Systems and methods for hemodynamic monitoring using computational surrogates of heart position
By using computational methods to correct the hemodynamic parameters sensed by the sensor to the level of the heart, the problem of inaccurate blood pressure readings caused by changes in sensor position is solved, achieving robust hemodynamic monitoring and improving the accuracy and consistency of monitoring.
Patent Information
- Application Number
- CN202580029629.5
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Priority Date
- 2024-04-26
- Filing Date
- 2025-04-25
- Publication Date
- 2026-11-17
AI Technical Summary
Existing hemodynamic monitoring technologies cannot achieve robust continuous monitoring because the vertical orientation of the sensor relative to the heart causes inaccurate blood pressure readings.
By using computational methods to correct the hemodynamic parameters sensed by the sensor to the level of the heart, and using a motion detector to update the correction amount in real time, the reliance on additional sensors and cables is reduced, and automatic compensation for changes in sensor position is achieved.
It improves the accuracy and consistency of hemodynamic monitoring, reduces errors caused by sensor position changes, and enhances the robustness of blood pressure monitoring and the reliability of patient outcomes.
Smart Images

Figure CN123124928A_ABST
Abstract
Description
Technical Field
[0001] This disclosure generally relates to systems and methods for hemodynamic monitoring, and more specifically, to systems and methods for adjusting sensed hemodynamic parameters to the cardiac level. Background Technology
[0002] Continuous noninvasive blood pressure monitors enable real-time measurement of blood pressure waves and the derivation of hemodynamic parameters. Commonly used techniques include volume clamping and tonometry. Other emerging technologies include pulse wave transit time, pulse arrival time, and capacitive pressure sensors. Each technique has its advantages and disadvantages, but generally, they all involve continuously monitoring blood pressure by placing one or more sensors on a patient's limb to sense the physiology associated with blood pressure and heart rate. In each case, the system acquires the blood pressure at the sensor placement site. The sensor's position relative to the heart can affect blood pressure readings due to variations in hydrostatic pressure, as gravity affects local blood pressure. Summary of the Invention
[0003] Systems and methods for continuous hemodynamic measurement may include a sensor system used on a patient. Hemodynamic data sensed by the sensor can be corrected based on the vertical orientation relative to the heart. A calculation method can calculate a correction based on the hemodynamic data generated by the sensor system. The calculation method can further identify changes in the sensor's vertical orientation and then calculate an updated correction based on the hemodynamic data generated by the sensor system. The correction can be continuously updated as the sensor orientation changes and can be used to continuously correct the sensed hemodynamic parameters.
[0004] In some implementations, a real-time method for correcting sensed hemodynamic data includes using a hemodynamic monitoring system to receive sensor signals acquired from a sensor system at the sensing site.
[0005] In some implementations, a real-time method for correcting sensed hemodynamic data includes using a hemodynamic monitoring system to determine a set of sensed hemodynamic parameters derived from sensor signals.
[0006] In some implementations, a real-time method for correcting sensed hemodynamic data includes using a hemodynamic monitoring system to calculate a correction amount that takes into account the vertical orientation of the sensor system.
[0007] In some implementations, a real-time method for correcting sensed hemodynamic data includes using a hemodynamic monitoring system to continuously correct sensed hemodynamic parameters using a correction factor that takes into account the vertical orientation of the sensor.
[0008] In some implementations, the step of using a hemodynamic monitoring system to calculate a correction amount that takes into account the vertical orientation of the sensor system further includes: using the hemodynamic monitoring system to input a set of sensed hemodynamic parameters into a trained computational model to generate corrected hemodynamic parameters.
[0009] In some implementations, the step of calculating the correction amount taking into account the vertical orientation of the sensor system using a hemodynamic monitoring system further includes: using the hemodynamic monitoring system, determining the correction amount taking into account the vertical orientation of the sensor system using corrected hemodynamic parameters within an invariant input equation. The inputs include the corrected hemodynamic parameters and the sensed hemodynamic parameters. The corrected hemodynamic parameters and the sensed hemodynamic parameters are the same hemodynamic parameters.
[0010] In some implementations, the invariant input equation is: in These are the corrected hemodynamic parameters. These are the sensed hemodynamic parameters, and It is the correction amount; among which and They are the same hemodynamic parameters.
[0011] In some implementation schemes, yes ,and yes .
[0012] In some implementations, the step of calculating a correction amount that takes into account the vertical orientation of the sensor system using a hemodynamic monitoring system further includes: using the hemodynamic monitoring system to receive a motion detector signal, wherein the motion detector signal indicates the vertical orientation of the sensor system relative to the level of the heart.
[0013] In some implementations, the step of calculating a correction amount that takes into account the vertical orientation of the sensor system using a hemodynamic monitoring system further includes: using the hemodynamic monitoring system, determining the correction amount using the vertical orientation of the sensor system relative to the level of the heart.
[0014] In some implementations, a real-time method for correcting sensed hemodynamic data includes using a hemodynamic monitoring system to identify vertical motion of the sensor system corresponding to changes in the vertical orientation of the sensor system.
[0015] In some implementations, a real-time method for correcting sensed hemodynamic data includes updating a correction amount that takes into account the vertical orientation of the sensor system, using a hemodynamic monitoring system.
[0016] In some implementations, the step of updating the correction amount that takes into account the vertical orientation of the sensor system using a hemodynamic monitoring system includes: using the hemodynamic monitoring system, re-inputting the updated set of sensed hemodynamic parameters into a trained computational model to generate updated corrected hemodynamic parameters.
[0017] In some implementations, the step of updating the correction amount that takes into account the vertical orientation of the sensor system using a hemodynamic monitoring system includes: using the hemodynamic monitoring system, determining the correction amount that takes into account the vertical orientation of the sensor system using updated corrected hemodynamic parameters within an invariant input equation. The inputs include the updated corrected hemodynamic parameters and the updated sensed hemodynamic parameters. The updated corrected hemodynamic parameters and the updated sensed hemodynamic parameters are the same hemodynamic parameters.
[0018] In some implementations, the step of using a hemodynamic monitoring system to identify motion of the sensor system includes: using the hemodynamic monitoring system to receive a motion detector signal, wherein the motion detector signal indicates that the amount of vertical motion is greater than a threshold.
[0019] In some implementations, the motion detector signal indicates a change in gravity greater than a threshold.
[0020] In some implementations, the motion detector signal is derived from a motion detector associated with the sensor system.
[0021] In some implementations, the steps for identifying motion of the sensor system using a hemodynamic monitoring system include: In some implementations, a hemodynamic monitoring system is used to detect the vertical motion of the sensor system from a computational model trained to detect changes in sensed hemodynamic parameters in response to the vertical repositioning of the sensor system.
[0022] In some implementations, the step of updating the correction amount that takes into account the vertical orientation of the sensor system using a hemodynamic monitoring system includes: using the hemodynamic monitoring system, receiving a motion detector signal, wherein the motion detector signal indicates a change in the vertical orientation of the sensor system relative to the level of the heart.
[0023] In some implementations, the step of updating the correction amount that takes into account the vertical orientation of the sensor system using a hemodynamic monitoring system includes: using the hemodynamic monitoring system, determining the correction amount based on the change in the vertical orientation of the sensor system relative to the level of the heart.
[0024] In some implementations, the sensor system includes at least one of the following: an applanation tonometer, a pressure cuff, a photoplethysmograph, a peripheral arterial line, one or more electrode leads, a pulse oximeter, or a capacitance-based pressure sensor.
[0025] In some implementations, the steps of using a hemodynamic monitoring system to determine a group of sensed hemodynamic parameters include: performing arterial tension measurement, performing volume clamping, performing catheter-based hemodynamic monitoring, performing pulse wave transit time or pulse arrival time, performing photoplethysmography-based heart rate monitoring, or performing capacitive arterial pressure sensing.
[0026] In some implementations, a real-time method for correcting sensed hemodynamic data includes using a hemodynamic monitoring system to calculate downstream hemodynamic parameters from the corrected sensed hemodynamic parameters.
[0027] In some implementations, downstream hemodynamic parameters are centralized hemodynamic parameters.
[0028] In some implementations, the central hemodynamic parameter is the blood pressure within the ascending aorta.
[0029] In some implementations, a real-time method for correcting sensed hemodynamic data includes using a hemodynamic monitoring system to generate waveforms using corrected sensed hemodynamic parameters.
[0030] In some implementations, a real-time method for correcting sensed hemodynamic data includes generating a waveform using downstream hemodynamic parameters with the aid of a hemodynamic monitoring system. Another real-time method for correcting sensed hemodynamic data includes displaying the waveform on a display of the hemodynamic monitoring system.
[0031] In some implementations, the hemodynamic monitoring system is a clinically used system, a portable system, or a wearable device.
[0032] In some implementations, the hemodynamic monitoring system excludes paired sensor systems, which are used to determine the vertical orientation of the sensing system relative to the heart based on the difference in vertical orientation between each sensor in the paired sensor system in order to calculate the correction amount.
[0033] In some implementations, the hemodynamic monitoring system excludes one or more sensors positioned at the level of the heart to establish the heart's vertical orientation in order to calculate the correction amount.
[0034] In some implementations, the hemodynamic monitoring system excludes the use of a central blood pressure sensor that is positioned outside the thoracic region to calculate the correction.
[0035] In some implementations, the hemodynamic monitoring system for real-time correction of sensed hemodynamic data includes a sensor system.
[0036] In some implementations, the hemodynamic monitoring system for real-time correction of sensed hemodynamic data includes a computing processing system digitally connected to the sensor system.
[0037] In some implementations, the computing system includes a processor system and a memory system, the memory system including one or more application programs that can instruct the processor system to perform the various computing methods described herein. Attached Figure Description
[0038] The specification and claims will be more fully understood with reference to the accompanying drawings and figures, which are given by way of example in this disclosure and should not be construed as a complete description of the scope of this disclosure.
[0039] Figure 1 An example of a sensor system for continuous blood pressure monitoring with a cardiac reference sensor is provided.
[0040] Figure 2 Examples of sensor systems for hemodynamic monitoring are provided.
[0041] Figure 3 Examples of methods for adjusting continuous hemodynamic measurements are provided.
[0042] Figure 4 Examples of computational methods for correcting hemodynamic monitoring parameters using computational alternatives based on cardiac orientation are provided.
[0043] Figure 5 Examples are provided on how to correct sensed hemodynamic data by calculating substitutes.
[0044] Figure 6 provides an example of a computational processing system for continuous hemodynamic monitoring.
[0045] Figures 7A to 7C Examples of continuous hemodynamic data are provided that compare the classic cardiac reference system with a calculated surrogate of the heart's location. Detailed Implementation
[0046] This disclosure details systems and methods for performing real-time continuous hemodynamic monitoring using sensor systems. These systems and methods improve hemodynamic monitoring by calculating a “cardiac-level correction” for hemodynamic parameters sensed by the sensor system and adjusting the hemodynamic data to compensate for hydrostatic pressure variations caused by the vertical difference between local blood pressure at the sensor site and pressure at the heart. Without correction, these hydrostatic differences can cause significant distortion in blood pressure readings. A hydrostatic difference may correspond to a pressure change of approximately 0.77 mmHg per centimeter of vertical displacement. Traditionally, various methods directly detect the position of the sensor system to compensate for vertical differences. As described herein, by utilizing computational alternatives to correct the sensed hemodynamic parameters to cardiac-level parameters, the need for additional sensors and cables is reduced while still achieving robust hemodynamic monitoring with hydrostatic pressure compensation. The described systems and methods are applicable to any mode employing sensor systems for hemodynamic monitoring because the compensation performed is related to the vertical difference in arterial pressure between the cardiac level and the actual site where sensing is performed.
[0047] Figure 1Provided is an example of a classic cardiac reference sensor used with a hemodynamic monitoring system including a sensing system. Patient 100 monitors their blood pressure via a sensor system 101 attached to the patient's fingers and wrist. As shown, if the vertical level of the patient 100's hand, to which the sensor system 101 is attached, is lower than the vertical level of the heart, pressure will be sensed at a higher level relative to the heart. Conversely, if the vertical level of the patient 100's hand, to which the sensor system 101 is attached, is higher than the vertical level of the heart, pressure will be sensed at a lower level relative to the heart. In this classic cardiac reference sensor (HRS) system, a first sensor 103 is strategically placed at or near the level of the patient's heart, and a second sensor 105 is provided at or near the sensor system 101. The vertical distance between the first sensor 103 and the second sensor 105 can be continuously or intermittently assessed by the sensor system 101 and used to correct for continuous hemodynamic monitoring performed by the sensor system 101. This ensures less variation in hemodynamic monitoring and better downstream applications. For example, continuous hemodynamic monitoring can be displayed on a screen to further calculate hemodynamic parameters (e.g., central blood pressure in the ascending aorta) and provide alerts based on parameter thresholds and trends. Such alerts can include mean arterial pressure (MAP), virtual MAP (vMAP), diastolic / systolic values, and / or deviations of other arterial pressure values. Therefore, reducing variations caused by the vertical orientation of the sensed area leads to greater consistency in hemodynamic monitoring and better patient outcomes.
[0048] In this document, various embodiments of the hemodynamic monitoring system may rely on computational surrogates to correct real-time hemodynamic data related to the relative vertical orientation of the sensing system. Therefore, in some embodiments, the hemodynamic monitoring system excludes the use of a classic cardiac reference sensor system. In some embodiments, the hemodynamic monitoring system excludes the use of paired sensor systems to establish the vertical orientation of the hemodynamic sensing system relative to the patient's heart based on the difference in vertical orientation between each sensor in the paired sensor system. And in some embodiments, the hemodynamic monitoring system excludes the use of one or more sensors positioned at the level of the heart to establish the vertical orientation of the patient's heart. In some embodiments, the sensing system excludes the use of hemodynamic sensors (e.g., central blood pressure sensors) for tracking central hemodynamic parameters, which may be positioned at the patient's center or peripheral location. For example, the sensing system may exclude central blood pressure sensors positioned outside the thoracic region. In some embodiments, the sensing system excludes the combined use of hemodynamic sensors for tracking central hemodynamic data and hemodynamic sensors for tracking peripheral hemodynamic data.
[0049] In some embodiments, the hemodynamic monitoring system includes a motion detection system for capturing motion at the sensing site. In some embodiments, motion can be detected from the captured data itself. For example, a predictive computational model can be trained to use features from the captured data to predict when vertical motion of the sensor system has occurred. In some embodiments, the captured data is used as input within the computational model as a computational substitute for vertical orientation correction. And in some embodiments, the detection of vertical motion of the sensing system triggers a computational step to adjust the vertical position correction amount. For example, when a patient raises or lowers their arm, detected acceleration and inferred orientation changes can serve as real-time signals to update the height bias correction term, even in the absence of absolute orientation measurements.
[0050] Figure 2 An example of a sensing system 12 attached to hand 14 is provided. While sensing is described as being adapted to the fingers and wrist, it should be understood that hemodynamic sensing can be performed at any location capable of having hemodynamic parameters sensed by the sensor system. In some embodiments, the sensor system is located in a non-cardiac-level vertical orientation. In some embodiments, the vertical orientation of the sensor system relative to the cardiac-level vertical orientation is adjustable. The sensor system can be centrally positioned (e.g., at the thoracic level) or peripherally positioned (e.g., in a non-thoracic position). Examples of available peripheral locations include, but are not limited to, the arm, fingers, thumb, wrist, ankle, leg, toes, ear, temple, etc. The sensor system can be invasive (including partially invasive and minimally invasive) or non-invasive.
[0051] like Figure 2 As shown in the example, sensor system 12 is a non-invasive hemodynamic sensor configured to acquire arterial blood pressure measurements via volumetric clamping. Sensor system 12 may include a housing 16, a connector 18, a cuff 20, and a pressure-adjustable bladder 22. In the example shown, cuff 20 is a loop or similar structure that wraps around or surrounds the fingers 24 of hand 14, while housing 16 is a wrist-worn device coupled to cuff 20 via connector 18. However, in the most common case, sensor system 12 may be coupled with... Figure 2 The layouts shown vary considerably. The sensor system 12 may include, for example, multiple separate connectors 18 between elements attached to the finger 24 (e.g., cuff 20), and / or the housing 16 may be repositioned to other locations (e.g., integrated with the cuff 20, or disposed separately in other locations), including the various configurations described herein.
[0052] The cuff 20 can be configured to fit onto various body attachments, including the arm, fingers, thumb, wrist, ankle, leg, toes, ear, temple, etc. In the example shown, the cuff 20 encircles the sensing area of the finger 24 of the hand 14. At least one artery 26 passes through the sensing area. The cuff 20 is also anchored to a pressure-increasing bladder 22, which can be, for example, an inflatable annular fluid bladder supplied by a fluid line included within the connector 18, or supplied from another source. Generally, the fluid used to inflate the bladder can be air. However, in the most general case, the pressure-increasing bladder 22 can be any kind of mechanism suitable for applying pressure to the finger 24 based on controls as described below. The sensor system 12 and the hand 14 together constitute a combined physical system 10 (sometimes referred to as a plant or plant system) responsive to both changes in the patient and changes in the control of the sensor system 12.
[0053] The sensor system 12 also includes a light emitter and a detector that can be disposed on the cuff 20. The light emitter can be configured to emit light of one or more discrete wavelength bands onto an artery within the finger 24, and the light detector can be configured to detect the reflection and refraction of the emitted light. The light emitter and detector system can be used to perform photoplethysmography, pulse oximetry, or other sensing activities to acquire or derive hemodynamic parameters, particularly blood pressure parameters.
[0054] A motion detection system can be used in conjunction with sensor system 12. For example, a camera system (e.g., an IR camera) can be positioned near and pointed at sensor system 12 to record its movement. A marker used to improve the camera system's detection of movement can be incorporated into sensor system 12, enabling the camera to easily track the movement of the marker. In another non-limiting example, the sensor system includes motion detection sensors (such as accelerometers, velocity sensors, gyroscope sensors, or any other sensors configured to detect their own movement) and can be attached to an accessory. A motion detector can be associated with or incorporated into the sensing system at or near the sensing location, enabling it to detect movement at that location. Return to Reference Figure 2 The motion detector can be integrated into the sensor system 12, cuff 20, connector 18, housing 16, or attached independently to an accessory. Alternatively, the motion detection system may lack a camera system and / or motion detection sensor, instead predicting motion from captured hemodynamic data.
[0055] For example, an accelerometer may be incorporated into and / or otherwise associated with sensor system 12 to detect changes in the orientation or position of a patient's appendages (e.g., hand, fingers) relative to the level of the heart. Such changes may be caused by movement or displacement of the patient in bed (e.g., tilting and / or rotating the bed during surgery or recovery), which may alter the hydrostatic pressure gradient between the heart and the measurement site. Upon detecting a change in the orientation and / or height of the appendages, the system may initiate a correction to adjust the measured blood pressure to reflect the pressure at the level of the heart.
[0056] In some embodiments, the system continuously acquires signals from a motion detector (e.g., an accelerometer, a gyroscope). The system may also acquire hemodynamic data, such as systolic blood pressure, diastolic blood pressure, mean arterial pressure (MAP), and / or derived characteristics, such as maximum dP / dt. The system can use this data to estimate the possible height difference between the pressure sensing site and the anatomical level of the heart. When the motion detector detects motion, the system can infer that a change in orientation has occurred and use the motion data as a gating signal to trigger a recalibration and / or correction of the MAP value.
[0057] The system can calculate virtual mean arterial pressure (vMAP). vMAP can be generated from a model, such as a trained model. The model can use heart rate, one or more cuff pressure values (e.g., systolic pressure, diastolic pressure, MAP, etc.), the difference between said values (e.g., systolic pressure minus diastolic pressure), the slope of the derived values (e.g., the slope of pressure over time (e.g., maximum dP / dt)), and / or other values. vMAP can be generated for each beat. In some embodiments, vMAP can represent a cleaned, smoothed MAP estimate across multiple beats. Additionally or alternatively, the system can determine the standard deviation of vMAP for each beat. The system can reject noisy and / or unreliable beats. In response to a detected motion event, the system can update the virtual heart reference sensor (vHRS) based on the vMAP value. Between motion events, vHRS remains static and is used to maintain an invariant relationship, such as vMAP + offset = cMAP + vHRS, where cMAP is the measured cuff MAP, and the offset term takes into account a previously known bias. The following reference to sMAP corresponds to cMAP.
[0058] In some embodiments, the system may incorporate a machine learning model trained to detect persistent changes in MAP that are not solely attributable to physiological variability. For example, if the system observes a consistent increase or decrease in blood pressure after detecting a change in orientation, it can learn (e.g., update the model) to associate that pattern with an upward or downward shift of the measurement site relative to the level of the heart. In some cases, without motion detection data, such changes may be misinterpreted as genuine changes in the patient's condition. Therefore, the motion detector can refine the trained model to attribute such variations to hydrostatic pressure drift and update the model accordingly.
[0059] In some embodiments, the system uses acceleration and / or rotation data to detect changes in the sensor's orientation relative to gravity. As an example, the system can identify changes in the x-axis and / or y-axis accelerometer signals to determine that the sensor (and clamp) has rotated. For instance, when an operating table is rotated or tilted, the accelerometer can detect corresponding changes in the signal relationship between the axes, indicating that the height of the pressure measurement site (e.g., the patient's hand) relative to heart level has changed. This rotational cue can act as a trigger to initiate algorithmic correction. In some embodiments, in the absence of detectable orientation drift from the accelerometer, the correction algorithm can remain gated closed, thereby reducing the risk of introducing noise and / or overcorrection due to physiological variations, patient-specific blood pressure drift, and / or sensor artifacts.
[0060] The system may incorporate other safety measures, such as user-initiated overrides and / or contextual filters. For example, when a known disturbance—such as the administration of a fluid bolus (which may result in a reasonable and sudden change in blood pressure and / or movement)—occurs, the user can press a button to signal the system to temporarily ignore algorithmic corrections. This temporary override may help avoid misinterpreting the event as altitude-induced pressure drift. Additionally or alternatively, the system may include an interface button that allows the virtual MAP to be recalibrated to cardiac levels to conform to values confirmed by a clinician. In some embodiments, the system may calculate and / or display quality metrics (e.g., vHRS confidence scores) derived from the variability and / or conformity between the measured vMAP and predictions from multiple trained models. For example, if the vMAP for a given beat falls outside the defined standard deviation range relative to a trained ensemble, the system may reject the data and suspend any updates to the correction factor.
[0061] Sensor system 12 is intended as a non-limiting example, and other configurations may be used. The sensor can be any sensor capable of acquiring hemodynamic data (especially blood pressure parameters) such that it can acquire or derive continuous hemodynamic pressure waveforms. The sensor system can be non-invasive, minimally invasive, partially invasive, or fully invasive. Examples of sensors and their applications include (but are not limited to): applanation tonometers for performing arterial tension measurements; pressure cuffs and photoplethysmography devices for performing volume clamping; peripheral arterial lines for performing catheter-based hemodynamic monitoring; internal pressure transducers for performing internal hemodynamic monitoring; one or more electrode leads and pulse oximeters for performing pulse wave transit time or pulse arrival time; photoplethysmography devices for performing heart rate monitoring via photoplethysmography; and capacitive pressure sensors for performing arterial pressure sensing via capacitance. In some embodiments, the sensor system is a device intended for clinical use within specialized healthcare facilities, such as catheter-based systems, internal systems, or Figure 2 Examples are depicted in the diagram. In some embodiments, the sensor system is a portable device designed for use in any environment, particularly outside of professional healthcare facilities (e.g., at home). In some embodiments, the sensor system is a wearable device designed for daily use, mobile use, or use during exercise. The sensor system may incorporate motion detectors, such as accelerometers or gyroscopes, to detect changes in orientation relative to gravity and thus determine changes in hydrostatic pressure due to vertical displacement of the sensing site. Furthermore, the motion detection system can be used with any sensor system, but particularly those capable of adjusting vertical height relative to heart level.
[0062] Figure 3 Provided is an example of a method for performing continuous hemodynamic measurements with vertical orientation correction relative to the heart level. Method 300 may begin by performing (301) continuous hemodynamic measurements on a recipient via a sensor system. Any sensor system for capturing or deriving hemodynamic parameters may be utilized. The sensor may be non-invasive (e.g., a blood pressure cuff) or invasive (e.g., an internal pressure transducer). Continuous hemodynamic measurements may be any hemodynamic measurement that is susceptible to gravity, such as (e.g.) blood pressure, systolic arterial pressure, diastolic arterial pressure, and mean arterial pressure. Almost all pressure-related hemodynamic parameters are affected by gravity, which may result in the need to correct measurements acquired at a non-heart-level vertical orientation for greater consistency. Such correction may be based on the determination of a vertical offset between the heart and the measurement site. The vertical offset may be calculated in response to orientation and / or motion data collected by the sensor system.
[0063] Any site that can be used for sensing and is in a different vertical orientation than the heart benefits from calibration, especially at sites where vertical movement is possible. Even minute movements, such as lifting an attachment or adjusting a hospital bed, can affect hemodynamic sensing. For example, changes in x-axis and / or y-axis accelerometer data can indicate that the measurement site has moved relative to gravity. This determination can trigger adjustments to the measured blood pressure. Therefore, the method can be performed at any central or peripheral site, including (but not limited to) within an artery, on the arm, on the finger, on the thumb, on the wrist, on the ankle, on the leg, on the toes, on the ear, on the temple, etc.
[0064] In many implementations, performing continuous hemodynamic measurements results in the capture of waveforms (e.g., arterial pressure waveforms), which can be displayed on a monitor. In many implementations, various other hemodynamic parameters are derived from the results of continuous hemodynamic measurements. Therefore, the vertical position of the sensor system or its vertical movement can affect various downstream calculations, visualizations, and evaluations. The system can calculate virtual mean arterial pressure (vMAP), smoothed MAP estimates, and / or deviations associated with measurements taken during movement.
[0065] Method 300 may optionally detect or predict (303) vertical movement at the site of sensing. Vertical movement can be detected by monitoring the movement of the sensor system using various mechanisms, such as a camera system or an attached motion detector. In some embodiments, motion detector signals are acquired continuously. Changes in acceleration and / or rotation can be used as gating events to initiate a recalibration of the measured pressure values. Alternatively, since changes in vertical orientation can lead to changes in the captured hemodynamic measurements, the captured hemodynamic data can be used as input within a predictive computational model to detect movement.
[0066] Method 300 can also correct (305) continuous hemodynamic measurements based on the vertical orientation of the sensing site. In some embodiments, a correction amount is determined in real time based on the sensed hemodynamic measurements, and this correction amount can then be used to correct the hemodynamic measurements for the vertical orientation. Furthermore, when vertical movement of the sensing site is detected, the correction amount can be adjusted to ensure that appropriate correction of the hemodynamic measurements for the vertical orientation is maintained. In some cases, computational correction of continuous hemodynamic measurements serves as an alternative to a cardiac reference sensor system. In some cases, computational correction of continuous hemodynamic measurements is performed instead of using a cardiac reference sensor system.
[0067] Figure 4The present invention provides a calculation method for calibrating continuous hemodynamic monitoring parameters. The calculation method 400 can be executed using a computing device equipped with instructions for performing the method. The computing device may be part of a hemodynamic monitoring system. The computing device may be connected to a hemodynamic measurement sensing system and, optionally, a motion detection system, enabling it to receive data from said systems in real time. The computing device can execute the steps of the calculation method 400 in real time, allowing it to continuously calibrate the hemodynamic parameters acquired by the sensing system. The calibrated hemodynamic parameters can be displayed on a screen for further downstream calculations and / or for providing real-time medical alerts, thereby enhancing the utility of hemodynamic monitoring using the hemodynamic measurement sensing system. In some embodiments, the system may additionally or alternatively calculate and / or display confidence scores for the calibrated values (e.g., vMAP, vHRS). The confidence scores can be derived from the variability of the sensed signal and / or conformity with the output of a trained model.
[0068] The computational method 400 can receive (401) sensor signals from a sensor system and optionally from a motion detector system. As described herein, any of a variety of sensor systems for acquiring hemodynamic parameters can be utilized. Sensors can be non-invasive or invasive; central (e.g., thoracic) or peripheral (e.g., non-thoracic). Generally, the sensor system can acquire sensor data such as capacitance-based pressure measurements, photoplethysmography, tension measurements, optical signals, and electronic signals, each of which can be utilized by various methods to generate hemodynamic data. Furthermore, the motion detector system can optionally be used to detect the occurrence of vertical motion. Signals from the motion detector system can be acquired via a camera system or a motion detector. For example, an accelerometer can be used to detect the orientation and / or height drift of the patient's appendages relative to the level of the heart, including changes caused by movement or tilting of the operating table. Generally, the motion detector system can be used to identify when the sensing site changes vertical orientation. Alternatively, the motion detector system can be a computation-based system that detects motion from acquired or derived hemodynamic measurements.
[0069] The calculation method 400 can also determine (403) hemodynamic parameters from the sensor system. Various methods of hemodynamic monitoring can continuously derive hemodynamic parameters using the acquired signals. For example, the volume clamp method uses optical plethysmography via optical signal data to determine the arterial volume within the appendage at the sensing site. Since each heartbeat causes the arterial volume to expand and contract, corresponding to systolic and diastolic pressures, a blood pressure cuff can be used in conjunction with an optical plethysmometer to continuously determine arterial pressure. For this purpose, the blood pressure cuff is continuously expanded and collapsed with fluid, so that the arterial volume within the appendage remains constant. When the arterial volume remains constant, the pressure within the cuff matches the arterial systolic and diastolic pressures, thereby providing an arterial pressure reading from the sensor system. In some embodiments, derived values, such as maximum dP / dt and / or other values disclosed herein, can be calculated / generated from beat-by-beat signal analysis. Further description of examples of various systems and methods for generating hemodynamic parameters from various sensor systems is provided in X. Quan et al. Sensors (Basel). 2021 Jun 22;21(13):4273 (the contents of which are incorporated herein by reference).
[0070] Calculation method 400 can also calculate (405) a correction to produce hemodynamic parameters that take into account the vertical orientation of the sensor system. Hemodynamic parameters acquired by the sensor system are affected by the relative vertical orientation. More specifically, blood pressure at the sensing site is partly determined by hydrostatic pressure. The effect of hydrostatic pressure is smaller when the sensor system is at a higher vertical orientation and larger when it is at a lower vertical orientation. To address these differences, it is ideal to standardize the pressure readings to a single vertical orientation, thereby reducing variability and improving downstream assessment and calculations. Any vertical orientation can be selected for standardization. In many cases, the goal is to standardize the hemodynamic parameters to the level of the heart.
[0071] In several implementations, one or more computational models are trained to predict corrected hemodynamic parameters (e.g., corrected pressure parameters from sensed hemodynamic parameters). To perform training, classical cardiac reference sensors can be used to obtain the corrected hemodynamic parameters. Data prior to correction (i.e., sensed hemodynamic data) is used as input within the computational model and correlated with the corrected hemodynamic parameters obtained using classical cardiac reference sensors. The computational model can be trained to predict the corrected hemodynamic parameters from the input data. The model can be updated in response to determining that motion has occurred. For example, the model can learn to associate upward and / or downward shifts with associated motion patterns. Otherwise, the system might incorrectly associate such shifts with actual changes in the patient's state. Additionally or alternatively, motion signals from motion detectors can be used as gating triggers for data recalibration and / or model updates. Any predictive computational model used to predict corrected hemodynamic parameters from sensed sensor hemodynamic data can be utilized, such as neural networks and linear regression.
[0072] Computational models can be used to predict any of a variety of outputs, such as, for example, corrected blood pressure, corrected systolic arterial pressure, corrected diastolic arterial pressure, or corrected mean arterial pressure. Furthermore, several sensed hemodynamic data features can be used as inputs, such as, for example, sensed blood pressure, features derived from sensed hemodynamic waveforms, sensed stroke volume, sensed pulse wave transit time, sensed pulse arrival time, sensed systolic arterial pressure, sensed diastolic arterial pressure, sensed mean arterial pressure, sensed dP / dt, sensed systolic / diastolic pressure ratio, etc. In some implementations, the input features are multinomial features, especially when using regression models. To improve predictive capabilities, the input features should be robust and insensitive to perturbations that may occur during continuous hemodynamic monitoring. For example, noisy and / or anomalous beats can be automatically rejected if they exceed a model-based beat-by-beat variability threshold and / or if they fall outside an alignment threshold with the ensemble prediction (e.g., exceeding one or two standard deviations). According to several embodiments, the computational model is designed to predict a single output or a set of defined outputs using several feature inputs. In some embodiments, several different computational models are trained individually and then combined to produce an ensemble model that generates a combined output (which may further produce standard deviations and other statistical metrics).
[0073] In an illustrative example, the computational model uses polynomial input features to predict the corrected mean arterial pressure (vMAP, where v represents a virtual prediction). Polynomial input features can include one or more of the following: heart rate, sensed systolic blood pressure, sensed diastolic blood pressure, sensed mean arterial pressure, sensed dP / dt, and sensed systolic / diastolic blood pressure ratio. Additionally or alternatively, the system can determine the standard deviation of the vMAP for each beat. The standard deviation can be based on values measured for a set number of beats. The system can reject noisy and / or unreliable beats that fall outside the standard deviation (or some other threshold).
[0074] To calculate the correction, an equation with invariant input can be used. In some implementations, the following equation is used: Equation 1 in These are virtual predictions of hemodynamic parameters derived from computational models. These are the sensed hemodynamic parameters, and This is the correction value (i.e., the correction value calculated virtually). It's important to note that the predicted hemodynamic parameters and the sensed hemodynamic parameters are the same hemodynamic parameters (e.g., both are mean arterial pressure). Therefore, it can be achieved by using... Offset and To calculate The offset can initially be set to any number (e.g., zero). Once calculated... , , and This allows for offset correction, as the offset remains constant. In some implementations, a low-pass filter (such as an exponential filter) is used to correct the offset, removing noise and disturbances that may occur during continuous hemodynamic monitoring. Calculated... This is used to correct the sensed hemodynamic data. For example, the arterial pressure waveform sensed by the sensor system can be used... Correction.
[0075] Returning to the illustrative example of calculating vMAP using a computational model, the following equations can be used: Equation 2 in It is a virtual predicted mean arterial pressure predicted from a computational model. It is the sensed mean arterial pressure, and It is the correction amount (i.e., the correction amount for virtual calculation).
[0076] When motion is detected, calculation method 400 may optionally update (407) the correction amount. As previously mentioned, the correction amount can remain unchanged when the sensing site remains stationary. However, when motion is detected, the correction amount can be adjusted to take into account the new vertical orientation of the sensing site. To update the correction amount, it is recalculated, which can be done using Equation 1 or Equation 2. In some embodiments, the calculation of the correction amount is gated after motion is detected. The gate may be based on a time period (e.g., 10 seconds) or a certain number of heartbeats (e.g., 10 heartbeats). Once the gate period ends, the calculated correction amount can remain unchanged until motion is detected again. For an example of a gate period, see [link to example]. Figure 5 And the accompanying description. In some implementations, motion events can trigger updates to the vHRS based on newly calculated vMAP to improve the stability of post-repositioning corrections.
[0077] In some implementations, motion is detected via a motion detection system, which may be a camera system, an attached motion detector, or a computational alternative. In some implementations using a motion detection system, a threshold may be used to determine whether motion requires a system update of the correction amount. In one illustrative example, the change in gravity caused by motion can be sensed (e.g., via a motion detector associated with a sensor system) or otherwise derived, and the process is triggered to update the correction amount when the change in gravity exceeds a threshold. In another illustrative example, vertical motion can be monitored (e.g., via a camera system) or otherwise derived, and the process is triggered to update the correction amount when the vertical motion exceeds a threshold.
[0078] In some implementations, the motion detection system uses sensed hemodynamic data as input to detect motion via a computational surrogate. When a change in the sensed hemodynamic parameters is identified (e.g., an overall increase or decrease in pressure waveform data), a computational model can be used to determine whether this change in the sensed hemodynamic parameters is due to repositioning of the sensing site or to other factors (e.g., cardiac events). For example, a neural network can be trained to detect vertical motion based on features derived from the sensed hemodynamic data. In some implementations, the sensed hemodynamic data features can be based on data trends over the immediate preceding time period, such as (e.g.) the immediate preceding 10 seconds or the immediate preceding 10 heartbeats.
[0079] In some implementations, the motion detection system directly determines the vertical orientation or changes in vertical orientation, which are then used to calculate the correction amount. In some such implementations, the motion detection system determines the vertical orientation, which may be determined relative to heart level. Based on the relative vertical orientation, the correction amount can be determined. In some such implementations, the motion detection system determines changes in vertical orientation, which can be used to determine adjustments to the correction amount. The system may initiate correction only in response to motion indicating a change in the hydrostatic gradient exceeding a threshold amount. When adjusting the correction amount, the amount of adjustment to the correction amount can be determined using the motion detection system or, as described herein, using Equation 1 or Equation 2.
[0080] The calculation method 400 can also continuously correct (409) the continuously sensed hemodynamic monitoring parameters. In some embodiments, the sensed hemodynamic monitoring parameters are corrected to the cardiac level. Continuous correction can employ any periodicity. In some embodiments, the sensed hemodynamic monitoring parameters are corrected with each heartbeat. In some embodiments, the correction amount is gated (i.e., (It cannot be greater than the upper threshold or less than the lower threshold). For example, if a sensor system is adapted to a finger, the maximum possible distance from the heart is typically less than 40 cm, so a threshold can be set accordingly. In some cases, when multiple computational models are used to determine... When, and when the standard deviation is greater than the threshold, an alternative is used. Such as (for example) previously calculated Other safety measures can also be adopted.
[0081] Figure 5 The example provided is for correcting sensed hemodynamic data. In this example, a blood pressure cuff, including an expandable balloon and an optical plethysmography device, is used to continuously capture arterial blood pressure. Figure 5 The graph above shows raw blood pressure data captured by the blood pressure cuff, including systolic arterial pressure (SYS), mean arterial pressure (MAP), and diastolic arterial pressure (DIA). The two horizontal dashed lines represent mean systolic pressure and mean diastolic pressure, respectively. At time 14:42:15, it can be seen that SYS, MAP, and DIA are rising, indicating that the blood pressure cuff is sensing higher blood pressure. Figure 5 As shown in the figure below, motion was detected at time 14:42:15, and the resulting increase in blood pressure was due to the vertical repositioning of the sensing site. The new vertical orientation caused the sensed pressure to remain elevated, thus requiring adjustment of the correction amount for the sensed blood pressure. This observed increase could be related to motion data detected by motion detection systems (e.g., accelerometers or gyroscopes) due to changes in the orientation or position of the sensing sensor relative to the heart level.
[0082] Figure 5The vertical dashed line in the diagram represents the gating period for recalculating the correction amount. Therefore, after the gating section, the correction amount remains unchanged. Figure 5 The figure below shows two separately determined correction values. The darker shaded line represents the correction value determined by the classical cardiac reference sensor (HRS) method (see [reference]). Figure 1 The lighter shaded lines represent the correction amount determined by the calculated substitute (vCorr). It can be seen that both the classic HRS and the substitute vCorr are adjusted downwards, resulting in a correction that reduces the sensed pressure. In some implementations, this gating is initiated based on orientation changes detected from the motion detection system (which may correspond to changes in vertical orientation).
[0083] Figure 5 The intermediate graph provides corrected blood pressure at the heart level. The darker shaded line represents sensed blood pressure corrected by the classic HRS system. The lighter shaded line represents blood pressure corrected by a computational substitute. This demonstrates that the computational substitute detects motion and adjusts the correction accordingly during the gating period, resulting in corrected blood pressure—highly representative of HRS-corrected blood pressure.
[0084] Although a specific example of a process for correcting sensed hemodynamic parameters has been described above, those skilled in the art will understand that the steps of this process can be performed in different orders, and that some steps may be optional depending on the various embodiments. Therefore, it should be understood that the individual steps of the process can be used appropriately according to the requirements of a specific application. Furthermore, any of a variety of processes for correcting sensed hemodynamic parameters suitable for the requirements of a given application can be employed in various embodiments.
[0085] Hemodynamic monitoring system The various systems and methods disclosed herein can be implemented or performed using a hemodynamic monitoring system. Generally, a hemodynamic monitoring system includes a sensing system and a computing system. Figure 6AThe provided example is a hemodynamic monitoring system 600 to be used as an accessory to an individual. The accessory includes a sensor 620 capable of capturing sensor signals for calculating hemodynamic parameters. It is noteworthy that the sensor 620 can be any sensor used to capture hemodynamic measurements and can be positioned at any location suitable for that sensor. For example, the sensor 620 may include a photoplethysmography (PPG) sensor, a pressure transducer, and / or an accelerometer. The hemodynamic monitoring system 600 can be a clinical system designed for use in professional healthcare settings, a portable system designed for use in any environment (including environments outside of professional healthcare settings, such as at home), or a wearable device designed for daily use, mobile use, or use during exercise. It is thus readily understood that various configurations of the hemodynamic monitoring system are considered suitable for their intended use.
[0086] Sensor 620 can be connected to hemodynamic monitoring system 600, allowing data to be transferred between them. Other systems suitable for sensing via sensor 620 can also be utilized, such as, for example, fluid pump systems, electrocardiograms, and / or power supplies. Sensor 620 can be connected via any means for transferring data, such as a wired connection via cable or wire or a wireless connection via Bluetooth, WiFi, etc.
[0087] The hemodynamic monitoring system 600 may include a computing system, which includes a processor system 602 and an I / O interface 604, a display screen, and a user interface for inputting and outputting data, such as data transferred between the hemodynamic monitoring system 600 and the sensor 620. It is readily understood that the processor system 602, the I / O interface 604, and the memory system 606 may be implemented using any of a variety of components suited to the requirements of a particular application, including (but not limited to) a CPU, GPU, ISP, DSP, wireless modem (e.g., Wi-Fi, Bluetooth modem), serial interface, volatile memory (e.g., DRAM), and / or non-volatile memory (e.g., SRAM and / or NAND flash memory).
[0088] Memory system 606 is capable of storing various data, applications, and models. It should be understood that the listed data, applications, and models are representative samples that can be stored in memory, and various memory systems can store some or all of the listed data, applications, and models. Furthermore, any combination of data, applications, and models can be stored, and in some embodiments, various data, applications, and / or models are temporarily stored. This data may include hemodynamic parameter data (e.g., blood pressure, heart rate), raw sensor data, motion data, historical trends, and metadata (e.g., timestamps, patient identifiers).
[0089] The hemodynamic monitoring system 600 can utilize multiple applications stored in a memory system 606, which are to be executed by a processor system 602 to perform a set of instructions that can perform various computational methods as described herein. Applications that can be stored in the memory system 606 include a real-time hemodynamic monitoring application 608, a motion detection application 610, and a hemodynamic data correction application 612, for performing the various methods and processes described herein. The various applications can be provided as individual processes or as an integration of processes, each of which can be used to provide real-time, continuous hemodynamic monitoring. Real-time hemodynamic data (e.g., corrected and / or sensed data) can also optionally be stored on its storage system 606 and / or displayed on a display screen via an I / O interface 604. The motion detection application 610 can utilize accelerometer and / or gyroscope data to identify repositioning events. The hemodynamic data correction application 612 can perform the vCorr algorithm and / or reference-based adjustments (e.g., via virtual and / or physical cardiac reference sensors). Data can be visualized (e.g., on local and / or remote displays) and / or transmitted to a remote computing system via I / O interface 604.
[0090] Despite the above reference Figure 6A Specific hemodynamic monitoring system configurations have been described; however, it should be readily understood that various hemodynamic monitoring systems used in the provision of hemodynamic monitoring can be implemented in any of these configurations. Therefore, the various systems and methods described herein should not be limited to any particular hemodynamic monitoring system configuration, but rather can be implemented using any kind of hemodynamic monitoring system configuration capable of sensing hemodynamic measurements.
[0091] Figure 6B An example hemodynamic monitoring system 630 integrating both pressure sensing and motion detection according to some embodiments is shown. The hemodynamic monitoring system 630 can correct for hydrostatic pressure drift caused by changes in attachment orientation and / or height. The hemodynamic monitoring system 630 can combine signal inputs from a volume clamp 632 and an accelerometer 634. The hemodynamic monitoring system 630 can implement algorithmic corrections to produce a physiologically representative pressure output. The hemodynamic monitoring system 630 may include one or more features of the sensor system 101 and / or the hemodynamic monitoring system 600.
[0092] The hemodynamic monitoring system 630 may include a volume clamp 632, which may be placed on a patient's attachment (e.g., finger, arm, toe, leg) to acquire continuous pressure waveforms. In addition to or as a substitute for the raw pressure measurements, the volume clamp 632 may generate data for pulsation detection 636. The pulsation detection 636 data may correspond to the chronometry and / or morphology of the cardiac cycle of the patient's attachment. These pulsation detections may be used to derive additional hemodynamic features, such as virtual mean arterial pressure (vMAP) 640 and / or a stable virtual heart reference sensor (vHRS). The volume clamp 632 may be part of and / or include one or more features of the cuff 20 and / or the compressible balloon 22 described above. vMAP may be calculated using a trained machine learning model that receives one or more variables—such as heart rate, cuff pressure value, and / or pressure variation over time (e.g., maximum dP / dt)—as input.
[0093] Accelerometer 634 can be positioned in association with the clamp to generate a motion signal associated with volume clamp 632 (e.g., attached to the same attachment). Accelerometer 634 detects changes in the orientation, tilt, or height of the attachment. The signal from accelerometer 634 is transmitted to motion detector 638, which determines when a significant motion event (e.g., tilting of the operating table, raising of an arm) has occurred, potentially introducing a hydrostatic gradient between the measurement site and the heart level. When such an event is detected, motion detector 638 signals post-processing unit 642. Accelerometer 634 may be part of and / or include one or more features of the motion detection sensor described above (e.g., of sensing system 12).
[0094] System 630 may include a post-processing module 642. Post-processing 642 may track vMAP over time and / or may use vMAP as input. Additionally or alternatively, post-processing 642 may include correction logic for vMAP. Post-processing 642 may receive input from one or more of pulsation detector 636 and / or motion detector 638. Based on one or more of these inputs, the system may calculate vMAP 640, taking into account the height of the appendage above the heart, which may include an estimated patient mean arterial pressure. Additionally or alternatively, the system may estimate a virtual heart reference sensor (vHRS). vHRS may represent the expected hydrostatic bias due to elevation or descent of the measurement site. In some embodiments, vHRS may remain static between motion events and be updated only after repositioning has been verified, thereby reducing noise from physiological drift and / or transient anomalousness.
[0095] Waveform correction module 644 can receive one or more pressure waveforms from volume clamp 632 and / or vHRS signals generated by post-processing 642. Based on one or more of these inputs, module 644 applies a height-based correction to the original pressure waveform, which can result in recalibration of the signal to reflect cardiac-level pressure. This correction step can improve the accuracy of pressure readings to better correspond to true physiological pressure, even in response to displacement of the patient's attachments. The corrected pressure waveform can then be output at 646 as a refined signal for display and / or further processing. Output 646 may additionally or alternatively include the correction factors / bias amounts described herein. Output 646 can be displayed on a display connected to system 630 via a wired and / or wireless data interface.
[0096] Figure 6C An example method 660 is illustrated for correcting blood pressure measurements based on pulsation level processing and motion detection according to some embodiments. Method 660 can be implemented by a hemodynamic monitoring system, such as hemodynamic monitoring system 630, sensor system 101, and / or hemodynamic monitoring system 600. As shown, the method can operate continuously or intermittently—for example, in response to a new cardiac pulsation identified from pressure waveform data acquired via a volume clamp placed on a patient accessory.
[0097] At box 662, the system senses a new heartbeat. The system may sense the new heartbeat, for example, in response to a beat detection module that identifies the start of the new heartbeat from a pressure waveform generated by volume clamp. At box 664, the system estimates the vMAP of the new beat. After compensating for the height of the appendage located above (or below) the heart, the vMAP may correspond to an estimate of the patient's actual MAP. In some embodiments, this vMAP may be generated by post-processing logic (e.g., post-processing 642) using one or more models, such as machine learning models.
[0098] At box 666, the system determines whether to reject the vMAP value of the pulsation. For example, the system can compare the newly estimated vMAP with a statistical norm, previous pulsations, or an ensemble of predictions from a trained model. If the vMAP falls outside a defined confidence interval (e.g., exceeding a variability threshold), the system can determine that the pulsation is noisy and / or simply unreliable and reject the vMAP value. If the system determines to reject the vMAP (path “true”), at box 668, the system does not take corrective action on the pulsation, and the method waits for the next sensed pulsation. In this way, noisy data is gated out to maintain the stability and reliability of downstream computations.
[0099] If vMAP is not rejected (path “false”), the method proceeds to box 670, where the system determines whether a motion event has occurred. This determination may be based on motion signals received from accelerometers and / or other motion detection modules (e.g., accelerometer 634, motion detector 638) that detect changes in the orientation, tilt, and / or height of the pressure sensor relative to gravity. For example, the system may infer from the motion signals that the patient’s arm has moved up or down, thus indicating a possible change in hydrostatic pressure relative to the level of the heart.
[0100] If motion is detected (path "true"), at box 672, the system updates the virtual heart reference sensor (vHRS). This vHRS can reflect a height offset based on the sensor's new orientation and can be used to adjust the measured pressure to reflect the pressure at the heart level. If no motion is detected (path "false"), method 660 proceeds to box 674, where the system updates an offset term. The offset term can include one or more corrections learned from a machine learning model. For example, this offset update can account for drift observed in the vMAP that is not attributable to motion and can be used to adjust for long-term signal trends. For example, the system can refine its internal model to distinguish physiological changes from significant changes due to slow sensor migration and / or calibration drift. In some embodiments, the system can continuously repeat method 660, using incoming beats to update stable and motion-aware pressure estimates in real time.
[0101] Experimental data Figures 7A to 7C Examples of correcting sensed data using computational substitutes according to the various systems and methods described herein are provided. In each figure, the top figure depicts sensed arterial blood pressure obtained via a plethysmography using a pressure cuff and optical plethysmography. The bottom figure depicts the motion detected at the sensing site, the correction amount via a classical cardiac reference sensor, the correction amount via a computational substitute, and the offset. The middle figure shows the corrected sensed data, comparing the correction amount via the HRS system (darker shaded line) with the correction amount via a computational substitute (lighter shaded line). These data show that the correction amount produced by the computational substitute is very comparable to that of the classical HRS system. This data demonstrates that sensed hemodynamic parameters can be successfully corrected via computational substitutes using the systems and methods of this disclosure.
[0102] Example Terms 1. A method for real-time correction of sensed hemodynamic data, the method comprising: A hemodynamic monitoring system is used to receive sensor signals from a sensor system at the sensing site; Using the hemodynamic monitoring system, a set of hemodynamic parameters derived from the sensor signals are determined; Using the hemodynamic monitoring system, a correction amount was calculated that took into account the vertical orientation of the sensors in the sensor system; and Using the hemodynamic monitoring system, hemodynamic parameters are continuously corrected by taking into account the vertical orientation of the sensor.
[0103] 2. The method according to Example 1, wherein calculating the correction amount taking into account the vertical orientation of the sensor system using the hemodynamic monitoring system further comprises: Using the hemodynamic monitoring system, a set of hemodynamic parameters are input into a trained computational model to generate corrected hemodynamic parameters; and Using the hemodynamic monitoring system, the correction amount, taking into account the vertical orientation of the sensor system, is determined using the corrected hemodynamic parameters within the invariant input equation, wherein the invariant input includes the corrected hemodynamic parameters and the sensed hemodynamic parameters, wherein the corrected hemodynamic parameters and the sensed hemodynamic parameters correspond to the same hemodynamic parameters.
[0104] 3. According to the method described in Example 2, the invariant input equation is: vHP + offset = sHP + vCorr Where vHP is the corrected hemodynamic parameter, sHP is the sensed hemodynamic parameter, and vCorr is the correction amount, and vHP and psHP are the same hemodynamic parameter.
[0105] 4. The method described in Example 3, wherein vHP is vMAP and sHP is sMAP.
[0106] 5. The method according to Example 1, wherein calculating the correction amount taking into account the vertical orientation of the sensor system further comprises: Using the hemodynamic monitoring system, signals from the motion detector are received; and Using the hemodynamic monitoring system and based on the motion detector signal, the correction amount is determined to correspond to the change in the vertical orientation of the sensor relative to the level of the heart.
[0107] 6. The method according to any one of Examples 1 to 5, further comprising: Using the hemodynamic monitoring system, the vertical motion of the sensor system corresponding to changes in its vertical orientation is identified; and Using the hemodynamic monitoring system, the correction value for the vertical orientation of the sensor system was updated.
[0108] 7. The method according to Example 6, wherein updating the correction amount that takes into account the vertical orientation of the sensor system includes: Using the hemodynamic monitoring system, updated sets of hemodynamic parameters are input into a trained computational model to generate updated, corrected hemodynamic parameters; and Using the hemodynamic monitoring system, the correction amount taking into account the vertical orientation of the sensor system is determined using the updated corrected hemodynamic parameters within the invariant input equation, wherein the invariant input includes the updated corrected hemodynamic parameters and the updated sensed hemodynamic parameters, wherein the updated corrected hemodynamic parameters and the updated sensed hemodynamic parameters correspond to the same hemodynamic parameters.
[0109] 8. The method according to Example 6 or 7, wherein identifying the motion of the sensor system comprises: Using the hemodynamic monitoring system, a motion detector signal is received, wherein the motion detector signal indicates that the amount of vertical motion is greater than a threshold.
[0110] 9. The method according to Example 8, wherein the motion detector signal indicates a change in gravity greater than a threshold.
[0111] 10. The method according to Example 8 or 9, wherein the motion detector signal is derived from a motion detector associated with the sensor system.
[0112] 11. The method according to Example 6 or 7, wherein identifying the motion of the sensor system comprises: using the hemodynamic monitoring system, detecting the vertical motion of the sensor system from a computational model, the computational model being trained to detect changes in sensed hemodynamic parameters in response to vertical repositioning of the sensor system.
[0113] 12. The method according to Example 6, wherein updating the correction amount that takes into account the vertical orientation of the sensor system includes: Using the hemodynamic monitoring system, signals from the motion detector are received; and Using the hemodynamic monitoring system and based on the motion detector signal, the correction amount is determined to correspond to the change in the vertical orientation of the sensor relative to the level of the heart.
[0114] 13. The method according to any one of Examples 1 to 12, wherein the sensor system comprises at least one of the following: a flattening arterial tension meter, a pressure cuff, an optical plethysmometer, a peripheral artery line, one or more electrode leads, a pulse oximeter, or a capacitance-based pressure sensor.
[0115] 14. The method according to any one of Examples 1 to 13, wherein determining the group of hemodynamic parameters comprises: Perform arterial tension measurement, perform volume clamping, perform catheter-based hemodynamic monitoring, perform pulse wave transit time or pulse arrival time, perform heart rate monitoring based on photoplethysmography, or perform arterial pressure sensing via capacitance.
[0116] 15. The method according to any one of Examples 1 to 14, further comprising: Using the hemodynamic monitoring system, downstream hemodynamic parameters are calculated from the calibrated sensed hemodynamic parameters.
[0117] 16. The method according to Example 15, wherein the downstream hemodynamic parameters include central hemodynamic parameters.
[0118] 17. The method according to Example 16, wherein the central hemodynamic parameter includes blood pressure in the patient's ascending aorta.
[0119] 18. The method according to any one of Examples 1 to 17, further comprising: Using the hemodynamic monitoring system, waveforms are generated based on the calibrated sensed hemodynamic parameters; or The hemodynamic monitoring system is used to generate waveforms using downstream hemodynamic parameters.
[0120] 19. The method according to Example 18, further comprising: The waveform is displayed on the monitor of the hemodynamic monitoring system.
[0121] 20. The method according to any one of Examples 1 to 19, wherein the hemodynamic monitoring system comprises at least one of: a clinically used system, a portable system, or a wearable device.
[0122] 21. The method according to any one of Examples 1 to 20, wherein the hemodynamic monitoring system excludes a paired sensor system for determining the vertical orientation of the sensing system relative to the heart based on the difference in vertical orientation between each sensor of the paired sensor system to calculate the correction amount.
[0123] 22. The method according to any one of Examples 1 to 21, wherein the hemodynamic monitoring system excludes one or more sensors positioned at the level of the heart to establish the vertical orientation of the heart in order to calculate the correction amount.
[0124] 23. The method according to any one of Examples 1 to 22, wherein the hemodynamic monitoring system excludes the use of a central blood pressure sensor positioned outside the thoracic region to calculate the correction amount.
[0125] According to any one of Examples 1 to 23, the motion detector includes at least one of an accelerometer or a gyroscope configured to detect azimuth drift of the sensor system.
[0126] 24. The method according to Example 6 or 7, wherein identifying the motion of the sensor system using the hemodynamic monitoring system includes: Using the hemodynamic monitoring system, a motion detector signal is received, wherein the motion detector signal indicates that the amount of vertical motion is less than a threshold; and Based on the fact that the amount of vertical motion indicated by the signal is less than the threshold, it is determined that the motion detector signal does not correspond to the change in the vertical orientation of the sensor system.
[0127] 25. The method according to any one of Examples 1 to 24, wherein determining the group of sensed hemodynamic parameters includes calculating the mean arterial pressure across multiple pulsations.
[0128] 26. The method according to Example 8, wherein the calculation of the correction amount taking into account the vertical orientation of the sensor system is based on receiving a motion detector signal that indicates vertical motion is greater than the threshold.
[0129] 27. The method according to any one of Examples 1 to 25, further comprising: It is confirmed that the takeover function has been executed; and In response to determining that the takeover function has been executed, the continuous correction of the sensed hemodynamic parameters is delayed.
[0130] 28. The method according to Example 27, wherein determining that the takeover function is performed includes determining to make a manual takeover selection.
[0131] 29. The method according to Example 28, wherein determining to make the manual takeover selection includes selecting the OK button.
[0132] 30. The method according to Example 28 further includes calculating a confidence score for the correction amount.
[0133] 31. A hemodynamic monitoring system for real-time correction of sensed hemodynamic data, the system comprising: Sensor system, which includes sensors; and A computing processing system digitally connected to the sensor system, the computing processing system comprising: Processor system; and A memory system comprising one or more application programs capable of directing the processor system: Receive sensor signals acquired from the sensor system at the sensing location; Determine a set of hemodynamic parameters derived from the sensor signals; The calculation takes into account the vertical orientation correction of the sensor system; and The sensed hemodynamic parameters are continuously corrected using the correction amount that takes into account the vertical orientation of the sensor.
[0134] 32. The hemodynamic monitoring system according to Example 31, wherein one or more applications are capable of further guiding the processor system: The group of sensed hemodynamic parameters are input into a trained computational model to generate corrected hemodynamic parameters; and The correction amount, taking into account the vertical orientation of the sensor system, is determined using the corrected hemodynamic parameters within the invariant input equation, wherein the invariant input includes the corrected hemodynamic parameters and the sensed hemodynamic parameters, wherein the corrected hemodynamic parameters and the sensed hemodynamic parameters correspond to the same hemodynamic parameters.
[0135] 33. The hemodynamic monitoring system according to Example 32, wherein the invariant input equation is: vHP + offset = sHP + vCorr Where vHP is the corrected hemodynamic parameter, sHP is the sensed hemodynamic parameter, and vCorr is the correction amount, and vHP and sHP are the same hemodynamic parameter.
[0136] 34. The hemodynamic monitoring system according to Example 33, wherein vHP is vMAP and sHP is sMAP.
[0137] 35. The hemodynamic monitoring system according to Example 31, wherein one or more applications are capable of further guiding the processor system: Receive motion detector signals, wherein the motion detector signals indicate the vertical orientation of the sensor system relative to the level of the heart; and The correction amount is determined using the vertical orientation of the sensor system relative to the level of the heart.
[0138] 36. The hemodynamic monitoring system according to any one of Examples 31 to 35, wherein one or more applications are capable of further guiding the processor system: Identify the motion of the sensor system corresponding to changes in its vertical orientation; and The update takes into account the correction amount for the vertical orientation of the sensor system.
[0139] 37. The hemodynamic monitoring system according to Example 36, wherein one or more applications are capable of further guiding the processor system: The updated set of sensed hemodynamic parameters is input into a trained computational model configured to generate updated, corrected hemodynamic parameters; and The correction amount taking into account the vertical orientation of the sensor system is determined using the updated corrected hemodynamic parameters in the input equation, wherein the input includes the updated corrected hemodynamic parameters and the updated sensed hemodynamic parameters, wherein the updated corrected hemodynamic parameters and the updated sensed hemodynamic parameters are the same hemodynamic parameters.
[0140] 38. The hemodynamic monitoring system according to Example 36 or 37, further comprising a motion detector, wherein one or more applications are capable of further guiding the processor system: Receive a motion detector signal, wherein the motion detector signal indicates that the amount of vertical motion is greater than a threshold.
[0141] 39. The hemodynamic monitoring system according to Example 31, wherein the motion detector signal indicates a change in gravity greater than a threshold.
[0142] 40. The hemodynamic monitoring system according to Example 38 or 39, wherein the motion detector is associated with the sensor system.
[0143] 41. The hemodynamic monitoring system according to Example 39 or 40, wherein one or more applications are capable of further guiding the processor system: The vertical motion of the sensor system is detected from a computational model trained to detect changes in sensed hemodynamic parameters in response to the vertical repositioning of the sensor system.
[0144] 42. The hemodynamic monitoring system according to Example 29, further comprising a motion detector, wherein one or more applications are capable of further guiding the processor system: Receive motion detector signals, wherein the motion detector signals indicate changes in the vertical orientation of the sensor system relative to the level of the heart; and The correction amount is determined using the change in the vertical orientation of the sensor system relative to the level of the heart.
[0145] 43. The hemodynamic monitoring system according to any one of Examples 31 to 42, wherein the sensor system comprises at least one of: a flattening arterial tension meter, a pressure cuff, an optical plethysmometer, a peripheral arterial line, one or more electrode leads, a pulse oximeter, or a capacitance-based pressure sensor.
[0146] 44. The hemodynamic monitoring system according to any one of Examples 31 to 43, wherein the group of sensed hemodynamic parameters is determined by at least one of: arterial tension measurement, volume clamping, catheter-based hemodynamic monitoring, pulse wave transit time or pulse arrival time, heart rate monitoring based on photoplethysmography, or arterial pressure sensing via capacitance.
[0147] 45. The hemodynamic monitoring system according to any one of Examples 31 to 44, wherein one or more applications are capable of further guiding the processor system: Downstream hemodynamic parameters are calculated from the corrected sensed hemodynamic parameters.
[0148] 46. The hemodynamic monitoring system according to Example 45, wherein the downstream hemodynamic parameter is a central hemodynamic parameter.
[0149] 47. The hemodynamic monitoring system according to Example 46, wherein the central hemodynamic parameter is blood pressure in the patient's ascending aorta.
[0150] 48. The hemodynamic monitoring system according to any one of Examples 31 to 47, wherein one or more applications are capable of further guiding the processor system: Waveforms are generated using the corrected sensed hemodynamic parameters; or Waveforms are generated using downstream hemodynamic parameters.
[0151] 49. The hemodynamic monitoring system according to Example 48, further comprising a display, wherein one or more applications are capable of further guiding the processor system: The waveform is displayed on the monitor.
[0152] 50. The hemodynamic monitoring system according to any one of Examples 31 to 49, wherein the hemodynamic monitoring system is at least one of: a clinically used system, a portable system, or a wearable device.
[0153] 51. A hemodynamic monitoring system according to any one of Examples 31 to 50, wherein the hemodynamic monitoring system excludes a paired sensor system for determining the vertical orientation of the sensing system relative to the heart based on the difference in vertical orientation between each sensor of the paired sensor system to calculate the correction amount.
[0154] 52. The hemodynamic monitoring system according to any one of Examples 31 to 51, wherein the hemodynamic monitoring system excludes one or more sensors positioned at the level of the heart to establish the vertical orientation of the heart in order to calculate the correction amount.
[0155] 53. The hemodynamic monitoring system according to any one of Examples 31 to 52, wherein the hemodynamic monitoring system excludes the use of a central blood pressure sensor positioned outside the thoracic region to calculate the correction amount.
Claims
1. A hemodynamic monitoring system for real-time correction of ambient sensed blood pressure data, the system comprising: Display screen; The first sensor system includes a blood pressure sensor configured to attach to a patient's wrist or hand; The second sensor system includes a motion sensor configured to detect the displacement of the first sensor system in the vertical direction; and A computing processing system digitally connected to the first sensor system and the second sensor system, the computing processing system comprising: Processor system; and A memory system comprising one or more application programs capable of directing the processor system: Receive the BP waveform signal acquired from the first sensor system; Determine a set of hemodynamic parameters derived from the BP waveform signal, including the sensed blood pressure parameters; The group of sensed hemodynamic parameters are input into a trained computational model to generate corrected blood pressure parameters that take into account changes in the vertical orientation of the first sensor system. The motion of the first sensor system is detected using the second sensor, the motion having an amplitude above a threshold along the vertical direction; Based on the detected motion of the first sensor system, the sensed hemodynamic parameters are corrected using the corrected blood pressure parameters; and The corrected blood pressure parameters are displayed on the screen.
2. The system of claim 1, wherein the sensed blood pressure parameter and the corrected blood pressure parameter are the same parameter and are one of the following: mean arterial pressure (MAP), systolic blood pressure, and diastolic blood pressure.
3. The system according to claim 1 or 2, wherein the memory system is further configured to adjust the BP waveform signal based on the corrected blood pressure parameters to correct for the change in the vertical orientation of the first sensor system.
4. A method for real-time correction of sensed hemodynamic data, the method comprising: A hemodynamic monitoring system is used to receive sensor signals from a sensor system at the sensing site; Using the hemodynamic monitoring system, a set of hemodynamic parameters derived from the sensor signals are determined; Using the hemodynamic monitoring system, a correction amount was calculated that took into account the vertical orientation of the sensors in the sensor system; and Using the hemodynamic monitoring system, hemodynamic parameters are continuously corrected by the correction amount that takes into account the vertical orientation of the sensor.
5. The method of claim 4, wherein calculating the correction amount taking into account the vertical orientation of the sensor system using the hemodynamic monitoring system further comprises: Using the hemodynamic monitoring system, a set of hemodynamic parameters are input into a trained computational model to generate corrected hemodynamic parameters; and Using the hemodynamic monitoring system, the correction amount, taking into account the vertical orientation of the sensor system, is determined using the corrected hemodynamic parameters within the invariant input equation, wherein the invariant input includes the corrected hemodynamic parameters and the sensed hemodynamic parameters, wherein the corrected hemodynamic parameters and the sensed hemodynamic parameters correspond to the same hemodynamic parameters.
6. The method according to claim 5, wherein the invariant input equation is: in These are the corrected hemodynamic parameters. These are the sensed hemodynamic parameters, and It is the correction amount, and where and They are the same hemodynamic parameters.
7. The method of claim 6, wherein yes ,and yes .
8. The method of claim 4, wherein calculating the correction amount taking into account the vertical orientation of the sensor system further comprises: The hemodynamic monitoring system is used to receive motion detector signals; and Using the hemodynamic monitoring system and based on the motion detector signal, the correction amount is determined to correspond to the change in the vertical orientation of the sensor relative to the level of the heart.
9. The method of claim 8, further comprising: Using the hemodynamic monitoring system, the vertical motion of the sensor system is identified in relation to changes in the vertical orientation of the sensor system. and Using the hemodynamic monitoring system, the correction value for the vertical orientation of the sensor system was updated.
10. The method of claim 9, wherein updating the correction amount that takes into account the vertical orientation of the sensor system comprises: Using the hemodynamic monitoring system, updated sets of hemodynamic parameters are input into a trained computational model to generate updated, corrected hemodynamic parameters. and Using the hemodynamic monitoring system, the correction amount taking into account the vertical orientation of the sensor system is determined using the updated corrected hemodynamic parameters within the invariant input equation, wherein the invariant input includes the updated corrected hemodynamic parameters and the updated sensed hemodynamic parameters, wherein the updated corrected hemodynamic parameters and the updated sensed hemodynamic parameters correspond to the same hemodynamic parameters.
11. The method of claim 9 or 10, wherein identifying the motion of the sensor system comprises: Using the hemodynamic monitoring system, a motion detector signal is received, wherein the motion detector signal indicates that the amount of vertical motion is greater than a threshold.
12. The method of claim 9 or 10, wherein identifying the motion of the sensor system comprises: Using the hemodynamic monitoring system, the vertical motion of the sensor system is detected from a computational model trained to detect changes in sensed hemodynamic parameters in response to the vertical repositioning of the sensor system.
13. The method according to any one of claims 4 to 12, wherein determining the group of hemodynamic parameters comprises: Perform arterial tension measurement, perform volume clamping, perform catheter-based hemodynamic monitoring, perform pulse wave transit time or pulse arrival time, perform heart rate monitoring based on photoplethysmography, or perform arterial pressure sensing via capacitance.
14. The method according to any one of claims 4 to 12, further comprising: Using the hemodynamic monitoring system, waveforms are generated using the calibrated sensed hemodynamic parameters; or The hemodynamic monitoring system is used to generate waveforms using downstream hemodynamic parameters.
