, the pneumatic diaphragm is in a preloaded stress state in the effective pressure transmission state, and the response of the pneumatic diaphragm under pressure change satisfies a preset stability condition.

[0015] The identification window determination module is used to determine an effective identification window for aortic pressure estimation based on the gas-side pressure signal and the pneumatic operating state of the percutaneous ventricular assist device when the pneumatic diaphragm is in the effective pressure transmission state; wherein, the effective identification window is the time interval corresponding to the gas-side pressure signal satisfying a preset pressure signal quality condition and the pneumatic operating state of the percutaneous ventricular assist device satisfying a preset operating state condition.

[0016] The dynamic pressure acquisition module is used to acquire the gas-side dynamic pressure sequence within the effective identification window;

[0017] The aortic pressure estimation module is used to determine the dynamic mapping relationship between the gas-side dynamic pressure sequence and the aortic pressure based on a pre-established multi-condition aortic pressure back-calculation model, and to obtain the aortic pressure estimation result.

[0018] Thirdly, embodiments of this application provide a percutaneous ventricular assist device, comprising:

[0019] The blood-side pump chamber is connected to the blood inlet side and the aortic side, and is used to contain the blood to be transported and to transport the blood from the blood inlet side to the aortic side when the volume changes;

[0020] A gas-side membrane cavity is used to contain gas, and the pressure change of the gas drives the volume change of the blood-side pump cavity;

[0021] A pneumatic diaphragm is disposed between the blood-side pump chamber and the gas-side diaphragm chamber, and is used to periodically change the volume of the blood-side pump chamber under the drive of the gas-side pressure in the gas-side diaphragm chamber, so as to drive the blood in the blood-side pump chamber to be transported from the blood inlet side to the aortic side;

[0022] The gas path and pressure regulating component are connected to the gas-side membrane cavity and are used to provide gas-side pressure to the gas-side membrane cavity and regulate the gas-side pressure in the gas-side membrane cavity so that the pneumatic diaphragm is in a preloaded state.

[0023] A gas-side pressure sensor, connected to the gas-side membrane cavity, is used to collect the gas-side pressure signal within the gas-side membrane cavity;

[0024] The controller, connected to the gas-side pressure sensor and the gas path and pressure regulating components, is configured to acquire the gas-side pressure signal and perform the aortic pressure estimation method for a percutaneous ventricular assist device as described in any embodiment of the first aspect.

[0025] Fourthly, embodiments of this application provide a computer-readable storage medium storing computer-executable instructions, which, when executed by a processor, are used to implement the aortic pressure estimation method for percutaneous ventricular assist devices as described in any embodiment of the first aspect.

[0026] The percutaneous ventricular assist device and its aortic pressure estimation method and apparatus provided in this application acquire gas-side pressure signals and determine whether the pneumatic diaphragm is in an effective pressure transmission state, enabling the diaphragm to operate under preload and stable conditions, providing repeatable boundary conditions for the transmission of aortic pressure to gas-side pressure. Based on this, an effective identification window is determined according to the gas-side pressure signal and pneumatic operating status, enabling the acquisition of gas-side dynamic pressure sequences within a low-disturbance, high-signal-quality time interval, effectively suppressing the influence of interference components such as pressure regulating valve action, gas source fluctuations, and gas flow on the estimation results. Furthermore, a pre-established multi-condition aortic pressure back-calculation model is used to process the gas-side dynamic pressure sequences, adapting to pressure transmission relationships under different heart rates, aortic pressures, system resistances, and compliance conditions. This allows for stable and accurate aortic pressure estimation results without adding a dedicated blood-side pressure sensor, reducing the cost of blood-contact components and the risk of complications such as thrombosis, and providing a reliable basis for evaluating the assistive effect, determining the output load, estimating flow rate, and adjusting operating parameters.

[0027] In addition, this application also provides a computer-readable storage medium, a computer program product, and a chip that have the same beneficial effects as the aortic pressure estimation method for percutaneous ventricular assist devices described above. Attached Figure Description

[0028] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this application and, together with the description, serve to explain the principles of this application.

[0029] Figure 1 This is a schematic diagram of the structure of a percutaneous ventricular assist device according to one embodiment of this application.

[0030] Figure 2 This is a flowchart of an aortic pressure estimation method for a percutaneous ventricular assist device provided in one embodiment of this application.

[0031] Figure 3 A flowchart of an aortic pressure estimation method for a percutaneous ventricular assist device provided in another embodiment of this application.

[0032] Figure 4 A flowchart of an aortic pressure estimation method for a percutaneous ventricular assist device provided in another embodiment of this application.

[0033] Figure 5 A flowchart of an aortic pressure estimation method for a percutaneous ventricular assist device provided in another embodiment of this application.

[0034] Figure 6 A flowchart of an aortic pressure estimation method for a percutaneous ventricular assist device provided in another embodiment of this application.

[0035] Figure 7 A flowchart of an aortic pressure estimation method for a percutaneous ventricular assist device provided in another embodiment of this application.

[0036] Figure 8 A flowchart of an aortic pressure estimation method for a percutaneous ventricular assist device provided in another embodiment of this application.

[0037] Figure 9 A flowchart of an aortic pressure estimation method for a percutaneous ventricular assist device provided in another embodiment of this application.

[0038] Figure 10 A flowchart of an aortic pressure estimation method for a percutaneous ventricular assist device provided in another embodiment of this application.

[0039] Figure 11 A flowchart of an aortic pressure estimation method for a percutaneous ventricular assist device provided in another embodiment of this application.

[0040] Figure 12 This is a schematic diagram of the aortic pressure estimation device for a percutaneous ventricular assist device provided in one embodiment of this application.

[0041] The accompanying drawings illustrate specific embodiments of this application, which will be described in more detail below. These drawings and descriptions are not intended to limit the scope of the concept in any way, but rather to illustrate the concept of this application to those skilled in the art through reference to particular embodiments. Detailed Implementation

[0042] The present invention will now be described in further detail with reference to specific embodiments and accompanying drawings. Similar elements in different embodiments are referred to by associated similar element reference numerals. In the following embodiments, many details are described to facilitate a better understanding of the invention. However, those skilled in the art will readily recognize that some features may be omitted in different situations, or may be replaced by other elements, materials, or methods. In some cases, certain operations related to the present invention are not shown or described in the specification. This is to avoid obscuring the core parts of the invention with excessive description. For those skilled in the art, detailed description of these related operations is not necessary; they can fully understand the related operations based on the description in the specification and general technical knowledge in the art.

[0043] Furthermore, the features, operations, or characteristics described in the specification can be combined in any suitable manner to form various embodiments. At the same time, the steps or actions in the method description can be rearranged or adjusted in a manner obvious to those skilled in the art. Therefore, the various orders in the specification and drawings are only for the clear description of a particular embodiment and do not imply a necessary order, unless otherwise stated that a particular order must be followed.

[0044] The terms "first," "second," etc., used in the specification and claims of this application are used to distinguish similar objects and not to describe a specific order or sequence. It should be understood that such use of data can be interchanged where appropriate so that embodiments of this application can be implemented in orders other than those illustrated or described herein, and the objects distinguished by "first," "second," etc., are generally of the same class, without limiting the number of objects; for example, a first object can be one or more. Furthermore, in the specification and claims, "and / or" indicates at least one of the connected objects, and the character " / " generally indicates that the preceding and following objects are in an "or" relationship. Unless otherwise specified, the terms "connection" and "linkage" used in this application include both direct and indirect connections (linkages).

[0045] The technical solution of this application and how the technical solution of this application solves the above-mentioned technical problems are described in detail below with specific embodiments. These specific embodiments can be combined with each other, and the same or similar concepts or processes may not be described again in some embodiments. The embodiments of this application will now be described with reference to the accompanying drawings.

[0046] Figure 1 This document provides a schematic diagram of a percutaneous ventricular assist device according to one embodiment of this application. Figure 1 As shown, the percutaneous ventricular assist device provided in this embodiment includes at least a blood-side pump chamber 101, a gas-side membrane chamber 102, a pneumatic diaphragm 103, a gas path and pressure regulating component 104, a gas-side pressure sensor 105, and a controller 106.

[0047] In this embodiment, the percutaneous ventricular assist device is a pneumatic diaphragm-type percutaneous ventricular assist device used to provide temporary or short-term circulatory support for patients with heart failure. The device delivers blood from the inlet side to the aortic side via interventional means, assisting the heart in maintaining systemic perfusion. Its core working principle is: the gas path and pressure regulating component 104 periodically provide positive and negative pressure to the gas-side diaphragm chamber 102, driving the pneumatic diaphragm 103 to reciprocate, thereby alternately changing the volume of the blood-side pump chamber 101, realizing the intake and pumping of blood.

[0048] During the operation of percutaneous ventricular assist devices (PVCs), aortic pressure is a crucial physiological parameter reflecting the patient's circulatory perfusion status and output load. Changes in aortic pressure directly affect the pumping process, single pump volume, and actual assist flow rate of the membrane pump. Therefore, obtaining an estimated aortic pressure value is essential for evaluating the assist effect, determining the output load, estimating flow rate, adjusting operating parameters, and displaying pressure trends. In existing technologies, aortic pressure is typically measured using an invasive arterial pressure catheter or a pressure sensor placed in the blood pathway. While these methods provide a reference pressure, they require the addition of a dedicated blood-side pressure sensor or pressure tapping structure in the blood pathway, increasing the cost of blood contact components, sealing interfaces, electrical connections, sterilization requirements, and disposable consumables. Furthermore, they may introduce issues related to thrombosis, zero-point drift, installation location, and long-term reliability. For the pneumatic diaphragm type 103 PVC, although a pressure sensor is provided on the gas side, this gas-side pressure signal is a composite signal influenced by multiple factors such as gas source pressure, valve status, gas flow, diaphragm position, and blood-side load, making it difficult to directly estimate aortic pressure.

[0049] This embodiment addresses the aforementioned issues by using existing gas-side pressure signals to stably and accurately obtain aortic pressure estimation results without adding a dedicated blood-side pressure sensor. This reduces the cost of blood contact components and the risk of complications such as thrombosis, providing a reliable basis for evaluating auxiliary effects, determining output load, estimating flow rate, and adjusting operating parameters.

[0050] In this embodiment, the blood-side pump chamber 101 is a cavity that directly contacts blood, with its two ends connected to the blood inlet side and the aortic side, respectively. The volume of the blood-side pump chamber 101 changes with the movement of the pneumatic diaphragm 103: when the pneumatic diaphragm 103 moves towards the blood side, the volume of the blood-side pump chamber 101 decreases, and the blood in the pump chamber is pushed to the aortic side; when the pneumatic diaphragm 103 moves towards the gas side, the volume of the blood-side pump chamber 101 increases, and blood is drawn into the pump chamber from the inlet side. The specific shape and volume of the blood-side pump chamber 101 can be designed according to the auxiliary flow requirements. Its inlet side and aortic side are usually equipped with one-way valves or equivalent structures to ensure unidirectional blood flow.

[0051] The gas-side diaphragm chamber 102 is used to contain the gas that drives the pneumatic diaphragm 103, typically air or an inert gas. Changes in gas pressure within the gas-side diaphragm chamber 102 directly drive the pneumatic diaphragm 103 to shift, thereby altering the volume of the blood-side pump chamber 101. The gas-side diaphragm chamber 102 is connected to a pressure regulating component via a gas path, and its internal pressure is regulated by the gas path and the pressure regulating component 104.

[0052] A pneumatic diaphragm 103 is disposed between the blood-side pump chamber 101 and the gas-side diaphragm chamber 102 to completely isolate blood from gas. Driven by gas-side pressure within the gas-side diaphragm chamber 102, it periodically changes the volume of the blood-side pump chamber 101. The pneumatic diaphragm 103 is made of a flexible material with good elasticity and biocompatibility, capable of reversible elastic deformation under gas-side pressure. When the pneumatic diaphragm 103 is subjected to a certain preload pressure, the diaphragm is in a tensile state. At this time, the diaphragm has a relatively stable effective area and local equivalent stiffness, which is beneficial for the stable transmission of aortic pressure fluctuations to the gas side via the diaphragm.

[0053] The gas path and pressure regulating component 104 is connected to the gas-side diaphragm chamber 102, and is used to provide gas-side pressure to the gas-side diaphragm chamber 102 and regulate the gas-side pressure within the gas-side diaphragm chamber 102. The gas path and pressure regulating component 104 may include a gas source, a pressure regulating valve, a control valve, and pipelines. The gas source provides positive or negative pressure gas; the pressure regulating valve regulates the pressure output to the gas-side diaphragm chamber 102; and the control valve controls the on / off state and flow direction of the gas path, thereby achieving precise control of the pressure in the gas-side diaphragm chamber 102 at different auxiliary stages. During normal auxiliary processes, the gas path and pressure regulating component 104 periodically switches between positive, negative, and holding states under the control of the controller 106, driving the pneumatic diaphragm 103 to reciprocate. When an effective pressure transmission state needs to be established, the controller 106 controls the gas path and pressure regulating component 104 to apply a controlled preload pressure to the gas-side diaphragm chamber 102, causing the pneumatic diaphragm 103 to enter a tensioned state and maintain stable gas-side pressure in this state, thereby establishing stable pressure transmission conditions.

[0054] A gas-side pressure sensor 105 is connected to a gas-side diaphragm 102 and is used to acquire gas-side pressure signals within the gas-side diaphragm 102. The gas-side pressure sensor 105 can be disposed on the wall of the gas-side diaphragm 102 or in a gas path connected to the gas-side diaphragm 102. The gas-side pressure sensor 105 converts pressure fluctuations within the gas-side diaphragm 102 into electrical signals and outputs them to the controller 106. This gas-side pressure signal includes both the pressure component generated by the gas path drive and the pressure component transmitted from the aorta to the gas side via the blood-side pump chamber 101 and the pneumatic diaphragm 103. Within the effective recognition window, the former is in a low-disturbance state, while the latter becomes the main fluctuation component in the gas-side pressure signal; therefore, the gas-side pressure signal can be used for aortic pressure estimation.

[0055] The controller 106 is connected to the gas-side pressure sensor 105 and the gas path and pressure regulating component 104. During routine assistance, the controller 106 controls the gas path and pressure regulating component 104 to periodically adjust the gas-side pressure in the gas-side diaphragm cavity 102 according to preset assistance modes and control parameters, driving the pneumatic diaphragm 103 to reciprocate and achieve blood pumping. Specifically, the controller 106 controls the gas path and pressure regulating component 104 to periodically apply positive and negative pressure to the gas-side diaphragm cavity 102, causing the pneumatic diaphragm 103 to reciprocate. During the positive pressure phase, the pneumatic diaphragm 103 moves towards the blood side, the volume of the blood-side pump cavity 101 decreases, and blood is pumped to the aortic side; during the negative pressure phase, the pneumatic diaphragm 103 moves towards the gas side, the volume of the blood-side pump cavity 101 increases, and blood is drawn into the blood-side pump cavity 101 from the inlet side. Through the above periodic process, the device continuously provides circulatory support to the patient.

[0056] The special feature of this embodiment is that, when aortic pressure estimation is required, the controller 106, based on the conventional control described above, performs aortic pressure estimation based on the gas-side pressure signal collected by the existing gas-side pressure sensor 105. Specifically, the controller 106 acquires the gas-side pressure signal collected by the gas-side pressure sensor 105, and determines whether the pneumatic diaphragm 103 is in an effective pressure transmission state based on the gas-side pressure signal; when the pneumatic diaphragm 103 is in an effective pressure transmission state, it determines an effective identification window based on the gas-side pressure signal and the pneumatic operation state of the percutaneous ventricular assist device; it acquires the gas-side dynamic pressure sequence within the effective identification window; and based on a pre-established multi-condition aortic pressure back-calculation model, it processes the gas-side dynamic pressure sequence to obtain the aortic pressure estimation result.

[0057] Understandably, when aortic pressure estimation is required, the controller 106, while ensuring safe operation of the equipment, switches to the aortic pressure estimation process. The controller 106 controls the air path and pressure regulating component 104 to apply controlled preload pressure to the gas-side diaphragm 102 in a gradually increasing manner, causing the pneumatic diaphragm 103 to enter a tensioned state. Once the gas-side pressure signal meets the preset stability conditions, the controller 106 determines an effective identification window based on the gas-side pressure signal and the pneumatic operating status. Within this window, it acquires the gas-side dynamic pressure sequence and processes it using a pre-established multi-condition aortic pressure back-calculation model to obtain the aortic pressure estimation result. After estimation, the controller 106 returns to the normal auxiliary mode and continues to drive the pneumatic diaphragm 103 according to the original auxiliary parameters, ensuring that the auxiliary process is not unnecessarily interrupted by the estimation operation.

[0058] It should be noted that the percutaneous ventricular assist device provided in this embodiment may also include other conventional functional modules / units such as a power module, display module, communication module, alarm module, and storage module. Their specific configuration and connection relationships are common knowledge in the art and will not be elaborated here. The innovation of this application lies in the aortic pressure estimation method executed by the controller 106 based on the gas-side pressure signal collected by the existing gas-side pressure sensor 105 and the operating status of the gas path and pressure regulating component 104, without adding a dedicated blood-side pressure sensor. The implementation process of this method will be further described in detail below with reference to the accompanying drawings.

[0059] Figure 2 This is a flowchart illustrating an aortic pressure estimation method for a percutaneous ventricular assist device according to one embodiment of this application. Figure 2 As shown, the aortic pressure estimation method for percutaneous ventricular assist devices provided in this embodiment specifically includes the following steps:

[0060] Step S201: Obtain the gas-side pressure signal collected by the gas-side pressure sensor.

[0061] During the operation of a percutaneous ventricular assist device (VCA), the state of the pneumatic diaphragm directly determines whether aortic pressure fluctuations can be stably transmitted to the gas side. The gas-side pressure signal is the primary source of information for judging the state of the pneumatic diaphragm and estimating aortic pressure. Specifically, the pneumatic diaphragm has different mechanical properties in different states such as relaxed, tense, wrinkled, or adhered to the wall. Whether it is in a suitable state for pressure transmission can be determined by the characteristics of the gas-side pressure signal. Simultaneously, after aortic pressure fluctuations are transmitted to the gas side via the blood-side pump chamber and the pneumatic diaphragm, a corresponding fluctuation component is formed in the gas-side pressure signal. This component is also important to consider during the inverse calculation of aortic pressure. The process of acquiring the gas-side pressure signal involves the controller controlling the gas-side pressure sensor connected to the gas-side diaphragm chamber to detect the gas-side pressure within the chamber in real time and outputting the gas-side pressure signal according to a preset sampling rate.

[0062] Step S202: Determine whether the pneumatic diaphragm is in an effective pressure transmission state based on the gas side pressure signal; wherein, when the pneumatic diaphragm is in an effective pressure transmission state, its response under pressure change satisfies the preset stability condition.

[0063] The pneumatic diaphragm is a crucial link in the transmission of blood-side pressure to the gas-side, and its mechanical state directly determines whether aortic pressure fluctuations can be stably and repeatedly transmitted to the gas-side pressure signal. Specifically, when the pneumatic diaphragm is in a relaxed, wrinkled, or adhered state, its effective area, local stiffness, and damping characteristics change significantly. This results in unstable amplitude and phase of minute pressure changes caused by aortic pulsations when transmitted to the gas side, making it difficult to establish a stable inverse calculation relationship even if pulsation fluctuations appear in the gas-side pressure signal. Conversely, when the pneumatic diaphragm is under appropriate gas-side preload and in a tensioned state, its local mechanical properties tend to stabilize, enabling a repeatable response to minute pressure changes caused by aortic pulsations. This allows aortic pressure fluctuations to be reflected more stably in the gas-side pressure signal. Therefore, before performing inverse calculations on the gas-side pressure signal, it is necessary to determine whether the pneumatic diaphragm is in this effective pressure transmission state suitable for pressure transmission based on the gas-side pressure signal.

[0064] Effective pressure transmission state refers to the state in which the response of the pneumatic diaphragm under pressure changes meets preset stability conditions. Specifically, in this state, the gas-side pressure signal exhibits only small fluctuations near the preload pressure, and the fluctuation pattern is consistent with the aortic pulsation cycle. The average pressure level of the gas-side pressure signal is within the expected range, and the pressure change rate and baseline drift are small, with the pulse fluctuation amplitude reaching a discernible level. Based on these characteristics of the gas-side pressure signal, it can be determined whether the pneumatic diaphragm has entered and maintained an effective pressure transmission state. When the gas-side pressure signal indicates that the pneumatic diaphragm meets the above stability conditions, it can be considered that the pneumatic diaphragm is in an effective pressure transmission state, thus providing stable pressure transmission boundary conditions for aortic pressure estimation and ensuring the reliability and repeatability of subsequent estimation results.

[0065] Step S203: When the pneumatic diaphragm is in an effective pressure transmission state, determine the effective identification window for aortic pressure estimation based on the gas-side pressure signal and the pneumatic operating state of the percutaneous ventricular assist device. The effective identification window is the time interval corresponding to when the gas-side pressure signal meets a preset pressure signal quality condition and the pneumatic operating state of the percutaneous ventricular assist device meets a preset operating state condition.

[0066] In practice, even when the pneumatic diaphragm is in an effective pressure transmission state, the gas-side pressure signal may still contain additional components generated by pneumatic disturbances such as pressure regulating valve operation, gas source fluctuations, gas flow changes, and motor pressure regulation. These disturbance components are unrelated to aortic pulsation but are superimposed on the gas-side pressure signal, reducing the discernibility of the aortic pulsation component. If the gas-side pressure signal under the entire effective pressure transmission state is used indiscriminately, these disturbances may be introduced into the back-calculation process, leading to deviations in the aortic pressure estimation results. Therefore, this embodiment further identifies a time interval with lower pneumatic disturbances and clearer aortic pulsation components when the pneumatic diaphragm is in an effective pressure transmission state, i.e., an effective identification window, for acquiring high-quality gas-side dynamic pressure sequences.

[0067] In this embodiment, determining the effective identification window requires combining information from both the gas-side pressure signal and the pneumatic operating status of the percutaneous ventricular assist device. Firstly, from the perspective of the gas-side pressure signal, it is determined whether it meets preset pressure signal quality conditions. These conditions measure whether the aortic pulsation component in the gas-side pressure signal is clear and stable, such as whether the baseline drift is sufficiently small, whether the pulse amplitude is identifiable, whether the waveforms of multiple consecutive pulsation cycles have good consistency, and whether there are significant interference components in the signal. When the gas-side pressure signal meets the preset quality conditions, it indicates that the aortic pulsation component is prominent during that time period, and the gas-side pressure signal is suitable for pressure inversion. Secondly, from the perspective of pneumatic operating status, it is determined whether the percutaneous ventricular assist device is currently in a low-disturbance state. This determination is based on whether the pressure regulating component is in a closed, held, or low-disturbance state, whether the valve is in a stable state, whether the gas source is supplying gas stably, and whether the device is in a non-active pressure regulating phase. When the pneumatic operation meets the preset conditions, it indicates that the pneumatic drive has minimal interference with the gas-side pressure signal during that time period, and the fluctuations in the gas-side pressure signal mainly originate from aortic pulsation. When both the gas-side pressure signal quality and the pneumatic operation meet the corresponding conditions, the corresponding time interval is the effective identification window. Within this window, the aortic pulsation component in the gas-side pressure signal has a high proportion and minimal interference, providing high-quality basic data for aortic pressure estimation.

[0068] Step S204: Obtain the gas-side dynamic pressure sequence within the effective identification window.

[0069] It is known that the transmission of aortic pressure to gas-side pressure signals is not a simple instantaneous correspondence, but rather involves dynamic characteristics such as inertia, damping, and time lag. The same gas-side pressure level may correspond to different aortic pressures at different waveform stages. Relying solely on an isolated instantaneous pressure value within the effective identification window cannot fully reflect the dynamic process of gas-side pressure changes with aortic pulsation, nor can it accurately calculate aortic pressure. Therefore, by continuously acquiring gas-side pressure signals within the effective identification window to form a time-sequential dynamic gas-side pressure sequence, more complete waveform information can be obtained. This dynamic gas-side pressure sequence consists of multiple continuous sampling points and can reflect the complete dynamic process of gas-side pressure changes with aortic pulsation, including the rate of pressure rise and fall, the shape of peaks and troughs, and the repeatability of each cycle. The pressure waveform morphology, rate of change, and statistical characteristics contained in the dynamic gas-side pressure sequence provide richer dynamic information for calculating aortic pressure, improving the accuracy of aortic pressure estimation.

[0070] Step S205: Based on the pre-established multi-condition aortic pressure back-calculation model, determine the dynamic mapping relationship between the gas-side dynamic pressure sequence and the aortic pressure, and obtain the aortic pressure estimation result.

[0071] In this embodiment, a multi-condition aortic pressure inverse calculation model is pre-established. This model simultaneously acquires reference aortic pressure and gas-side pressure signals under multiple aortic pressure conditions, extracts historical sequences, differentials, and statistical features of the gas-side pressure signals, and establishes a dynamic mapping relationship between the gas-side dynamic pressure sequence and the aortic pressure. These multiple system conditions may include different heart rates, flow rates, system resistance, compliance, preload pressure, membrane pump types, and tubing parameters. When estimating the aortic pressure under the current condition, the gas-side dynamic pressure sequence within the effective identification window is input into the pre-established multi-condition aortic pressure inverse calculation model. Based on the historical pressure values, trends, and statistical features in the gas-side dynamic pressure sequence, the model determines the aortic pressure corresponding to the current gas-side dynamic pressure sequence, thereby obtaining an aortic pressure estimation result. This estimation result can be an instantaneous aortic pressure waveform, or it can be systolic pressure, diastolic pressure, mean pressure, pulse pressure, or pressure trend obtained from the waveform.

[0072] Therefore, the aortic pressure estimation method provided in this embodiment acquires the gas-side pressure signal and determines whether the pneumatic diaphragm is in an effective pressure transmission state, ensuring that the diaphragm operates under tension and meets stability conditions. This provides stable and repeatable boundary conditions for the transmission of aortic pressure to the gas side. Based on this, an effective identification window is determined by combining the gas-side pressure signal and the pneumatic operating state. The gas-side dynamic pressure sequence is collected within a time interval with low pneumatic disturbance and clear aortic pulsation components, effectively suppressing the influence of interference such as pressure regulating valve action and gas source fluctuations on the estimation results. Furthermore, the gas-side dynamic pressure sequence is processed using a pre-established multi-condition aortic pressure back-calculation model, which can adapt to changes in pressure transmission relationships under different heart rates, aortic pressures, system resistances, and compliance conditions. Thus, without adding a dedicated blood-side pressure sensor, the aortic pressure estimation results are obtained stably and accurately, providing a reliable basis for auxiliary effect evaluation, output load judgment, flow estimation, and operating parameter adjustment.

[0073] Figure 3 A flowchart illustrating an aortic pressure estimation method for a percutaneous ventricular assist device, provided as another embodiment of this application. Figure 3 As shown, in some embodiments, step S202 in the above embodiments, determining whether the pneumatic diaphragm is in an effective pressure transmission state based on the gas-side pressure signal, specifically includes the following steps:

[0074] Step S301: Analyze the gas-side pressure signal to obtain the corresponding pressure change characteristics. These pressure change characteristics include at least one of the following: gas-side pressure change rate, gas-side pressure change period, gas-side pressure fluctuation range, adjacent pressure change trends, and pressure response consistency across multiple periods.

[0075] The gas-side pressure signal is a continuous signal that varies over time. Directly observing this signal makes it difficult to objectively and quantitatively determine the mechanical state of the pneumatic diaphragm. Therefore, when analyzing the gas-side pressure signal, it is necessary to extract various pressure change characteristics that reflect the response state of the pneumatic diaphragm. Specifically, the gas-side pressure signal can be processed through filtering, differential analysis, period detection, and statistical calculations to obtain characteristics such as the rate of change of gas-side pressure, the period of change of gas-side pressure, the range of fluctuation of gas-side pressure, the trend of adjacent pressure changes, and the consistency of pressure response between multiple periods. These characteristics can then be used as a basis for judging the state of the pneumatic diaphragm.

[0076] Among them, the rate of change of gas-side pressure reflects the instantaneous fluctuation speed of gas-side pressure; the period of change of gas-side pressure reflects whether the rhythm of pressure fluctuation corresponds to the aortic pulsation period; the range of gas-side pressure fluctuation reflects the magnitude of pressure fluctuation; the trend of adjacent pressure changes reflects whether the pressure waveform transitions smoothly; and the consistency of pressure response between multiple cycles reflects whether the diaphragm maintains stable response characteristics under continuous pulsation. These characteristics together describe the response performance of the pneumatic diaphragm under the action of gas-side pressure.

[0077] Step S302: Determine the response state of the pneumatic diaphragm under the action of gas side pressure based on the pressure change characteristics.

[0078] It is known that the response of a pneumatic diaphragm to changes in gas-side pressure varies significantly under different states. For example, when the pneumatic diaphragm is in a relaxed or wrinkled state, its effective area and local stiffness are unstable, and the pulsating fluctuations in the gas-side pressure signal may exhibit characteristics such as unstable amplitude, unclear period, or large baseline drift. When the pneumatic diaphragm is in a tensile state and its mechanical properties are stable, the pulsating fluctuations in the gas-side pressure signal exhibit stable amplitude, clear period, and a stable baseline. Therefore, after obtaining the pressure change characteristics, we can further determine the current response state of the pneumatic diaphragm based on these characteristics. Specifically, we can compare each extracted feature with preset judgment conditions or thresholds, and combine the comprehensive performance of multiple features to determine the response state of the pneumatic diaphragm under gas-side pressure, such as whether it is in a stable tensile state, a relaxed state, or an unstable state.

[0079] Step S303: When the response state of the pneumatic diaphragm meets the preset stability condition, it is determined that the pneumatic diaphragm is in an effective pressure transmission state.

[0080] After obtaining the response state of the pneumatic diaphragm, it is further determined whether this state meets the preset stability conditions required for aortic pressure estimation. The preset stability conditions are criteria used to measure whether the pneumatic diaphragm has stable and repeatable pressure transmission characteristics in its current state. When the response state of the pneumatic diaphragm meets the preset stability conditions, it indicates that its tension is appropriate, its mechanical properties are stable, and aortic pressure fluctuations can be stably transmitted to the gas side through the diaphragm, thus making the pulsating component in the gas-side pressure signal clear and repeatable. If the response state of the pneumatic diaphragm does not meet the preset stability conditions, it indicates that the current diaphragm state is not suitable for pressure back-calculation, and it is necessary to continue adjusting the preload pressure or wait for the equipment operating state to naturally meet the conditions until they are met before proceeding with subsequent processing.

[0081] In some embodiments, the stability conditions of the pneumatic diaphragm's response state can be defined from two aspects: pressure stability and pulse wave quality. For example, the pressure stability of the gas-side pressure signal meets a first preset requirement, and / or, the pulse wave quality in the gas-side pressure signal that matches the heart rate or trigger cycle meets a second preset requirement. The first preset requirement includes that the gas-side average pressure is within the allowable range of the target preload pressure, the pressure change rate is less than a corresponding threshold, and the baseline drift is less than a corresponding threshold. The second preset requirement includes that the pulse amplitude reaches a preset amplitude threshold, the signal-to-noise ratio reaches a preset signal-to-noise ratio threshold, and the amplitude and morphology consistency of multiple consecutive cycles reaches a preset consistency threshold.

[0082] Specifically, the first preset requirement is used to evaluate the overall stability of the gas-side pressure, mainly reflecting whether the pneumatic diaphragm has stopped large-scale movement and tended to stabilize under gas-side preload. Specifically, the average gas-side pressure being within the allowable range of the target preload pressure indicates that the current gas-side pressure has basically reached the preload level required for the diaphragm to enter a tensile state; it is neither too low, leading to diaphragm relaxation, nor too high, leading to excessive stretching or adhesion to the wall. The pressure change rate being less than the corresponding threshold indicates that the gas-side pressure has not undergone rapid, active adjustment-induced changes during the observation period, and the diaphragm has not undergone large-scale stroke movement driven by the gas path. The baseline drift being less than the corresponding threshold indicates that the slow-varying component of the gas-side pressure remains within an acceptable range, eliminating the influence of slow interferences such as temperature changes, minor gas leaks, or sensor drift on the pressure baseline. When all of the above conditions are met, it indicates that the mechanical boundary conditions of the diaphragm corresponding to the gas-side pressure signal have basically stabilized, meeting the prerequisites for pressure back-calculation.

[0083] The second preset requirement is used to evaluate the identifiability of the aortic pulsation component in the gas-side pressure signal. It primarily reflects whether aortic pressure fluctuations have been effectively transmitted to the gas side through the diaphragm and form a clear pulse wave. Specifically, a pulse amplitude reaching a preset amplitude threshold indicates the presence of reliably identifiable pulsation fluctuations in the gas-side pressure signal, with a sufficiently large amplitude to avoid being overwhelmed by noise; a signal-to-noise ratio reaching a preset signal-to-noise ratio threshold indicates a sufficiently high proportion of the pulsation component relative to aerodynamic disturbances and sensor noise, ensuring the quality of the input signal for subsequent inverse calculation; and amplitude and morphological consistency across multiple consecutive cycles reaching a preset consistency threshold indicates that the diaphragm's response to aortic pulsations remains stable and repeatable across multiple cardiac cycles, without significant amplitude attenuation, waveform distortion, or cycle mismatch. When the second preset requirement is met, it indicates that the pulse wave component in the gas-side pressure signal possesses sufficient stability and clarity for inverse calculation of aortic pressure.

[0084] In practical implementation, the first preset requirement can be used for judgment alone, the second preset requirement can be used alone, or both can be used in combination. When the gas-side pressure signal does not meet the preset stability condition, it indicates that the pneumatic diaphragm has not yet entered or stabilized in an effective pressure transmission state. At this time, the preload pressure can be adjusted, for example, by gradually adjusting the gas-side pressure through the gas path and pressure regulating components until the pressure stability and / or pulse wave quality meet the requirements; or the equipment's operating status can be allowed to change naturally, allowing the pneumatic diaphragm to automatically enter the tension stage that meets the conditions during the current auxiliary process, until the conditions are met before proceeding with the subsequent determination of the effective identification window and aortic pressure back-calculation. This ensures the reliability of the effective pressure transmission state judgment and avoids invalid signal acquisition and back-calculation when the diaphragm's pressure transmission conditions are not yet met.

[0085] Figure 4 A flowchart illustrating an aortic pressure estimation method for a percutaneous ventricular assist device, as provided in another embodiment of this application. Figure 4 As shown, in some embodiments, the pressure change characteristics corresponding to the gas-side pressure signal in the above embodiments and the response state of the pneumatic diaphragm are obtained through an active pressure build-up process, which includes the following steps:

[0086] Step S401: Control the pressure regulating component to apply controlled preload pressure to the gas-side diaphragm cavity through the air passage, so that the pneumatic diaphragm enters the predetermined tension state.

[0087] During routine diaphragm operation, the pneumatic diaphragm undergoes periodic, large-stroke motion driven by the airflow, spending most of its time in a relaxed, wrinkled, or rapidly moving state. In these states, the effective area, local stiffness, and damping characteristics of the diaphragm are unstable, making it difficult to reliably transmit minute pressure changes caused by aortic pulsation to the gas side. Therefore, when it is necessary to estimate aortic pressure using gas-side pressure signals, it is essential to first switch the pneumatic diaphragm from its routine motion state to a relatively stable tensioned state.

[0088] During active pressure build-up, the pressure regulating component first applies a controlled preload pressure to the gas-side diaphragm cavity via the gas path. This preload pressure is intentionally controlled and applied gradually in a slow or step-by-step manner to avoid sudden pressure changes that could impact the diaphragm and blood side. Under the preload pressure, the pneumatic diaphragm is pushed to a defined equilibrium position, the diaphragm material is tensioned, the pleats are unfolded, and the diaphragm enters a predetermined tension state. Once in the tension state, the diaphragm possesses a relatively stable effective area and local equivalent stiffness, making its response to minute changes in blood-side pressure repeatable, thus ensuring a high-quality gas-side pressure response.

[0089] Step S402: After the pneumatic diaphragm is in a tensioned state, the pressure regulating component is sequentially controlled to apply multiple preset candidate gas side pressures to the gas side diaphragm cavity through the gas path.

[0090] It is understandable that the diaphragm enters a tensile state under a certain preload pressure, but different preload pressures will cause the diaphragm to be in different equilibrium positions. Under different equilibrium positions, the local tangential stiffness, effective area, gas-side volume, and the diaphragm's transmission gain to blood-side pressure fluctuations may all differ, thus affecting the pressure transmission quality. Using only a single preload pressure point may not achieve optimal pressure transmission. Therefore, after the pneumatic diaphragm has entered a tensile state, multiple preset candidate gas-side pressures are applied sequentially. These candidate gas-side pressures can be a set of discrete pressure values ​​or a continuous pressure sequence formed by stepwise adjustments. After each candidate gas-side pressure is applied, the pneumatic diaphragm reaches a new force equilibrium under that pressure, corresponding to a different diaphragm operating point. By traversing multiple candidate pressures, the gas-side pressure response of the diaphragm at multiple operating points can be obtained, facilitating the selection of the operating point with the optimal pressure transmission quality.

[0091] Step S403: Obtain the gas-side pressure response of the pneumatic diaphragm under tension state when the gas-side pressure is formed in each candidate gas.

[0092] Under each candidate gas-side pressure, the pneumatic diaphragm is in a corresponding tensile state. At this time, aortic pressure fluctuations are transmitted to the gas side through the blood-side pump chamber and the diaphragm, causing corresponding fluctuations in the gas-side pressure. This gas-side pressure fluctuation is the gas-side pressure response of the pneumatic diaphragm under that candidate pressure. However, the diaphragm's ability to transmit aortic pulsations differs under different candidate pressures, resulting in variations in pulsation amplitude, waveform morphology, signal-to-noise ratio, and periodicity in the gas-side pressure response. Therefore, under different tensile states created by various candidate gas-side pressures, a gas-side pressure sensor is used to collect pressure signals within the gas-side diaphragm chamber in real time, recording the corresponding gas-side pressure response signals at each candidate pressure point. This allows for an objective evaluation of the pressure transmission quality at each pressure point.

[0093] Step S404: Based on the gas-side pressure response, determine the pressure change characteristics and the response state of the pneumatic diaphragm.

[0094] After obtaining the gas-side pressure response under each candidate gas-side pressure, the response signal is analyzed to extract characteristic parameters reflecting the pressure transmission state of the diaphragm. Specifically, pressure change characteristics can be determined from the gas-side pressure response, such as the rate of change of gas-side pressure, the range of gas-side pressure fluctuations, the trend of adjacent pressure changes, periodic fluctuation characteristics, and the consistency of pressure response between multiple cycles. Simultaneously, based on these pressure change characteristics, the response state of the pneumatic diaphragm under the candidate pressure can be determined, i.e., whether the diaphragm is in a stable state of tensioned pressure transmission, and whether its response to aortic pulsation is clear, stable, and repeatable.

[0095] In step S404, the raw pressure responses collected under each candidate gas side pressure are converted into characteristic information and state conclusions that can be used to determine the effective pressure transmission state. If the pressure change characteristics and response state corresponding to a certain candidate pressure meet the preset stability conditions, then the candidate pressure can be determined as the effective pressure transmission operating point; if the current candidate pressure does not meet the conditions, the preload pressure can be adjusted or the equipment can be waited for to run naturally until the conditions are met. In this way, through the active pressure building process, the effective pressure transmission state of the pneumatic diaphragm can be systematically found and confirmed.

[0096] Figure 5 A flowchart illustrating an aortic pressure estimation method for a percutaneous ventricular assist device, as provided in another embodiment of this application. Figure 5 As shown, in some embodiments, the pressure change characteristics and response state of the pneumatic diaphragm corresponding to the gas-side pressure signal in the above embodiments can also be obtained through an operational status identification process. Unlike the active pressure build-up method, the operational status identification process does not actively change the current control parameters of the gas path and pressure regulating components. Instead, it identifies the time stage in the existing routine operation of the percutaneous ventricular assist device where the pneumatic diaphragm has met the tension and stability conditions, and uses the gas-side pressure signal corresponding to that stage to determine the pressure change characteristics and response state. The process of identifying the response state through operational status includes the following steps:

[0097] Step S501: Obtain the pneumatic operation status during the operation of the percutaneous ventricular assist device and determine the current operation stage.

[0098] During routine percutaneous ventricular assist devices (PCA), the gas path and pressure regulating components periodically adjust the gas-side pressure according to a preset control strategy, driving the pneumatic diaphragm to reciprocate. Each assist cycle can be divided into several different operating phases, such as a positive pressure drive phase, a negative pressure backflow phase, and a holding phase. The stress state of the pneumatic diaphragm, the on / off state of the gas path valves, and the stability of the gas-side pressure all differ in different operating phases. By acquiring the pneumatic operating status of the PCA device, such as the current state of the pressure regulating components and valves, the drive phase, the gas-side pressure level, and the duration of the current phase, the system can determine which operating phase of the assist cycle the device is currently in (e.g., the holding phase or the low-flow phase), and identify the conditions that currently satisfy the pneumatic diaphragm tension and low aerodynamic disturbance, thus identifying the effective pressure transmission state.

[0099] Step S502: Obtain the gas-side pressure signal corresponding to the current operating stage, and determine the force state, pressure stability and aerodynamic disturbance of the pneumatic diaphragm based on the gas-side pressure signal.

[0100] After determining the current operating phase, the gas-side pressure signal collected by the gas-side pressure sensor during this phase is acquired. Based on this signal, it is determined from three dimensions whether the current phase is suitable for aortic pressure estimation.

[0101] The stress state is used to evaluate whether the pneumatic diaphragm is under tension under the current gas-side pressure. This can be determined by whether the average gas-side pressure reaches a certain level and whether the gas-side pressure is sufficient to tension the diaphragm. If the gas-side pressure is too low, the diaphragm may be in a relaxed or wrinkled state; if the gas-side pressure is too high, the diaphragm may stick to the wall or reach its stroke limit. Only when the gas-side pressure is within a moderate range can the diaphragm be under tension. Pressure stability is used to evaluate whether the gas-side pressure has basically stabilized, rather than being in a process of active pressure regulation or rapid change. This can be measured by the gas-side pressure change rate and baseline drift. A large gas-side pressure change rate indicates that the gas path may be undergoing active pressure regulation or the diaphragm is moving; a significant baseline drift indicates the presence of slow disturbances such as temperature or leakage. When pressure stability meets the requirements, it means that the gas-side pressure has stabilized at a certain level and has stable transmission boundary conditions. Pneumatic disturbance is used to evaluate whether there are significant active disturbances on the gas path side at the current stage, such as pressure regulating valve operation, frequent valve switching, or gas source fluctuations. This can be determined by the valve status and drive status. If the pressure regulating component is closed or held, the valve does not operate frequently, and there is no obvious active pressure regulation in the gas path, then the aerodynamic disturbance is small, and the fluctuation in the gas side pressure signal mainly comes from the aortic pulsation.

[0102] Step S503: Determine the pressure change characteristics and pneumatic diaphragm response state corresponding to the current operating stage based on the stress state, pressure stability, and aerodynamic disturbance.

[0103] After obtaining information from three dimensions—stress state, pressure stability, and aerodynamic disturbance—an overall assessment of the current operating stage is made. Specifically, if the stress state indicates that the pneumatic diaphragm is under tension, the pressure stability indicates that the gas-side pressure is basically stable, and the aerodynamic disturbance indicates that there is no significant active interference on the gas path side in the current stage, then the gas-side pressure signal in this stage has good pressure transmission conditions. At this time, the corresponding pressure change characteristics can be extracted from the gas-side pressure signal, such as the gas-side pressure change rate, fluctuation range, periodic fluctuation characteristics, and consistency between multiple cycles, and it can be determined that the pneumatic diaphragm is in a stable tension response state in this stage. Conversely, if the pneumatic diaphragm is not under tension, the gas-side pressure is unstable, or the aerodynamic disturbance is significant, the pressure change characteristics corresponding to this stage will exhibit irregularity, unstable amplitude, or low signal-to-noise ratio, and the response state of the pneumatic diaphragm will not meet the requirements for effective pressure transmission. Through the above comprehensive assessment, the operating state identification process can determine the gas-side pressure signal characteristics and diaphragm response state that can be used for aortic pressure estimation from the normal operating process without generating additional disturbances to the auxiliary control.

[0104] Figure 6 A flowchart illustrating an aortic pressure estimation method for a percutaneous ventricular assist device, as provided in another embodiment of this application. Figure 6 As shown, in some embodiments, step S203 in the above embodiments, when the pneumatic diaphragm is in an effective pressure transmission state, determines the effective identification window for aortic pressure estimation based on the gas-side pressure signal and the pneumatic operation state of the percutaneous ventricular assist device, specifically including the following steps:

[0105] Step S601: Analyze the gas-side pressure signal to obtain the corresponding signal quality information.

[0106] Although the gas-side pressure signal already contains the fluctuation component caused by aortic pulsation, the clarity and stability of this component vary at different times. Even if the pneumatic diaphragm is in an effective pressure-transmitting state overall, the gas-side pressure signal may still be affected by residual disturbances, noise, or waveform instability in local time periods. Directly using signals from these time periods for inverse calculation may lead to fluctuations or deviations in the estimation results. Therefore, it is necessary to first analyze the gas-side pressure signal to obtain quality information that reflects whether the signal is suitable for pressure estimation, in order to assess whether it is clear and stable within the current time period and whether it meets the basic signal conditions for aortic pressure estimation.

[0107] Signal quality information can include at least one of the following: pressure change rate, pressure baseline variation, pressure signal stability, pulse amplitude, pressure signal-to-noise ratio (SNR), and pressure period consistency. Specifically, the pressure change rate measures how quickly the gas-side pressure changes over a short period; an excessively large rate of change indicates the presence of rapid components caused by active pressure regulation or diaphragm movement. The pressure baseline variation measures whether the pressure baseline is slowly drifting; excessive drift can affect the accurate extraction of the pulsating component. Pressure signal stability comprehensively evaluates whether the signal fluctuations within a certain window are smooth. Pulse amplitude measures whether the heart rate-related pulsating component in the signal is sufficiently strong. The pressure SNR measures the proportion of the pulsating component relative to noise and residual disturbances. Pressure period consistency measures whether the waveform of multiple consecutive pulsating cycles has good repeatability. These pieces of information collectively reflect the performance of the gas-side pressure signal.

[0108] Step S602: Obtain the pneumatic operating status information corresponding to the percutaneous ventricular assist device.

[0109] The quality of gas-side pressure signals depends not only on the signal itself but also on the direct impact of the equipment's gas path operating status. Pressure regulation actions, valve switching, and changes in drive cycles on the gas path side all introduce disturbance components into the gas-side pressure signal, reducing the discernibility of the aortic pulsation component. Therefore, signal quality information alone is insufficient to fully determine whether the current time period is suitable for pressure estimation; understanding the current operating status of the equipment's gas path is also necessary.

[0110] The pneumatic operating status information includes at least one of the following: gas-side driving pressure status, gas path switching status, driving cycle status, and percutaneous ventricular assist device operating mode. Specifically, the gas-side driving pressure status indicates whether the gas path is actively regulating or maintaining pressure; the gas path switching status indicates whether the relevant valves are in the process of closing, maintaining, or switching; the driving cycle status indicates which stage of the auxiliary cycle the device is currently in, such as the positive pressure driving stage, the negative pressure backflow stage, or the holding stage; and the operating mode indicates whether the device is in a normal auxiliary mode or another special mode. This information allows for the determination of whether there is active disturbance on the gas path side and whether the current stage is a low-disturbance phase.

[0111] Step S603: Determine the time interval that meets the aortic pressure estimation conditions based on the signal quality information and pneumatic operation status information.

[0112] After obtaining signal quality information and pneumatic operating status information separately, the two are combined for comprehensive judgment. Specifically, the signal quality information is used to determine whether the gas-side pressure signal meets the preset pressure signal quality conditions, such as whether the pulse amplitude is sufficient, whether the signal-to-noise ratio meets the standard, whether the period consistency is good, and whether the baseline change is within the allowable range. The pneumatic operating status information is used to determine whether the equipment is in a low-disturbance state, such as whether the pressure regulating components are in a closed or held state, whether the valves are not frequently switched, and whether the drive cycle is in a stable phase. When both signal quality information and pneumatic operating status information simultaneously meet the corresponding conditions, the corresponding continuous time interval is the time interval that meets the aortic pressure estimation conditions. Within this interval, the aortic pulsation component in the gas-side pressure signal has a high proportion, and the aerodynamic disturbance is small, making it an ideal time period for pressure back-calculation.

[0113] Step S604: Determine the time interval that meets the aortic pressure estimation conditions as the effective identification window.

[0114] After determining the time interval that meets the aortic pressure estimation conditions, this time interval is formally designated as an effective identification window for subsequent gas-side dynamic pressure sequence acquisition and aortic pressure inverse calculation. This transforms the results of the aforementioned condition judgment into a clear time boundary, avoiding the introduction of low-quality signals or high-perturbation signals outside the window into the inverse calculation process.

[0115] In some embodiments, the effective identification window satisfies at least one of the following conditions: the average pressure of the gas-side pressure signal is within the allowable range of the candidate preload pressure; the pressure change rate and baseline drift are both less than the corresponding thresholds; the pressure regulating component is in a closed, held, or low-disturbance state; there is a periodic fluctuation in the gas-side pressure signal that matches the heart rate or trigger cycle, and the amplitude and morphology consistency of multiple consecutive cycles reaches the threshold; and the gas-side pressure signal does not reach the saturation range of the gas-side pressure sensor, and the pneumatic diaphragm is not in a wall-attached or stroke-limited state.

[0116] Specifically, the average pressure of the gas-side pressure signal is within the allowable range of the candidate preload pressure to ensure that the gas-side pressure is at a reasonable preload level within the current time interval. If the average gas-side pressure is too low, the pneumatic diaphragm may not be sufficiently stretched and may remain in a relaxed or wrinkled state, making it difficult to stably transmit aortic pulsations. If the average gas-side pressure is too high, the diaphragm may be overstretched or even adhere to the wall, which is also detrimental to pressure transmission. An average pressure within the allowable range of the candidate preload pressure indicates that the diaphragm has an appropriate degree of tension under the current gas-side pressure, providing a stable mechanical basis for the transmission of aortic pulsations to the gas side.

[0117] The rate of change of pressure reflects the speed of change of gas-side pressure over a short period of time, while baseline drift reflects the slow shift of the pressure baseline over a longer timescale. If the rate of change of pressure is too large, it indicates that the gas path may still be actively regulating pressure or the diaphragm may still be moving; in this case, the pulsating component in the gas-side pressure signal will be masked by the rapidly changing pressure component. If the baseline drift is too large, it indicates the presence of slow instability factors in the gas-side pressure, such as temperature changes or minor gas leaks, which will affect the accurate extraction of the pulsating component. Both values ​​being less than their corresponding thresholds indicate that the gas-side pressure remains stable over time, and the fluctuations mainly originate from the aortic pulsation itself.

[0118] Even if the gas-side pressure signal itself appears stable, new pressure disturbances may still occur on the gas path side if the pressure regulating component continues to operate frequently or the valve continues to switch. These disturbances will be superimposed on the gas-side pressure signal, disrupting the integrity and consistency of the pulsation waveform. When the pressure regulating component is in a closed, held, or low-disturbance state, it means that there is no active pressure regulation action on the gas path side. The fluctuation component of the gas-side pressure signal mainly originates from the pressure changes transmitted from the blood side through the diaphragm, thus ensuring the singularity of the signal source.

[0119] If the gas-side pressure signal exhibits periodic fluctuations that match the heart rate or trigger cycle, and the amplitude and morphology of multiple consecutive cycles are consistent, it confirms the presence of an aortic pulsation component within the gas-side pressure signal, rather than other aerodynamic disturbances or noise. Aortic pulsations are periodic and synchronized with the heart rate; therefore, the gas-side pressure signal should contain pressure fluctuations of the same rhythm. Furthermore, the amplitude and morphology of multiple consecutive pulsation cycles should show good consistency, indicating that the diaphragm's response to each pulsation is stable and repeatable. If the waveform amplitude fluctuates wildly or the morphology changes periodically, it indicates that the diaphragm's pressure transmission characteristics are not yet stable, and using this signal for inverse calculations will introduce significant errors.

[0120] Gas-side pressure sensors have a limited measurement range. If the gas-side pressure approaches or reaches the sensor's saturation range, the sensor output will not accurately reflect the actual pressure change, resulting in signal distortion and rendering it unusable for pressure calculation. Simultaneously, if the pneumatic diaphragm is in a state of contact with the wall or reaches its travel limit, its degrees of freedom are restricted, preventing it from generating a normal elastic response to aortic pulsation and disrupting the pressure transmission relationship. Therefore, ensuring that the gas-side pressure signal is not saturated and that the diaphragm is not in contact with the wall or at its limit is a fundamental prerequisite for guaranteeing signal accuracy and the diaphragm's normal responsiveness.

[0121] The above conditions ensure the quality of the effective identification window from different dimensions: first, by ensuring the stability of the pressure baseline based on the gas-side pressure itself; second, by eliminating interference from active disturbances based on the gas pathway; third, by confirming the clarity and stability of the aortic pulsation component based on the signal content; and fourth, by eliminating signal distortion and limited diaphragm response based on the physical state of the sensor and diaphragm. Each condition can be used individually or in combination. When at least one condition is met, the time interval can be determined as the effective identification window for subsequent acquisition of gas-side dynamic pressure sequences and inverse calculation of aortic pressure.

[0122] Figure 7 A flowchart illustrating an aortic pressure estimation method for a percutaneous ventricular assist device, as provided in another embodiment of this application. Figure 7 As shown, in some embodiments, step S204 of the above embodiments, obtaining the gas-side dynamic pressure sequence within the effective identification window, specifically includes the following steps:

[0123] Step S701: Within the effective identification window, collect gas-side pressure signals for no less than a predetermined number of heartbeat cycles, and simultaneously acquire the heart rate or trigger cycle of the percutaneous ventricular assist device.

[0124] Understandably, the gas-side pressure signal from a single cardiac cycle contains limited information and is susceptible to interference from random factors, such as an irregular beat, brief signal fluctuations, or local noise spikes. If inverse calculations are performed based on only one or a very small number of cycles, the estimation results are easily and significantly affected by these random factors, resulting in insufficient reliability and stability. Therefore, by acquiring signals from multiple consecutive cardiac cycles and utilizing the repeatability between cycles to suppress random interference, aortic pressure estimation based on a sufficient number of complete cardiac cycles can be performed to improve accuracy.

[0125] The predetermined number of cardiac cycles can be set according to actual needs, such as at least three, five, or more complete cardiac cycles. During the acquisition process, the heart rate or trigger cycle of the percutaneous ventricular assist device is acquired synchronously to determine the start and end times of the heartbeat. The heart rate can be sourced from existing ECG signals, trigger signals, or duty cycle signals from the device. Synchronizing the gas-side pressure signal with the heart rate or trigger cycle ensures that the pressure data within each cardiac cycle is correctly segmented, avoiding the impact of inaccurate cycle boundaries on feature extraction and inverse calculation results.

[0126] Step S702: The gas-side pressure signals within a predetermined number of cardiac cycles are arranged in chronological order to form a gas-side dynamic pressure sequence, and the gas-side dynamic pressure sequence is processed to obtain input data for the multi-condition aortic pressure back-calculation model.

[0127] Although the raw gas-side pressure signals acquired within the effective identification window have passed the screening process, they may still contain local signal defects, such as high-frequency noise, occasional abnormal sampling points, slow baseline shifts, or baseline differences between different cycles. If these defects are directly input into the inverse calculation model without processing, they may interfere with the model's identification of the aortic pulsation component, reducing estimation accuracy. Furthermore, the numerical range of the raw gas-side pressure signals may vary with different preload pressures and operating conditions; direct input into the model may cause inconsistencies in characteristic scales. Therefore, necessary data processing of the acquired raw signals is required.

[0128] Data processing includes at least one of the following: pressure signal filtering, outlier removal, period segmentation, baseline processing, and data normalization. Pressure signal filtering removes high-frequency noise and irrelevant interference components from the gas-side pressure signal, retaining the effective fluctuation components related to aortic pulsation. Filtering methods can include low-pass filtering, band-pass filtering, or other processing methods suitable for physiological pressure signals. Outlier removal detects and removes abnormal sampling points that significantly deviate from the normal range, such as spikes or breaks caused by transient sensor interference or signal transmission errors, preventing these outliers from affecting subsequent statistical feature calculations and model input. Period segmentation divides the gas-side dynamic pressure sequence into several corresponding pulsation period segments based on heart rate or trigger cycle, ensuring each period segment corresponds to a complete heartbeat process, facilitating feature extraction and waveform analysis by cycle. Baseline processing eliminates slow baseline drift in the gas-side pressure signal, ensuring the pressure waveforms of each cycle are at a consistent baseline level, improving consistency between cycles and the comparability of features. Data normalization maps gas-side pressure data to a uniform numerical range, eliminating differences in values ​​caused by variations in preload pressure levels or operating conditions, and giving the input features of the back-calculation model a uniform scale.

[0129] Through the above data processing, the gas-side dynamic pressure sequence is converted into input data that can be used in multi-condition aortic pressure inverse calculation models. This input data, while preserving the dynamic information of aortic pulsation, reduces noise, outliers, and baseline shifts, and makes data from different conditions comparable, thus ensuring improved accuracy and robustness of aortic pressure estimation.

[0130] Figure 8 A flowchart illustrating an aortic pressure estimation method for a percutaneous ventricular assist device, as provided in another embodiment of this application. Figure 8 As shown, in some embodiments, step S205 in the above embodiments, which involves determining the dynamic mapping relationship between the gas-side dynamic pressure sequence and the aortic pressure based on a pre-established multi-condition aortic pressure back-calculation model, and obtaining the aortic pressure estimation result, specifically includes the following steps:

[0131] Step S801: Construct historical sampling values, first-order difference sequences, and window statistical features based on the gas-side dynamic pressure sequence.

[0132] The gas-side dynamic pressure sequence reflects the continuous change of gas-side pressure over time, containing waveform information of the aortic pulsation transmitted through the diaphragm. In this embodiment, three types of features are constructed based on the gas-side dynamic pressure sequence. Historical sampled values ​​are used to retain the complete waveform information of the gas-side pressure over a recent period, enabling the model to perceive the current waveform stage and historical pressure trajectory. The first-order difference sequence is used to reflect the rate and direction of change of the gas-side pressure, distinguishing between pressure rise and fall segments, and compensating for the inability of a single sampled value to reflect the dynamic changes of the waveform. Window statistical features are used to characterize the overall properties of the pressure signal within the entire window, such as mean, standard deviation, minimum, maximum, and window span, enabling the model to understand the position of the pressure baseline, the magnitude of fluctuations, and the overall dispersion of the signal.

[0133] Step S802: Select the corresponding heart rate partition parameters in the multi-condition aortic pressure back-calculation model according to the current heart rate or trigger cycle.

[0134] The transmission characteristics of aortic pressure to gas-side pressure signals are not the same across all heart rates. As heart rate changes, the period and waveform morphology of the aortic pulsation alter, and the time scale and phase relationship of the corresponding pulsation component in the gas-side pressure signal also change. For example, at faster heart rates, the pulsation period shortens, gas-side pressure fluctuations become more frequent, the diaphragm's recovery time between adjacent pulsations decreases, and the pressure transmission gain and phase lag may differ compared to slower heart rates. Therefore, using a single-parameter inverse model makes it difficult to maintain consistent estimation accuracy across a wide range of heart rates.

[0135] When the multi-condition aortic pressure inverse calculation model was established offline, corresponding model parameters were trained separately for different heart rate zones. Each heart rate zone has independent feature-standardized parameters and regression parameters to adapt to the characteristics of pressure transmission relationships within that heart rate range. In this embodiment, the current heart rate or trigger cycle is obtained, its corresponding heart rate zone is determined, and the corresponding partition parameters are selected from the multi-condition aortic pressure inverse calculation model. In this way, the model parameters are matched with the current physiological rhythm, which can more accurately reflect the mapping relationship between the gas-side dynamic pressure sequence and aortic pressure under the current heart rate condition, improving the specificity and accuracy of the estimation.

[0136] Step S803: Standardize the historical sampled values, first-order difference sequence and window statistical features using heart rate zoning parameters, and input the standardized features into the multi-condition aortic pressure back-calculation model to calculate the aortic pressure estimate.

[0137] While the historical sampling values, first-order difference sequences, and window statistical features constructed in step S801 already reflect the dynamic information of the gas-side pressure signal, there may be significant differences in the dimensions and numerical ranges between different features. For example, the magnitude of the pressure value may be much larger than that of the first-order difference sequence, and the standard deviation and span in the window statistical features may not be on the same scale as the absolute value of the pressure value. If the unprocessed features are directly input into the model, features with larger numerical ranges may dominate the model output, causing the model to be insensitive to features with smaller numerical ranges, thus affecting the estimation accuracy. To solve the above problems, this embodiment uses the heart rate zone parameters selected in step S802 to standardize the features. The purpose of standardization is to map each feature to a uniform numerical range, eliminate differences in dimensions and magnitudes, and make the contribution of each feature to the model output more balanced. The parameters used in the standardization process (such as the mean and standard deviation of each feature) are predetermined for the corresponding heart rate zone during the offline calibration stage, reflecting the statistical distribution characteristics of the features within that zone.

[0138] After standardization, the processed features are input into a multi-condition aortic pressure inverse calculation model. Based on the mapping relationships pre-trained using multi-condition data, the model calculates the input features and outputs corresponding aortic pressure estimates. These estimates may include instantaneous aortic pressure, aortic pressure variation trends, and at least one of the following: systolic pressure, diastolic pressure, mean pressure, and pulse pressure, obtained from the instantaneous aortic pressure waveform. Instantaneous aortic pressure provides a pressure estimate at each sampling point or time interval, reflecting the continuous change of aortic pressure over time. Systolic and diastolic pressures correspond to the maximum and minimum aortic pressure values ​​within a cardiac cycle, respectively. Mean pressure reflects the average perfusion pressure level within a cycle. Pulse pressure reflects the difference between systolic and diastolic pressure and is related to stroke volume and vascular compliance. These outputs provide multi-dimensional pressure information for evaluating auxiliary effects, determining output load, estimating flow rate, and adjusting operating parameters.

[0139] Figure 9 A flowchart illustrating an aortic pressure estimation method for a percutaneous ventricular assist device, as provided in another embodiment of this application. Figure 9 As shown in some embodiments, the multi-condition aortic pressure back-calculation model used in the above embodiments for aortic pressure estimation can be obtained through offline calibration using the following steps:

[0140] Step S901: Under multiple aortic pressure conditions and multiple system conditions, simultaneously acquire the reference aortic pressure and the gas-side pressure signal from the gas-side pressure sensor.

[0141] The transmission relationship between aortic pressure and gas-side pressure signals is not constant. When physiological conditions such as aortic pressure level, heart rate, pulse pressure, peripheral vascular resistance, and vascular compliance change, the response gain and phase of the pneumatic diaphragm to blood-side pressure fluctuations will also change accordingly. Similarly, equipment-side factors, such as the diaphragm operating point under different preload pressures, the structural parameters of the membrane pump, the gas-side volume, and the length and resistance of the tubing, will also affect the pressure transmission characteristics. If a calibration model is established under a single operating condition, the mapping relationship learned by the model is only valid for states that are the same as or similar to the calibration conditions. Once the actual usage conditions deviate from the calibration conditions, the estimation error may increase significantly, or even cause the model to fail.

[0142] Therefore, it is necessary to acquire calibration data under multiple operating conditions so that the model can learn the changing patterns of the mapping relationship between gas-side pressure signals and aortic pressure under different conditions. In this embodiment, data acquisition is performed under multiple aortic pressure conditions and multiple system conditions. Multiple aortic pressure conditions refer to simulating or setting different aortic pressure levels, such as different combinations of systolic pressure, diastolic pressure, mean pressure, and pulse pressure, to cover different output load states from hypotension to hypertension. Multiple system conditions refer to changing the relevant factors affecting the pressure transmission relationship, specifically including at least one of heart rate, flow rate, system resistance, compliance, preload pressure, membrane pump type, and tubing parameters. Among them, heart rate affects the pulse cycle and waveform time scale, flow rate affects the pressure drop characteristics of the blood-side pathway, system resistance and compliance reflect the load characteristics of the peripheral circulation, preload pressure determines the operating point of the diaphragm, and membrane pump type and tubing parameters reflect the structural differences on the device side.

[0143] Under each calibration condition, reference aortic pressure and gas-side pressure signals from the gas-side pressure sensor are acquired simultaneously. The reference aortic pressure can be obtained from a reference pressure sensor installed in the test equipment or simulated circulatory system, providing a true reference for aortic pressure. The purpose of simultaneous acquisition is to ensure a one-to-one temporal correspondence between the two sets of data, enabling subsequent model training to accurately match the characteristics of the gas-side pressure signal with the actual aortic pressure at the same time.

[0144] Step S902: Mark the time period when the pneumatic diaphragm is in an effective pressure transmission state under each working condition, and construct the historical sequence, differential and statistical characteristics of the gas side pressure signal.

[0145] In the calibration data collected in step S901, not all gas-side pressure signals over all time periods are suitable for establishing the aortic pressure inverse relationship. As mentioned earlier, only when the pneumatic diaphragm is in an effective pressure transmission state can the aortic pulsation be stably transmitted to the gas side, and only then will the pulsation component in the gas-side pressure signal have clear and repeatable characteristics. Therefore, this embodiment first marks the calibration data under each operating condition, identifying and marking the time periods when the pneumatic diaphragm is in an effective pressure transmission state. The marking criteria are consistent with the aforementioned conditions for online judgment of the effective pressure transmission state, such as whether the average gas-side pressure is within the allowable range of preload pressure, whether the pressure change rate and baseline drift meet the stability requirements, and whether the pulse wave quality reaches a discernible level. By marking, the model training data is limited to the time period that coincides with the physical state and the online effective identification window, ensuring that the mapping relationship learned by the model matches the signal conditions during online application. After labeling is completed, the historical sequence, differential and statistical features of the gas-side pressure signal are constructed within the effective pressure transmission time period of the labeling. These features are consistent with the features constructed during online estimation, ensuring that the model training input and the online inference input are consistent in form and distribution.

[0146] Step S903: Using the reference aortic pressure as the target output and the historical sequence, differential and statistical features of the gas-side pressure signal as input, a multi-condition aortic pressure inverse calculation model is trained.

[0147] After feature construction, the calibration data is organized into a training sample set. The input of each training sample consists of historical sequences, differential and statistical features constructed from the gas-side pressure signal, and the target output is the reference aortic pressure collected synchronously at the same time. The goal of training is to enable the model to learn the mapping relationship between the input features and the target output, that is, to be able to output the corresponding aortic pressure value as accurately as possible given the dynamic features of the gas-side pressure signal.

[0148] Model training can be achieved using various regression or machine learning methods. During training, by learning from a large number of samples covering different aortic pressures and system operating conditions, the model can capture the changing relationship between gas-side pressure signal features and aortic pressure as the operating conditions change, and encode this relationship into the model parameters. Thus, during online estimation, by inputting the dynamic gas-side pressure sequence features within the current effective identification window, the model can output an estimated aortic pressure value based on the learned mapping relationship, eliminating the need to rely on a reference pressure sensor.

[0149] Through the above offline calibration process, the multi-condition aortic pressure back-calculation model can maintain its estimation capability over a wide range of physiological and equipment conditions, solving the problem of insufficient adaptability of single-condition calibration coefficients under complex usage conditions, and providing a model basis for online aortic pressure estimation.

[0150] In some embodiments, the multi-condition aortic pressure back-calculation model can employ a ridge regression model with heart rate zoning, a 64-point finite historical window, and diagonal quadratic feature expansion.

[0151] Because the period length, waveform morphology, and diaphragm recovery time between adjacent pulsations of aortic pulsations differ at different heart rates, the time scale of the pulsating component and the phase relationship of pressure transmission in the gas-side pressure signal also change accordingly. If the same set of model parameters is used for all heart rate ranges, the model will struggle to adapt to the transmission characteristics at both fast and slow heart rates. Therefore, the heart rate is divided into multiple intervals, and independent model parameters are established for each interval, enabling the model to map more accurately to the current heart rate range. During online estimation, the interval is determined based on the current heart rate or trigger cycle, and the parameters for the corresponding interval are selected for calculation, thereby improving the estimation accuracy and adaptability under different heart rate conditions.

[0152] The transmission between gas-side pressure signals and aortic pressure exhibits inertia and time lag. The current aortic pressure is related not only to the current gas-side pressure but also to the changes in gas-side pressure over a previous period. Employing a 64-point finite history window—that is, taking the gas-side pressure values ​​of 64 consecutive sampling points up to and including the current moment as the historical sequence—retains sufficient waveform information to cover the main changes in the aortic pulsation wave. The finite history window effectively characterizes the dynamic changes in the pressure signal while avoiding the computational burden and reduced real-time performance caused by excessively long windows. Furthermore, the 64-point window provides ample waveform context information while meeting the real-time processing requirements of the embedded controller, helping the model determine the current waveform stage of the pressure.

[0153] The relationship between gas-side pressure signals and aortic pressure is not entirely linear. The tangential stiffness of the diaphragm, gas compressibility, and pressure transmission gain change with pressure conditions, leading to a nonlinear pressure transmission relationship. To maintain model simplicity while introducing nonlinear expressiveness, a diagonal quadratic expansion is performed on the basic features. The basic features include historical sampled values, first-order differences, and window statistics. The diagonal quadratic expansion involves calculating the squared term for each basic feature and combining the first-order term with the squared term as the expanded feature. This preserves the linear relationship reflected by the first-order term while introducing nonlinear variation information for individual features through the squared term, enabling the description of the nonlinear characteristics of pressure transmission gain changing with pressure conditions. Furthermore, the diagonal expansion avoids generating cross-product terms between different features, effectively controlling the feature dimensionality and preventing excessive model complexity and overfitting.

[0154] During the training phase, the synchronously acquired reference aortic pressure is used as the target output, and the features after diagonal quadratic expansion are used as input. Ridge regression is employed for model training. Ridge regression introduces a regularization term on top of ordinary linear regression, which can suppress excessively large values ​​of model parameters, reduce the risk of overfitting, and improve the model's generalization ability under unknown conditions. Furthermore, when there is a certain correlation between input features, ridge regression exhibits better numerical stability, avoiding solution instability caused by ill-conditioned feature matrices. After training, the model parameters are stored as feature weights and biases; the calculation process involves only feature standardization and multiply-accumulate operations, facilitating real-time execution in the embedded controller.

[0155] The model can be exported in a parametric form compatible with the controller platform and run on embedded controllers such as the STM32F429. During online estimation, standardized and regression parameters for the corresponding partition are selected based on the current heart rate. The features constructed from the 64-point window are diagonally expanded and standardized, and then the aortic pressure estimate is obtained through linear multiplication and addition operations. This implementation method, while ensuring estimation performance, has low computational complexity and good real-time performance, meeting the requirements of percutaneous ventricular assist devices for online aortic pressure estimation.

[0156] Figure 10 A flowchart illustrating an aortic pressure estimation method for a percutaneous ventricular assist device, as provided in another embodiment of this application. Figure 10 As shown, in some embodiments, after determining the effective identification window for aortic pressure estimation in step S203, the following steps are further included:

[0157] Step S1001: Control the pressure regulating component to apply multiple candidate preload pressures to the gas-side membrane cavity in sequence, and obtain the corresponding gas-side dynamic pressure sequence under each candidate preload pressure.

[0158] Although the pneumatic diaphragm has entered an effective pressure transmission state at a certain preload pressure, its operating point changes with different preload pressures. Different preload pressures will place the diaphragm in different equilibrium positions, corresponding to different local tangential stiffness, effective area, and gas-side volume. These differences will affect the gain and phase characteristics of the aortic pulsation transmitted to the gas side, and thus affect the amplitude, clarity, and stability of the pulsation component in the gas-side pressure signal. Therefore, a single preload pressure point is not necessarily the operating point with optimal pressure transmission quality.

[0159] In this embodiment, the pressure regulating component sequentially applies multiple candidate preload pressures to the gas-side membrane cavity. These candidate preload pressures can be a set of preset discrete pressure values ​​or a continuous pressure sequence formed by stepwise adjustment. After each candidate preload pressure is applied and the gas-side pressure stabilizes, the dynamic pressure sequence of the gas side is acquired within the corresponding effective identification window at that pressure. This allows for the acquisition of dynamic gas-side pressure sequences of the diaphragm at multiple different operating points, providing a data foundation for subsequent pressure transmission quality evaluation and optimal operating point selection.

[0160] Step S1002: Calculate the pressure transmission quality index for the gas-side dynamic pressure sequence corresponding to each candidate preload pressure; the pressure transmission quality index is determined based on at least one of pulse amplitude, signal-to-noise ratio, consistency between adjacent periods, baseline drift, and pressure overrun penalty.

[0161] The pressure transmission quality of gas-side dynamic pressure sequences acquired under different candidate preload pressures varies. To objectively compare the quality of each pressure point, it is necessary to calculate a pressure transmission quality index for each sequence, integrating multiple dimensions of signal quality into a comparable value.

[0162] The pressure transmission quality index is determined based on at least one of the following: pulse amplitude, signal-to-noise ratio (SNR), adjacent cycle consistency, baseline drift, and pressure overrun penalty. Specifically, pulse amplitude reflects the strength of the aortic pulsation component in the gas-side pressure signal; a larger amplitude indicates a more prominent pulsation component, which is more conducive to accurate back-calculation. SNR reflects the proportion of the pulsation component relative to noise and residual disturbances; a higher SNR indicates a cleaner signal. Adjacent cycle consistency reflects the similarity of waveforms between multiple consecutive pulsation cycles; better consistency indicates more stable diaphragm pressure transmission. Baseline drift reflects the degree of slow deviation of the pressure baseline; smaller drift indicates a more stable pressure reference. The pressure overrun penalty is used to penalize abnormal situations such as gas-side pressure exceeding the sensor range, diaphragm adhesion, or travel limit, ensuring that the pressure transmission quality index reflects these adverse factors. By combining the above items according to preset weights, the pressure transmission quality index corresponding to each candidate preload pressure can be obtained. A higher index value indicates that the gas-side dynamic pressure sequence at that pressure point is more suitable for aortic pressure back-calculation.

[0163] Step S1003: Select the gas-side dynamic pressure sequence under the candidate preload pressure with the best pressure transmission quality index for aortic pressure back-calculation; or, perform weighted fusion of aortic pressure estimation results under multiple candidate preload pressures based on the pressure transmission quality index of each candidate preload pressure.

[0164] After obtaining the pressure transmission quality index of each candidate preload pressure, there are two ways to utilize it. The first method is to select the candidate preload pressure with the optimal pressure transmission quality index, and use the gas-side dynamic pressure sequence collected at that pressure point as input to the inverse calculation model to obtain the aortic pressure estimation result. This method is simple and direct, ensuring that the input signal comes from the operating point with the best pressure transmission quality, reducing the influence of low-quality signals on the estimation result. The second method is to perform inverse calculations separately using the gas-side dynamic pressure sequences at multiple candidate preload pressures to obtain multiple aortic pressure estimation results, and then weight and fuse these results according to the pressure transmission quality index of each pressure point. The estimation result corresponding to the pressure point with a higher pressure transmission quality index is assigned a larger weight, while the pressure point with a lower pressure transmission quality index has a smaller weight. This comprehensively utilizes information from multiple operating points, avoiding the random bias that may exist at a single operating point, and further improving the stability and reliability of the estimation result.

[0165] Through the above process, based on the effective identification window, a comparison and fusion mechanism of multiple preload pressures is introduced, which can screen out the conditions with the best pressure transmission quality from different diaphragm operating points, or use the complementary information of multiple operating points to make the gas-side dynamic pressure sequence used for aortic pressure back-calculation have higher signal quality, ultimately improving the accuracy and robustness of aortic pressure estimation.

[0166] Figure 11 A flowchart illustrating an aortic pressure estimation method for a percutaneous ventricular assist device, as provided in another embodiment of this application. Figure 11 As shown, in some embodiments, after obtaining the aortic pressure estimation result in step S205, the following steps are also included:

[0167] Step S1101: Perform a consistency analysis on the aortic pressure estimation results obtained from multiple consecutive effective identification windows.

[0168] Considering that single aortic pressure estimates may be affected by random factors, such as transient aerodynamic disturbances, individual irregular heartbeats, or local signal fluctuations, even after filtering through an effective identification window, the influence of these random factors on the estimation results cannot be completely eliminated. Outputting the estimate based solely on a single window's result may lead to deviations from the true aortic pressure if unidentified anomalies exist within that window, affecting subsequent assessments of auxiliary effects and adjustments to operating parameters. Therefore, it is necessary to perform a consistency analysis on the estimation results obtained from multiple consecutive effective identification windows, using the repeatability between multiple estimates to evaluate the stability of the results.

[0169] Consistency analysis can be performed by comparing the differences in systolic blood pressure, diastolic blood pressure, mean blood pressure, and pulse pressure between adjacent windows to see if they are within a preset range; calculating the correlation between the estimated waveforms of multiple windows; or statistically analyzing the dispersion of the estimation results for each window. When the estimation results of multiple consecutive windows are close to each other and have small fluctuations, it indicates that the estimation process is stable and the possibility of accidental interference is low. When the results of each window differ significantly and have a large degree of dispersion, it suggests that there may be unresolved disturbances or that the current pressure transmission conditions have changed, and the reliability of the estimation results is questionable.

[0170] Step S1102: Determine the reliability of the aortic pressure estimation results based on the consistency analysis results.

[0171] Credibility is a quantitative evaluation of the reliability of the current estimation result, based on the consistency of the estimation results of multiple consecutive windows in step S1101. Good consistency means that the estimation results obtained under different time windows corroborate each other, the estimated values ​​have high repeatability, and the credibility is correspondingly high; poor consistency means that there are obvious discrepancies between the estimations, the estimation results may be affected by unstable factors, and the credibility is correspondingly low.

[0172] The reliability of the aortic pressure estimate can be determined based on a combination of indicators. Besides consistency between consecutive windows, other factors include whether the characteristics of the current gas-side pressure signal fall within the model calibration range, the number of effective cardiac cycles, and the dispersion of the pressure estimation results across multiple cardiac cycles. By integrating these factors, a reliability level or score for the current aortic pressure estimate is obtained, providing a basis for deciding whether to output the result.

[0173] Step S1103: When the confidence level meets the preset output conditions, output the aortic pressure estimation result.

[0174] When the established confidence level reaches or exceeds the preset output condition, it indicates that the current estimation result has undergone cross-validation across multiple windows, with consistent results across different windows, demonstrating high repeatability and stability of the estimated value. At this point, the estimation result can be output for auxiliary effect evaluation, output load judgment, flow estimation, operating parameter adjustment, or pressure trend display. By setting confidence output conditions, unreliable pressure values ​​can be avoided when the estimation result is unstable or interfered with, thereby reducing the risk of misjudgment due to estimation errors. The output result may include at least one of the following: instantaneous aortic pressure waveform, systolic pressure, diastolic pressure, mean pressure, pulse pressure, or pressure change trend.

[0175] Step S1104: When the confidence level does not meet the preset output conditions, redetermine the effective identification window and / or reacquire the gas-side dynamic pressure sequence within the effective identification window, and reacquire the aortic pressure estimation result based on the multi-condition aortic pressure back-calculation model.

[0176] When the reliability does not meet the preset output conditions, it indicates that the current estimation result is not reliable enough and should not be output directly. In this case, signal acquisition and estimation need to be repeated to obtain more reliable results. Specifically, the effective identification window can be redefined, i.e., a time interval that meets the pressure signal quality conditions and aerodynamic operating conditions can be found in subsequent time periods; alternatively, the gas-side dynamic pressure sequence can be reacquired within the currently determined effective identification window to eliminate any abnormal data that may exist in a single acquisition. After reacquiring the gas-side dynamic pressure sequence, it is processed again based on the multi-condition aortic pressure back-calculation model to obtain new aortic pressure estimation results. Consistency analysis and reliability judgment are then performed again to form a closed loop until a reliable estimation result that meets the output conditions is obtained, or if the conditions are not met after multiple attempts, a state of no output or invalid output is selected, thus ensuring that the output aortic pressure estimation result has sufficient reliability.

[0177] In summary, the aortic pressure estimation method for percutaneous ventricular assist devices provided in the above embodiments acquires gas-side pressure signals and determines whether the pneumatic diaphragm is in an effective pressure transmission state, enabling the diaphragm to operate under preload tension and meet stability conditions, thus providing stable and repeatable boundary conditions for the transmission of aortic pressure to the gas side. Based on this, an effective identification window is determined by combining the gas-side pressure signal quality and the device's pneumatic operating status. Gas-side dynamic pressure sequences are collected within time intervals with low pneumatic disturbance and clear aortic pulsation components, effectively suppressing the impact of interference such as pressure regulating valve action and gas source fluctuations on the estimation results. Furthermore, the back-calculation model obtained through offline calibration under multiple operating conditions is used to process the historical sampling values, first-order differences, and statistical characteristics of the gas-side dynamic pressure sequences, adapting to changes in pressure transmission relationships under different heart rates, aortic pressures, system resistance, and compliance conditions. Simultaneously, by comparing and fusing the pressure transmission quality under multiple candidate preload pressures, the diaphragm operating point selection is optimized, and the reliability of the output results is ensured through multi-window consistency analysis and a reliability gating mechanism. The entire method can stably and accurately obtain aortic pressure estimation results using only the existing gas-side pressure signal of the percutaneous ventricular assist device without adding a dedicated blood-side pressure sensor. This avoids the problems caused by adding blood-side sensors, such as increased blood contact parts, thrombosis risk, sealing requirements, sterilization costs, and long-term reliability. It provides a reliable pressure basis for evaluating the assist effect, judging the output load, estimating the flow rate, and adjusting the operating parameters.

[0178] The following is a specific implementation example, illustrating the complete process from offline calibration to online estimation. The process is divided into two stages: the offline stage utilizes a simulated circulatory system and a reference pressure sensor to establish a multi-condition aortic pressure back-calculation model; the online stage uses signals collected by a gas-side pressure sensor, and through effective pressure transmission state judgment, effective identification window determination, feature construction, and model calculation, obtains the aortic pressure estimation result.

[0179] The offline phase includes multi-condition calibration data collection and inverse calculation model training.

[0180] During offline calibration, the percutaneous ventricular assist device (PVC) is connected to a simulated circulatory system capable of adjusting aortic pressure, flow rate, system resistance, compliance, or heart rate. A reference pressure sensor is placed at the aortic location of the simulated circulatory system to provide the true aortic pressure value. The signal from the reference pressure sensor can be acquired via testing equipment or synchronously using the testing sampling channel of the PVC host device.

[0181] Synchronization data set under the j-th calibration condition It can be represented as:

[0182]

[0183] in, This is the gas-side pressure value. For reference aortic pressure values, The target pressure or actual stable pressure value is the preloaded pressure. For heart rate or trigger cycle, For valve status / drive phase and operating parameters, Configuration parameters for membrane pumps, pipelines, equipment, or simulated cycles.

[0184] The calibration data should cover at least two different aortic pressure conditions, and preferably include different heart rates, pulse pressures, flow rates, system resistances, compliance, preload pressures, and membrane pump or tubing configurations. Each data set should record valid pressure transmission status markers to ensure consistency between the model training data and the physical state corresponding to the product's online identification window.

[0185] The pneumatic diaphragm is a crucial component in transmitting pressure from the blood side to the gas side. Let the effective area of ​​the pneumatic diaphragm be... The gas-side membrane pressure is The blood pump chamber pressure is The displacement of the pneumatic diaphragm is x (the positive direction is taken as the gas side pointing to the blood side).

[0186] The force balance relationship of the diaphragm is as follows:

[0187]

[0188] in, For the elastic recovery force of the diaphragm, For equivalent damping, For equivalent quality, Additional forces resulting from friction, wrinkling, and material hysteresis.

[0189] Equilibrium point established after gas-side preload The diaphragm's velocity and acceleration are zero, satisfying:

[0190]

[0191] make , , When the diaphragm is in the predetermined tension region and the pulsating displacement is relatively small compared to the diaphragm stroke, the force balance equation is locally linearized near the equilibrium point, resulting in:

[0192]

[0193] in, The tangential stiffness of the diaphragm at the current preload position. This refers to the local equivalent damping of the diaphragm material, gas-liquid damping, and friction. The preload pressure changes the equilibrium position. This changes the effective area of ​​the membrane. Tangential stiffness and equivalent damping By placing the diaphragm in a predetermined tension region, relatively stable local dynamic parameters can be obtained within the identification time window, thereby providing stable mechanical boundary conditions for pressure transmission from the blood side to the gas side.

[0194] The relationship between the blood-side pump chamber pressure and the aortic pressure can be expressed as:

[0195]

[0196] in, and These are the equivalent flow resistance and equivalent inertia of the blood pathway, respectively. Blood flow during the identification phase.

[0197] During the maintenance phase, low flow phase, or identification window where flow rate changes are limited, the blood-side pathway pressure drop term is within the predetermined range. When residual pressure drop exists, flow rate, operating phase, or relevant operating parameters can be used as model inputs, or multi-condition calibration can be used to compensate for the pressure drop.

[0198] The gas on the gas side satisfies the ideal gas law. Expanding with a small perturbation at the equilibrium point, the relative change in gas-side pressure can be expressed as:

[0199]

[0200] in, The changes in gas mass caused by valve opening and closing and gas flow. For temperature changes, This refers to the change in gas-side volume caused by diaphragm displacement. Within the effective identification window, the pressure regulation pathway can be closed, maintained, or limited to allow for... Within the predetermined range; the recognition window is relatively short compared to the thermal drift time, making... The main manifestation is slow-changing components that can be filtered out or calibrated.

[0201] Short-time gas processes can be represented by polyvariate relationships. If the change in gas-side volume is caused by diaphragm displacement, that is... The relationship between the gas-side pressure change and the diaphragm displacement is as follows:

[0202]

[0203] in, This is the pressure displacement coefficient of the air chamber.

[0204] Substituting the above relationship into the linearized force balance equation of the membrane and performing a Laplace transform, we obtain the local transfer function from the blood-side pump chamber pressure to the gas-side membrane chamber pressure:

[0205]

[0206] Considering the transmission characteristics of the gas-side pressure sensor and sampling channel Then the gas-side pressure signal can be expressed as:

[0207]

[0208] in, The pneumatic component generated by the movement of valves, air supply, and motor. For temperature and baseline variation components, For sensor and sampling noise. The above transmission relationship shows that aortic pressure fluctuations can be transmitted to the gas-side pressure sensor via blood-side fluid, diaphragm, and gas-side gas; the transmission gain and phase are simultaneously affected by preload pressure, gas-side volume, diaphragm stiffness, damping, mass, and signal frequency. Therefore, online estimation requires selecting a time window with low aerodynamic disturbance and stable diaphragm state to ensure... and To minimize the proportion of aortic pulsation component in the gas-side pressure signal.

[0209] Within the effective identification window, the above continuous transfer function is discretized into a finite memory form:

[0210]

[0211] in, The discrete dynamic coefficients are formed by the membrane, gas, blood pathway, and sampling system. For the remaining disturbance within the allowable range, The noise is considered. This formula indicates that the current gas-side pressure sample value can be represented as a weighted sum of current and historical aortic pressures plus perturbations and noise. Accordingly, the pressure inverse calculation model uses the absolute level of gas-side pressure, historical sequence, differential and periodic statistical features as inputs to simultaneously characterize the pressure baseline, waveform phase, dynamic changes, and pulsation amplitude.

[0212] This embodiment uses a ridge regression model with heart rate zoning, a 64-point finite historical window, and diagonal quadratic feature expansion as the specific aortic pressure back-calculation model.

[0213] Let the current and historical gas-side pressure sampling sequences be as follows: Construct the first-order difference The basic feature vector is:

[0214]

[0215] in, These represent the mean, standard deviation, minimum, and maximum values ​​for a 64-point window. The window span is specified. The basic feature set consists of 64 stress features, 64 difference features, and 5 statistical features, totaling 133 features.

[0216] Diagonal quadratic expansion of the basic features:

[0217]

[0218] This extension preserves the first-order and square terms of each feature to describe the nonlinearity of diaphragm stiffness, gas compressibility, and pressure transfer gain as a function of pressure state, while not generating cross-products between all features.

[0219] Heart rate information is preferentially obtained from the host's existing ECG, trigger cycle, or stable working cycle. In this embodiment, three heart rate intervals z are defined according to HR<100, 110≤HR<120, and HR≥120. Each heart rate interval has its own feature standardization parameters and regression parameters. The standardization formula is:

[0220]

[0221] During the offline calibration phase, aortic pressure was used as a synchronous reference. To find the target value, solve for the ridge regression parameters within each heart rate zone:

[0222]

[0223] The corresponding analytical solution for the ridge regression parameters is:

[0224]

[0225] in, This is a matrix composed of the standardized features of all samples within this heart rate range. The target mean.

[0226] Through the above offline calibration process, standardized parameters, regression coefficients, and biases of the heart rate zones were obtained, which served as the multi-condition aortic pressure back-calculation model used for online estimation.

[0227] The online phase includes effective pressure transmission status identification and aortic pressure estimation.

[0228] During online operation, the gas-side pressure signal acquired by the gas-side pressure sensor is obtained. According to the aforementioned physical model, the blood-side aortic pulsation can only be stably transmitted to the gas side when the pneumatic diaphragm is in an effective pressure transmission state. Therefore, the first step is to determine whether the pneumatic diaphragm is in an effective pressure transmission state based on the gas-side pressure signal. The judgment criteria are consistent with the effective pressure transmission state markers used during offline calibration, including the gas-side average pressure being within the allowable range of the target preload pressure, the pressure change rate being less than the first threshold, the baseline drift being less than the second threshold, and the pulse wave quality meeting the requirements.

[0229] Provided the pneumatic diaphragm is in an effective pressure transmission state, an effective identification window is further determined by combining the gas-side pressure signal quality and the pneumatic operating status. An effective identification window satisfies at least one of the following conditions: the average gas-side pressure is within the allowable range of the candidate preload pressure; the pressure change rate and baseline drift are both less than the corresponding thresholds; the pressure regulating component is in a closed, held, or low-disturbance state; the gas-side pressure signal exhibits periodic fluctuations matching the heart rate or trigger cycle, and the amplitude and morphology consistency of multiple consecutive cycles reaches the threshold; the gas-side pressure signal does not reach the saturation range of the gas-side pressure sensor, and the pneumatic diaphragm is not in a wall-attached or stroke-limited state. Within the above window, the aortic pulsation component in the gas-side pressure signal has a high proportion, and the pneumatic disturbance is low, making it suitable for pressure inverse calculation.

[0230] Within the effective identification window, gas-side pressure signals are acquired for no less than a predetermined number of heartbeat cycles, while simultaneously acquiring heart rate or trigger cycles. After data processing such as filtering, outlier removal, cycle segmentation, baseline processing, and normalization, the acquired raw signals are used to form a dynamic gas-side pressure sequence in chronological order.

[0231] For this sequence, construct 64 historical sample values, first-order differences, and window statistical features, and form a basic feature vector according to the above formula. and extended feature vectors .

[0232] To further optimize the diaphragm operating point, multiple candidate preload pressures can be applied sequentially. At each candidate preload pressure, the dynamic pressure sequence of the gas side within the effective identification window is obtained, and the pressure transmission quality index is calculated.

[0233]

[0234] in, This refers to the pulse amplitude. For signal-to-noise ratio, For consistency between adjacent periods, For baseline drift, This is a penalty amount for pressure exceeding limits, abnormal waveforms, or period mismatch. Preset weights.

[0235] Select pressure transmission quality indicators The optimal candidate preload pressure corresponds to the gas-side dynamic pressure sequence used for inverse calculation, or based on each pressure point. The estimation results at multiple pressure points are weighted and fused.

[0236] For the selected gas-side dynamic pressure sequence, the corresponding heart rate partition z is selected based on the current heart rate, and the extended features are standardized to obtain:

[0237]

[0238] Then, the estimated aortic pressure is calculated:

[0239]

[0240] When the historical window is not full, the model is in a warm-up state; after the window is full, the estimated aortic pressure values ​​are output for each sampling point.

[0241] After obtaining the aortic pressure estimation waveform at each sampling point, for the nth effective cardiac cycle... The estimated values ​​within the range are used to calculate systolic blood pressure, diastolic blood pressure, mean blood pressure, and pulse pressure:

[0242]

[0243]

[0244] The pressure parameters mentioned above can be used for equipment display, auxiliary effect evaluation, output load judgment, flow estimation input, operating parameter evaluation, or pressure trend indication.

[0245] Through the above complete process, this embodiment starts with offline multi-condition calibration, uses the diaphragm pressure transmission physical model and multi-condition data training to obtain the aortic pressure back calculation model; in the online stage, through effective pressure transmission state judgment, effective identification window screening, multi-preload pressure optimization and feature standardization, it realizes aortic pressure estimation relying only on gas side pressure signal, without the need for additional blood side pressure sensor.

[0246] Figure 12 This is a schematic diagram of the aortic pressure estimation device for a percutaneous ventricular assist device provided in one embodiment of this application. Figure 12 As shown, the aortic pressure estimation device for percutaneous ventricular assist devices provided in this embodiment includes a gas-side pressure acquisition module 1201, a diaphragm state determination module 1202, an identification window determination module 1203, a dynamic pressure acquisition module 1204, and an aortic pressure estimation module 1205.

[0247] In this embodiment, the gas-side pressure acquisition module 1201 is used to acquire the gas-side pressure signal collected by the gas-side pressure sensor. This module receives the pressure data output in real time by the gas-side pressure sensor and converts the pressure fluctuations within the gas-side diaphragm cavity into a time-series signal suitable for subsequent processing according to a preset sampling rate. The acquired gas-side pressure signal is the fundamental data for determining the state of the pneumatic diaphragm and estimating the aortic pressure.

[0248] The diaphragm state determination module 1202 is used to determine whether the pneumatic diaphragm is in an effective pressure transmission state based on the gas-side pressure signal. In the effective pressure transmission state, the pneumatic diaphragm is in a preloaded state, and its response to pressure changes meets preset stability conditions. This module analyzes the gas-side pressure signal to determine whether the pneumatic diaphragm has entered a preloaded state and whether its response to pressure changes meets preset stability conditions. Only when the pneumatic diaphragm is in an effective pressure transmission state can the aortic pulsation be stably transmitted to the gas side, providing reliable boundary conditions for subsequent pressure back-calculation.

[0249] The identification window determination module 1203 is used to determine an effective identification window for aortic pressure estimation based on the gas-side pressure signal and the pneumatic operation status of the percutaneous ventricular assist device when the pneumatic diaphragm is in an effective pressure transmission state. The effective identification window is the time interval corresponding to when the gas-side pressure signal meets preset pressure signal quality conditions and the pneumatic operation status of the percutaneous ventricular assist device meets preset operation status conditions. This module simultaneously judges from two dimensions: signal quality and pneumatic disturbance, filtering out time intervals where the aortic pulsation component in the gas-side pressure signal is clear and the airway disturbance is low. Determining the effective identification window can eliminate the influence of interference such as pressure regulating valve action and gas source fluctuations on the estimation results, ensuring that the subsequently acquired dynamic pressure sequence has a high signal-to-noise ratio and waveform consistency.

[0250] The dynamic pressure acquisition module 1204 is used to acquire the gas-side dynamic pressure sequence within the effective identification window. This module continuously acquires gas-side pressure signals within the effective identification window and forms a dynamic pressure sequence containing multiple cardiac cycles in chronological order. Compared with single-point pressure values, the dynamic pressure sequence carries the morphology, rate of change, and statistical characteristics of the pressure waveform, providing more complete dynamic information for the inverse calculation model.

[0251] The aortic pressure estimation module 1205 is used to determine the dynamic mapping relationship between the gas-side dynamic pressure sequence and the aortic pressure based on a pre-established multi-condition aortic pressure inverse calculation model, and obtain the aortic pressure estimation result. This module selects the corresponding model parameters according to the current heart rate or trigger cycle, performs feature construction and standardization processing on the dynamic pressure sequence, and calculates the aortic pressure estimate. The estimation result may include the instantaneous aortic pressure waveform, systolic pressure, diastolic pressure, mean pressure, pulse pressure, or pressure change trend.

[0252] In some embodiments, the aortic pressure estimation device for percutaneous ventricular assist devices further includes a reliability evaluation module 1206, which performs consistency analysis on the aortic pressure estimation results obtained from multiple consecutive effective identification windows; determines the reliability of the aortic pressure estimation results based on the consistency analysis results; and outputs the aortic pressure estimation results when the reliability meets preset output conditions. This module avoids interference from random factors that may occur in a single estimation by cross-verifying the estimation results from multiple windows, thus ensuring the reliability of the output results.

[0253] The working process of each module in the above device embodiment is the same as the implementation of the corresponding steps in the aforementioned method embodiment of this application. The technical problems solved, the technical means adopted, and the beneficial effects obtained are also the same. To avoid repetition, they will not be described again here.

[0254] This application also provides a readable storage medium storing a program or instructions that, when executed by a processor, implement the various processes of any embodiment of the aortic pressure estimation method for percutaneous ventricular assist devices described above, and achieve the same technical effect. To avoid repetition, these will not be described again here.

[0255] The processor can be a central processing unit (CPU), an application-specific integrated circuit (ASIC), or one or more integrated circuits configured to implement the embodiments of this application. The readable storage medium includes computer-readable storage media, such as computer read-only memory (ROM), random access memory (RAM), magnetic disk, or optical disk.

[0256] In the above embodiments, implementation can be achieved, in whole or in part, by software, hardware, firmware, or any combination thereof. Furthermore, as those skilled in the art will understand, the principles herein can be reflected in a computer program product on a computer-readable storage medium pre-loaded with computer-readable program code. Any tangible, non-transitory computer-readable storage medium may be used, including magnetic storage devices (hard disks, floppy disks, etc.), optical storage devices (CDs, DVDs, Blu-ray discs, etc.), flash memory, and / or the like. These computer program instructions can be loaded onto a general-purpose computer, a special-purpose computer, or other programmable data processing apparatus to form a machine, such that instructions executing on the computer or other programmable data processing apparatus can generate means for performing a specified function. These computer program instructions can also be stored in a computer-readable storage medium that can instruct the computer or other programmable data processing apparatus to operate in a particular manner, such that instructions stored in the computer-readable storage medium can form an article of manufacture, including means for implementing the specified function. The computer program instructions can also be loaded onto a computer or other programmable data processing apparatus to perform a series of operational steps on the computer or other programmable apparatus to produce a computer-implemented process, such that instructions executing on the computer or other programmable apparatus can provide steps for implementing the specified function.

[0257] This document describes various exemplary embodiments with reference to them. However, those skilled in the art will recognize that changes and modifications can be made to the exemplary embodiments without departing from the scope of this document. For example, various operational steps and components for performing operational steps can be implemented in different ways depending on the specific application or considering any number of cost functions associated with the operation of the system (e.g., one or more steps can be deleted, modified, or combined with other steps).

[0258] While the principles herein have been illustrated in various embodiments, numerous modifications to the structures, arrangements, proportions, elements, materials, and components, particularly suited to specific environments and operational requirements, may be used without departing from the principles and scope of this disclosure. These modifications and other alterations or alterations will be included within the scope of this document. Those skilled in the art will recognize that many changes can be made to the details of the above embodiments without departing from the fundamental principles of the invention.

Claims

1. A method for estimating aortic pressure in a percutaneous ventricular assist device, the percutaneous ventricular assist device comprising a blood-side pump chamber, a gas-side membrane chamber, a pneumatic diaphragm separating the blood-side pump chamber and the gas-side membrane chamber, a gas path and pressure regulating component communicating with the gas-side membrane chamber, and a gas-side pressure sensor communicating with the gas-side membrane chamber; the pneumatic diaphragm periodically changes the volume of the blood-side pump chamber under the drive of the gas-side pressure in the gas-side membrane chamber, thereby delivering blood from the inlet side to the aortic side; Its features are, The aortic pressure estimation method includes: Acquire the gas-side pressure signal collected by the gas-side pressure sensor; Based on the gas-side pressure signal, it is determined whether the pneumatic diaphragm is in an effective pressure transmission state; wherein, when the pneumatic diaphragm is in the effective pressure transmission state, its response to pressure changes satisfies a preset stability condition. When the pneumatic diaphragm is in the effective pressure transmission state, an effective identification window for aortic pressure estimation is determined based on the gas-side pressure signal and the pneumatic operation state of the percutaneous ventricular assist device; wherein, the effective identification window is the time interval corresponding to the gas-side pressure signal meeting the preset pressure signal quality condition and the pneumatic operation state of the percutaneous ventricular assist device meeting the preset operation state condition. Obtain the gas-side dynamic pressure sequence within the effective identification window; Based on a pre-established multi-condition aortic pressure back-calculation model, the dynamic mapping relationship between the gas-side dynamic pressure sequence and the aortic pressure is determined, and the aortic pressure estimation result is obtained.

2. The aortic pressure estimation method for percutaneous ventricular assist devices according to claim 1, characterized in that, The step of determining whether the pneumatic diaphragm is in an effective pressure transmission state based on the gas-side pressure signal includes: Analyze the gas-side pressure signal to obtain the corresponding pressure change characteristics; Based on the pressure change characteristics, determine the response state of the pneumatic diaphragm under gas-side pressure; When the response state of the pneumatic diaphragm meets the preset stability condition, the pneumatic diaphragm is determined to be in an effective pressure transmission state. The pressure change characteristics include at least one of the following: gas-side pressure change rate, gas-side pressure change period, gas-side pressure fluctuation range, adjacent pressure change trend, and pressure response consistency among multiple periods. The preset stability conditions include the pressure stability of the gas-side pressure signal meeting a first preset requirement, and / or the pulse wave quality in the gas-side pressure signal that matches the heart rate or trigger cycle meeting a second preset requirement; the first preset requirement includes the gas-side average pressure being within the allowable range of the target preload pressure, the pressure change rate being less than the corresponding threshold, and the baseline drift being less than the corresponding threshold; the second preset requirement includes the pulse amplitude reaching a preset amplitude threshold, the signal-to-noise ratio reaching a preset signal-to-noise ratio threshold, and the amplitude and morphology consistency of multiple consecutive cycles reaching a preset consistency threshold.

3. The aortic pressure estimation method for percutaneous ventricular assist devices according to claim 2, characterized in that, The pressure change characteristics and the response state of the pneumatic diaphragm are obtained through an active pressure build-up process, which includes: The pressure regulating component is controlled to apply a controlled preload pressure to the gas-side diaphragm cavity through the air passage, so that the pneumatic diaphragm enters a predetermined tension state; After the pneumatic diaphragm is in the tensioned state, the pressure regulating component is sequentially controlled to apply multiple preset candidate gas side pressures to the gas-side diaphragm cavity through the air passage; Obtain the gas-side pressure response of the pneumatic diaphragm under tension state corresponding to the pressure formation of each candidate gas side; Based on the gas-side pressure response, the pressure change characteristics and the response state of the pneumatic diaphragm are determined.

4. The aortic pressure estimation method for percutaneous ventricular assist devices according to claim 2, characterized in that, The pressure change characteristics and the response state of the pneumatic diaphragm are obtained through an operational status identification process, which includes: Acquire the pneumatic operating status of the percutaneous ventricular assist device during operation and determine the current operating stage; The gas-side pressure signal corresponding to the current operating stage is obtained, and the force state, pressure stability and aerodynamic disturbance of the pneumatic diaphragm are determined based on the gas-side pressure signal. The pressure change characteristics and the response state of the pneumatic diaphragm corresponding to the current operating stage are determined based on the stress state, the pressure stability, and the aerodynamic disturbance.

5. The aortic pressure estimation method for percutaneous ventricular assist devices according to claim 1, characterized in that, The step of determining an effective identification window for aortic pressure estimation based on the gas-side pressure signal and the pneumatic operating status of the percutaneous ventricular assist device includes: The gas-side pressure signal is analyzed to obtain corresponding signal quality information; wherein, the signal quality information includes at least one of the following: pressure change rate, pressure baseline change degree, pressure signal stability, pulse amplitude, pressure signal-to-noise ratio, and pressure period consistency. Obtain the pneumatic operating status information corresponding to the percutaneous ventricular assist device; wherein, the pneumatic operating status information includes at least one of the following: gas-side driving pressure status, gas path switching status, driving cycle status, and percutaneous ventricular assist device operating mode; Based on the signal quality information and the pneumatic operating status information, determine the time interval that meets the aortic pressure estimation conditions; The time interval that satisfies the aortic pressure estimation conditions is determined as the effective identification window.

6. The aortic pressure estimation method for percutaneous ventricular assist devices according to claim 5, characterized in that, The effective identification window satisfies at least one of the following conditions: The average pressure of the gas-side pressure signal is within the allowable range of the candidate preload pressure. The pressure change rate and baseline drift of the gas-side pressure signal are both less than the corresponding threshold. The voltage regulating component is in a closed, held, or low-disturbance state; The gas-side pressure signal contains periodic fluctuations that match the heart rate or trigger cycle, and the amplitude and morphology of multiple consecutive cycles are consistent to a threshold. The gas-side pressure signal did not reach the saturation range of the gas-side pressure sensor, and the pneumatic diaphragm was not in a state of contact with the wall or in a stroke limit state.

7. The aortic pressure estimation method for percutaneous ventricular assist devices according to claim 1, characterized in that, The acquisition of the gas-side dynamic pressure sequence within the effective identification window includes: Within the effective identification window, gas-side pressure signals for no less than a predetermined number of cardiac cycles are collected, and the heart rate or trigger cycle of the percutaneous ventricular assist device is acquired simultaneously. The gas-side pressure signals within a predetermined number of cardiac cycles are arranged in chronological order to form the gas-side dynamic pressure sequence, and the gas-side dynamic pressure sequence is processed to obtain input data for the multi-condition aortic pressure inverse calculation model. The data processing includes at least one of pressure signal filtering, abnormal data removal, periodic segmentation, baseline processing, and data normalization.

8. The aortic pressure estimation method for percutaneous ventricular assist devices according to claim 1, characterized in that, The multi-condition aortic pressure back-calculation model is pre-established through the following steps: Under multiple aortic pressure conditions and multiple system conditions, the reference aortic pressure and the gas-side pressure signal of the gas-side pressure sensor are simultaneously acquired; wherein, the system conditions include at least one of heart rate, flow rate, system resistance, compliance, preload pressure, membrane pump type and tubing parameters; The time periods during which the pneumatic diaphragm is in an effective pressure transmission state under each operating condition are marked, and the historical sequence, differential, and statistical characteristics of the gas-side pressure signal are constructed. The multi-condition aortic pressure inverse calculation model is trained by using the reference aortic pressure as the target output and the historical sequence, difference, and statistical features of the gas-side pressure signal as input. The multi-condition aortic pressure back-calculation model is divided into zones according to heart rate intervals, and feature standardization parameters and regression parameters are set for each heart rate interval.

9. The aortic pressure estimation method for percutaneous ventricular assist devices according to claim 8, characterized in that, The method, based on a pre-established multi-condition aortic pressure back-calculation model, determines the dynamic mapping relationship between the gas-side dynamic pressure sequence and the aortic pressure, and obtains the aortic pressure estimation result, including: Historical sampling values, first-order difference sequences, and window statistical features are constructed based on the gas-side dynamic pressure sequence. Select the corresponding heart rate partition parameter in the multi-condition aortic pressure back calculation model based on the current heart rate or trigger cycle; The historical sampled values, first-order difference sequences, and window statistical features are standardized using the heart rate zoning parameters. The standardized features are then input into the multi-condition aortic pressure inverse calculation model to calculate the aortic pressure estimate. The aortic pressure estimate includes instantaneous aortic pressure, aortic pressure change trend, and at least one of systolic pressure, diastolic pressure, mean pressure, and pulse pressure obtained from the instantaneous aortic pressure waveform.

10. The aortic pressure estimation method according to claim 1, characterized in that, Also includes: The pressure regulating component is controlled to sequentially apply multiple candidate preload pressures to the gas-side membrane cavity, and the corresponding gas-side dynamic pressure sequence is obtained under each candidate preload pressure. For each candidate preload pressure, a pressure transmission quality index is calculated for the gas-side dynamic pressure sequence corresponding to the pressure-side. The pressure transmission quality index is determined based on at least one of pulse amplitude, signal-to-noise ratio, consistency between adjacent cycles, baseline drift, and pressure overrun penalty. The gas-side dynamic pressure sequence under the candidate preload pressure with the best pressure transmission quality index is selected for aortic pressure back-calculation; or, the aortic pressure estimation results under multiple candidate preload pressures are weighted and fused according to the pressure transmission quality index of each candidate preload pressure.

11. The aortic pressure estimation method according to claim 1, characterized in that, After obtaining the aortic pressure estimation result, the following is also included: Consistency analysis was performed on the aortic pressure estimation results obtained from multiple consecutive effective identification windows. Based on the consistency analysis results, the reliability of the aortic pressure estimation results is determined; When the confidence level meets the preset output conditions, the aortic pressure estimation result is output; When the confidence level does not meet the preset output conditions, the effective identification window is redefined and / or the gas-side dynamic pressure sequence within the effective identification window is reacquired, and the aortic pressure estimation result is re-obtained based on the multi-condition aortic pressure back-calculation model.

12. An aortic pressure estimation device for percutaneous ventricular assist devices, characterized in that, include: The gas-side pressure acquisition module is used to acquire the gas-side pressure signal collected by the gas-side pressure sensor. The diaphragm state determination module is used to determine whether the pneumatic diaphragm is in an effective pressure transmission state based on the gas-side pressure signal; wherein, the pneumatic diaphragm is in a preloaded stress state in the effective pressure transmission state, and the response of the pneumatic diaphragm under pressure change satisfies a preset stability condition. The identification window determination module is used to determine an effective identification window for aortic pressure estimation based on the gas-side pressure signal and the pneumatic operating state of the percutaneous ventricular assist device when the pneumatic diaphragm is in the effective pressure transmission state; wherein, the effective identification window is the time interval corresponding to the gas-side pressure signal satisfying a preset pressure signal quality condition and the pneumatic operating state of the percutaneous ventricular assist device satisfying a preset operating state condition. The dynamic pressure acquisition module is used to acquire the gas-side dynamic pressure sequence within the effective identification window; The aortic pressure estimation module is used to determine the dynamic mapping relationship between the gas-side dynamic pressure sequence and the aortic pressure based on a pre-established multi-condition aortic pressure back-calculation model, and to obtain the aortic pressure estimation result.

13. The aortic pressure estimation device for percutaneous ventricular assist devices according to claim 12, characterized in that, It also includes a credibility evaluation module; The credibility evaluation module is used to perform consistency analysis on the aortic pressure estimation results obtained from multiple consecutive effective identification windows; determine the credibility of the aortic pressure estimation results based on the consistency analysis results; and output the aortic pressure estimation results when the credibility meets the preset output conditions.

14. A percutaneous ventricular assist device, characterized in that, include: The blood-side pump chamber is connected to the blood inlet side and the aortic side, and is used to contain the blood to be transported and to transport the blood from the blood inlet side to the aortic side when the volume changes; A gas-side membrane cavity is used to contain gas, and the pressure change of the gas drives the volume change of the blood-side pump cavity; A pneumatic diaphragm is disposed between the blood-side pump chamber and the gas-side diaphragm chamber, and is used to periodically change the volume of the blood-side pump chamber under the drive of the gas-side pressure in the gas-side diaphragm chamber, so as to drive the blood in the blood-side pump chamber to be transported from the blood inlet side to the aortic side; The gas path and pressure regulating component are connected to the gas-side membrane cavity and are used to provide gas-side pressure to the gas-side membrane cavity and regulate the gas-side pressure in the gas-side membrane cavity so that the pneumatic diaphragm is in a preloaded state. A gas-side pressure sensor, connected to the gas-side membrane cavity, is used to collect the gas-side pressure signal within the gas-side membrane cavity; The controller, connected to the gas-side pressure sensor and the gas path and pressure regulating components, is configured to acquire the gas-side pressure signal and perform the aortic pressure estimation method for a percutaneous ventricular assist device as described in any one of claims 1-11.

15. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer-executable program or instructions, which, when executed by a processor, are used to implement the aortic pressure estimation method for a percutaneous ventricular assist device as described in any one of claims 1-11.

Images