System and method for computer-aided measurement of lung capillary wedge pressure

Abstract

Systems and methods for computer-aided analysis of pulmonary capillary wedge pressure (PCWP) measurement acquisition in an individual are provided. Various systems and methods measure wedge pressure via a lung catheter and determine the quality of the wedge pressure measurements. In some cases, the systems and methods utilize a trained computational model to assess PCWP quality. The systems and methods also involve determining transitions between wedge and non-wedge positions, which may identify such transitions based on one or more haemodynamic features using fuzzy logic.

Description

[0001] Cross - Reference to Related Applications

[0002] This application claims priority to U.S. Provisional Patent Application No. 63 / 389,331, filed Jul. 14, 2022, by Al Hatib et al. and entitled “Systems and Methods for Machine Learning Enabled Measurement of Pulmonary Capillary Wedge Pressure” and U.S. Provisional Patent Application No. 63 / 486,914, filed Feb. 24, 2023, by Angel et al. and entitled “Systems and Methods for Computer Enabled Measurement of Pulmonary Capillary Wedge Pressure”; the disclosures of which are incorporated herein by reference in their entireties. Technical Field

[0003] The present invention generally relates to systems and methods for computer - assisted acquisition and analysis of pulmonary capillary wedge pressure (PCWP). Background Art

[0004] Left atrial pressure is an important measurement for individuals with left ventricular dysfunction and / or valvular disease. While direct methods for determining left atrial pressure are possible, these methods (such as transseptal access) have inherent risks. Thus, indirect methods are commonly used, including using a pulmonary artery catheter (PA catheter) (also known as a Swan - Ganz catheter) to determine the pulmonary capillary wedge pressure (PCWP), also known as the pulmonary artery occlusion pressure (PAOP).

[0005] Although PCWP measurement is considered a low - risk procedure, accurate interpretation of pulmonary artery pressure (PAP) and PCWP waveforms generally requires an expert. Thus, there is a need for methods and systems that can help ensure accurate acquisition and analysis of PAP and PCWP measurements. Summary of the Invention

[0006] In some embodiments, a computational method is used to perform a pulmonary capillary wedge pressure (PCWP) measurement of an individual and to evaluate its quality. The method includes acquiring a blood pressure waveform of the individual, where the waveform includes blood pressure measurements acquired from a pulmonary artery location and blood pressure measurements acquired from a wedge location. The blood pressure waveform is acquired using a pulmonary artery catheter. The method includes determining a PCWP measurement using a computational processing system based on the blood pressure measurements from the wedge location. The method includes using a computational processing system to determine a quality assessment for the PCWP measurement using a machine learning model configured to provide a quality assessment for the PCWP measurement based on the blood pressure measurements acquired from the pulmonary artery location and the blood pressure measurements acquired from the wedge location.

[0007] In some embodiments, a hemodynamic monitoring system is used to perform a pulmonary capillary wedge pressure (PCWP) measurement and to evaluate its quality. The system includes a pulmonary artery catheter configured to acquire blood pressure measurements from a pulmonary artery location and blood pressure measurements from a wedge location. The pulmonary artery catheter includes an inflatable balloon. The system includes a computational processing system connected to the pulmonary artery catheter. The computational system includes a processor system, a display screen digitally connected to the processor system, and a memory including one or more applications. One or more applications may direct the processor to acquire blood pressure measurements at the pulmonary artery location. One or more applications may direct the processor to inflate the balloon. Inflating the balloon allows the catheter to migrate to the wedge location. One or more applications may direct the processor to acquire blood pressure measurements at the wedge location. One or more applications may direct the processor to generate a blood pressure waveform based on the blood pressure measurements acquired from the pulmonary artery location and the blood pressure measurements acquired from the wedge location. One or more applications may direct the processor to determine a PCWP measurement based on the blood pressure measurements from the wedge location. One or more applications may direct the processor to use a machine learning model to determine a quality assessment for the PCWP measurement, the machine learning model being configured to provide a quality assessment for the PCWP measurement based on the blood pressure measurements acquired from the pulmonary artery location and the blood pressure measurements acquired from the wedge location. One or more applications may direct the processor to display on the display screen one or more of the following: the blood pressure waveform, the PCWP measurement, or the quality assessment for the PCWP measurement.

[0008] In some embodiments, the blood pressure waveform is generated and displayed on the display screen in real time.

[0009] In some embodiments, the PCWP measurement is determined and displayed on the display screen in real time.

[0010] In some embodiments, the quality assessment for the PCWP measurement is determined and displayed on the display screen in real time.

[0011] In some embodiments, a computational method for performing a pulmonary capillary wedge pressure (PCWP) measurement of an individual. The method includes collecting a blood pressure waveform of the individual using a pulmonary artery catheter. The waveform includes blood pressure measurements collected from a pulmonary artery location. The pulmonary artery catheter is connected to a hemodynamic monitoring system. The method includes, while collecting blood pressure measurements from the pulmonary artery location, using the hemodynamic monitoring system to evaluate the blood pressure measurements collected from the pulmonary artery location for artifacts. The method includes, after determining that the blood pressure measurements collected from the pulmonary artery location are free of artifacts, using the hemodynamic monitoring system to inflate a balloon at or near the distal end of the pulmonary artery catheter to allow the pulmonary artery catheter to migrate to a wedge position. The method further includes collecting a blood pressure waveform of the individual using the pulmonary artery catheter, wherein the waveform includes blood pressure measurements collected from the wedge position. The method includes using the hemodynamic monitoring system to determine a PCWP measurement based on the blood pressure measurements from the wedge position.

[0012] In some embodiments, a hemodynamic monitoring system for performing a pulmonary capillary wedge pressure (PCWP) measurement. The system includes a pulmonary artery catheter configured to collect blood pressure measurements from a pulmonary artery location and blood pressure measurements from a wedge position. The pulmonary artery catheter includes an inflatable balloon. The system includes a computational processing system connected to the pulmonary artery catheter. The computational system includes a processor system, a display screen digitally connected to the processor system, and a memory system including one or more applications. The one or more applications may direct the processor system to collect blood pressure measurements in the pulmonary artery location. The one or more applications may direct the processor system to evaluate the blood pressure measurements collected from the pulmonary artery location for artifacts. The one or more applications may direct the processor system to inflate the balloon after determining that the blood pressure measurements collected from the pulmonary artery location are free of artifacts, wherein inflating the balloon allows the catheter to migrate to the wedge position. The one or more applications may direct the processor system to collect blood pressure measurements in the wedge position. The one or more applications may direct the processor system to generate a blood pressure waveform based on the blood pressure measurements from the pulmonary artery location and the blood pressure measurements from the wedge position. The one or more applications may direct the processor system to determine a PCWP measurement based on the blood pressure measurements from the wedge position. The one or more applications may direct the processor system to display the PCWP measurement on the display screen.

[0013] In some embodiments, a computational method for detecting a transition between a pulmonary artery position and a wedge position. The method includes acquiring a blood pressure waveform of an individual using a pulmonary artery catheter. The waveform includes blood pressure measurements acquired from the pulmonary artery position. The pulmonary artery catheter is connected to a hemodynamic monitoring system. The method includes inflating a balloon at or near the distal end of the pulmonary artery catheter of the hemodynamic monitoring system to allow the pulmonary artery catheter to migrate to the wedge position. The method includes using the hemodynamic monitoring system to detect the transition from the pulmonary artery position to the wedge position. The method further includes acquiring a blood pressure waveform of the individual using the pulmonary artery catheter. The waveform includes blood pressure measurements acquired from the wedge position. The method includes deflating the balloon at or near the distal end of the pulmonary artery catheter using the hemodynamic monitoring system to allow the pulmonary artery catheter to migrate back to the pulmonary artery position. The method includes using the hemodynamic monitoring system to detect the transition from the wedge position to the pulmonary artery position.

[0014] In some embodiments, a hemodynamic monitoring system for detecting a transition between a pulmonary artery position and a wedge position. The system includes a pulmonary artery catheter configured to acquire blood pressure measurements acquired from the pulmonary artery position and blood pressure measurements acquired from the wedge position. The pulmonary artery catheter includes an inflatable balloon. The system includes a computational processing system connected to the pulmonary artery catheter. The computational processing system includes a processor system, a display screen digitally connected to the processor system, and a memory system including one or more applications. One or more applications may direct the processor system to acquire blood pressure measurements in the pulmonary artery position. One or more applications may direct the processor system to inflate the balloon. Inflating the balloon allows the catheter to migrate to the wedge position. One or more applications may direct the processor system to detect the transition from the pulmonary artery position to the wedge position. One or more applications may direct the processor system to acquire blood pressure measurements in the wedge position. One or more applications may direct the processor system to deflate the balloon at or near the distal end of the pulmonary artery catheter to allow the pulmonary artery catheter to migrate back to the pulmonary artery position. One or more applications may direct the processor system to detect the transition from the wedge position to the pulmonary artery position.

[0015] In some embodiments, PCWP measurements and quality assessments are each determined in real time.

[0016] In some embodiments, the computational processing system and the pulmonary artery catheter are part of a hemodynamic monitoring system.

[0017] In some embodiments, acquiring the blood pressure waveform includes inserting the pulmonary artery catheter into the central vein of an individual. Acquiring the blood pressure waveform includes guiding the pulmonary artery catheter to the pulmonary artery. Acquiring the blood pressure waveform includes inflating a balloon at or near the distal end of the pulmonary artery catheter to allow the pulmonary artery catheter to migrate to the wedge position.

[0018] In some embodiments, the method includes, prior to inflating the balloon, using a computational processing system to evaluate, in real time, blood pressure measurements acquired from a pulmonary artery location for artifacts.

[0019] In some embodiments, an artifact includes one or more of the following: a measurement obtained during patient movement, a measurement obtained during catheter flushing, and a measurement indicating a flat line, a measurement indicating underdamping, and a measurement indicating overdamping.

[0020] In some embodiments, if the PAP maximum value (PAP max ) is greater than a threshold, if the PAP minimum value (PAP min ) is less than a threshold, and / or if the PAP maximum value minus the PAP minimum value (PAP max - PAP min ) is greater than a threshold, then a detection of an artifact based on flushing, improper zeroing, or patient movement is detected; where PAP is a blood pressure measurement acquired from a pulmonary artery location.

[0021] In some embodiments, if the PAP maximum value minus the PAP minimum value (PAP max - PAP min ) is less than a threshold, then a detection of an artifact based on a flat line or overdamping is detected; where PAP is a blood pressure measurement acquired from a pulmonary artery location.

[0022] In some embodiments, the pulmonary artery catheter includes a lumen configured to measure blood pressure.

[0023] In some embodiments, the pulmonary artery catheter is a Swan - Ganz catheter.

[0024] In some embodiments, the machine - learning model is based on at least one feature selected from the group consisting of: waveform phase features, determinable features, and morphological features. The waveform phase features are selected from: contractility, pulmonary artery compliance, stroke volume, vascular tone, afterload, and a complete cardiac cycle. The determinable features are determined based on the waveform phase features and are selected from: phase average, maximum value, minimum value, duration, area, standard deviation, slope, derivative, trend, deviation from the trend, variability, and variance. The morphological features are based on the waveform features and are selected from: frequency components, skewness, and kurtosis.

[0025] In some embodiments, the machine - learning model is based on one or more of the following: approximate entropy, sample entropy, baroreflex sensitivity, variability, and change.

[0026] In some embodiments, acquiring a PCWP measurement includes determining an individual's respiratory cycle. Determining the PCWP measurement is based on the individual's respiratory cycle.

[0027] In some embodiments, determination of the PCWP includes determining a blood pressure measurement from a wedge position at the end of expiration of the respiratory cycle.

[0028] In some embodiments, determination of the PCWP includes averaging blood pressure measurements from a wedge position over one or more respiratory cycles.

[0029] In some embodiments, the method includes using a computing processing system to detect in real time a transition from a pulmonary artery position to a wedge position or from a wedge position to a pulmonary artery position by: extracting one or more hemodynamic features from blood pressure measurements in a pulmonary position and blood pressure measurements in a wedge position; and for each of the one or more hemodynamic features, using a fuzzy logic membership function to determine a fuzzy logic value based on a change in value between the blood pressure measurements in the pulmonary position and the blood pressure measurements in the wedge position.

[0030] In some embodiments, the one or more features include at least one of the following: statistical moments, percentile values of data, histograms of data, diastolic blood pressure, mean pressure, average pressure, systolic blood pressure, pulse pressure, Shannon entropy, number of peaks above a percentile, number of valleys below a percentile, average number of crossings, area under the curve, singular vector coefficients after principal component analysis, subsampled signals, determinable features, waveform phase features, and morphological features.

[0031] In some embodiments, the morphological features are selected from: frequency components, skewness, and kurtosis; wherein the waveform phase features are selected from: contractility, pulmonary artery compliance, stroke volume, vascular tone, afterload, and a complete cardiac cycle; and wherein the determinable features are selected from: phase average, maximum value, minimum value, duration, area, standard deviation, slope, derivative, trend, deviation from the trend, variation, and variance.

[0032] In some embodiments, the one or more features include average pressure and pulse pressure.

[0033] In some embodiments, the method further includes segmenting blood pressure measurements collected from a pulmonary artery position and blood pressure measurements collected from a wedge position into time windows. The method further includes extracting features of the blood pressure measurements collected from a pulmonary artery position and the blood pressure measurements collected from a wedge position from the time windows. A machine learning model is trained to detect whether the extracted features of the time windows are derived from blood pressure measurements collected from a pulmonary artery position or from a wedge position. Using a computing processing system, a quality assessment for the PCWP measurement is determined using the machine learning model, including inputting the extracted features from the time windows into the machine learning model to produce a quality assessment for each time window. The quality assessment is based on whether the extracted features for each time window can be distinguished as being derived from a pulmonary artery position or from a wedge position.

[0034] In some embodiments, the quality assessment is categorical.

[0035] In some embodiments, the categorization is a qualitative ranking.

[0036] In some embodiments, using a computing processing system, determining a quality assessment for a PCWP measurement using a machine learning model includes determining whether one or more extracted features of a time window are above or below a threshold.

[0037] In some embodiments, one or more of the extracted features include: PCWP Mean , PAP Diastolic and PCWP PulsePress , where PCWP Mean is the average of pressure measurements taken in the wedge position; where PAP Diastolic is the diastolic pressure in the pulmonary artery position; and where PCWP PulsePress is the pulse pressure in the wedge position.

[0038] In some embodiments, high quality is indicated when: PCWP Mean < PAP Diastolic and a × PCWP PulsePress + b × PCWP Mean ≤ c, where a, b, and c are decisive values.

[0039] In some embodiments, medium quality is indicated when: PCWP Mean < PAP Diastolic and a × PCWP PulsePress + b × PCWP Mean > c, where a, b, and c are decisive values.

[0040] In some embodiments, low quality is indicated when: PCWP Mean ≥ PAP Diastolic .

[0041] In some embodiments, the method further includes using a computing processing system to assign its quality assessment to each time window.

[0042] In some embodiments, the method further includes displaying the quality assessment of one or more time windows on a display screen in digital communication with the computing processing system.

[0043] In some embodiments, determining a PCWP measurement using a computing processing system includes averaging blood pressure measurements taken from the wedge position for time windows determined to have a quality above a threshold.

[0044] In some embodiments, determining a PCWP measurement using a computing processing system includes excluding blood pressure measurements taken from a wedge position during a time window determined to have a quality below a threshold, and averaging the non-excluded blood pressure measurements taken from the wedge position during the time window.

[0045] In some embodiments, the method further includes displaying, on a display screen in digital communication with the computing processing system, a quality assessment of one or more time windows.

[0046] Additional embodiments and features are set forth in part in the description below, and will in part be obvious to those of ordinary skill in the art upon examination of the specification, or may be learned by practice of the present disclosure. A further understanding of the nature and advantages of the present disclosure can be realized by reference to the remaining portions of the specification and drawings that form a part of the present disclosure. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] The description will be more fully understood with reference to the following drawings, which are exemplary embodiments of the invention and should not be construed as a complete statement of the scope of the invention, wherein:

[0048] Figure 1 An example of a catheter located at a wedge position is provided.

[0049] Figures 2A - 2D Examples of PA catheter positions within the heart and the pressures obtainable by the PA catheter at these positions are provided.

[0050] Figures 3A - 3B Examples of proper ( Figure 2A ) wedge placement and of a wedge position in the case of an over-inflated balloon ( Figure 2B ) are provided.

[0051] Figures 3C - 3D Examples of waveforms during PCWP measurement are provided.

[0052] Figures 4A - 4B Examples of methods and computational processes for acquiring PWCP measurements and assessing their quality are provided.

[0053] Figure 5 Examples of computational methods for detecting artifacts in a PAP waveform are provided.

[0054] Figure 6 Examples of computational methods for determining transition points for balloon inflation or balloon deflation are provided.

[0055] Figures 7A - 7E Examples of plotting using fuzzy logic membership functions are provided.

[0056] Figure 8Examples of methods for generating fuzzy logic membership functions are provided.

[0057] Figure 9 Examples of clinical data for a group of patients are provided, and the relationship between pulse pressure and mean pressure is plotted.

[0058] Figure 10 Examples of computational methods for performing PCWP measurements and quality assessments are provided.

[0059] Figures 11A - 11B Examples of quality metrics are provided.

[0060] Figure 12 Examples of plotting quality assessments on wedge position waveforms are provided.

[0061] Figures 13A - 13B Examples of pressure waves are provided.

[0062] Figures 14A - 14F Examples of various parameters derived from pressure waves are provided.

[0063] Figure 15 Examples of computational processing systems are provided.

[0064] Figure 16 Examples of hemodynamic monitoring systems are provided. Detailed implementation

[0065] Turning now to the drawings, systems and methods for computer-aided acquisition and analysis of pulmonary capillary wedge pressure (PCWP) are described. In many cases, a hemodynamic monitoring system including a pulmonary artery catheter system (e.g., a Swan-Ganz catheter) is utilized to acquire PCWP. Computational methods can be performed to assist in and analyze the acquisition of PCWP. In some cases, computational methods are utilized to detect when a pulmonary artery catheter is ready to acquire PCWP (e.g., ready to inflate a balloon on the end of the pulmonary catheter). In some cases, computational methods are utilized to detect the balloon inflation point and / or detect the balloon deflation point. In some cases, computational methods are utilized to detect respiratory signals and / or perform PCWP measurements. In some cases, a trained computational model is utilized to evaluate the quality of the acquisition of PCWP. In some cases, a computational system is configured to run one or more of the described computational methods. The computational system can be part of a hemodynamic monitoring system or utilized in conjunction with a hemodynamic monitoring system.

[0066] PCWP is often used to evaluate left ventricular filling, represent left atrial pressure, and assess mitral valve function. In many cases, PCWP is also an estimate of left ventricular end-diastolic pressure (LVEDP). Physiologically normal PCWP is between 4 mmHg and 12 mmHg, and elevated levels of PCWP may indicate severe left ventricular failure or severe mitral stenosis.

[0067] To measure PCWP, a multi-lumen catheter with a balloon at its tip (such as a Swan-Ganz catheter) can be inserted through a central vein (such as the femoral vein, subclavian vein, internal jugular vein, or any other suitable vein) and advanced through the superior vena cava or inferior vena cava to reach the right atrium. The catheter can be advanced from the right atrium through the tricuspid valve into the right ventricle. Once in the right ventricle, the balloon at the tip of the catheter is inflated, advancing the catheter into the right ventricular outflow tract and then across the pulmonary valve to reach the pulmonary artery. Once the catheter enters the main pulmonary artery, the balloon is deflated. Then, the tip of the catheter is located in the main pulmonary artery, where the balloon can be re-inflated as needed, which allows the catheter to migrate downstream (e.g., through the circulatory system in the direction of blood flow), where the balloon can occlude a branch of the pulmonary artery, the "wedged" position. Then, the catheter can provide a measurement of PCWP, which should be equal to the pressure in the left atrium.

[0068] Figure 1 An example diagram of a pulmonary artery catheter in the wedged position is provided. The inflated balloon 102 occludes a branch of the pulmonary artery 104 to allow one or more ports located on the catheter 106 to obtain pressure measurements. The pressure measurements can be single measurements, continuous measurements, sustained measurements, and combinations thereof.

[0069] As the catheter is advanced through the chambers of the heart and / or circulatory system, various blood pressure waveforms can be obtained and / or specific measurements can be allowed within that chamber, blood vessel, or other circulatory component. Figures 2A - 2D An example is provided showing the progression of waveforms of the catheter passing through the path from the right atrium ( Figure 2A ) to the wedged position ( Figure 2D ). Figure 2A The pressure obtained in the right atrium is shown, where the physiologically normal pressure is approximately 2 - 6 mmHg and the average value is 4 mmHg. Waveform 202 shows an exemplary ECG of an individual during the PCWP procedure, while waveform 204 provides an exemplary right atrial pressure waveform measured by the catheter 206 (such as a Swan-Ganz catheter). Peaks a, c, and v represent atrial contraction, backward bulging after tricuspid valve closure, and atrial filling during ventricular contraction, respectively.

[0070] At Figure 2BIn [the context], the catheter 206 is advanced into the right ventricle, which results in a change in the pattern of the pressure waveform 204. Since the catheter 206 is in the right ventricle, the right ventricular systolic pressure (RVSP) and the right ventricular diastolic pressure (RVDP) can be measured, where the physiologically normal RVSP is approximately 15 - 25 mmHg, and the physiologically normal RVDP is approximately 0 - 8 mmHg.

[0071] Figure 2C An example of the pressure waveform 204 as the catheter 206 is advanced into the pulmonary artery is shown, where the pulmonary artery systolic pressure (PASP), the pulmonary artery diastolic pressure (PADP), and the mean pulmonary artery pressure (MPA) can be measured, where the physiologically normal PASP is approximately 15 - 25 mmHg, the physiologically normal PADP is approximately 8 - 15 mmHg, and the physiologically normal MPA is approximately 10 - 20 mmHg.

[0072] Figure 2D An example of the pressure waveform 204 generated at the wedge position to produce the PCWP (also known as the pulmonary artery occlusion pressure (PAOP)) is shown. As previously mentioned, the PCWP is equivalent to the left atrial pressure, where the physiologically normal PCWP is approximately 6 - 12 mmHg. Additionally, the pressure waveform 204 indicates peaks a and v, representing atrial contraction and atrial filling during ventricular contraction, respectively.

[0073] Abnormal waveforms can be caused by a variety of reasons, especially incorrect catheter placement. For example, being placed too close to the right ventricle, being placed too far from the right ventricle, and / or the catheter being unable to migrate from the pulmonary artery to the wedge position. Improper inflation of the balloon (including over - inflation and under - inflation) may result in additional abnormalities. Non - clinicians (such as nurses, doctors, internists, surgeons, and / or other medical practitioners) may not be able to recognize abnormal waveforms, which may lead to incorrect measurement interpretation or abandonment of the procedure altogether.

[0074] Figure 3A and Figure 3B Examples of pressure waveforms with correct and incorrect PCWP measurements are provided. Specifically, Figure 3A a proper wedge is shown, where the waveform indicates appropriate a and v waves, such as Figure 2D shown. However, Figure 3B depicts a waveform generated by an over - inflated balloon, resulting in an inaccurate PCWP waveform (indicated by the gradual rise of the waveform), with less distinct a and v peaks.

[0075] Figure 3C and Figure 3D Examples of pressure waveforms generated according to the PCWP procedure are provided. Figure 3C Examples of accurately acquired waveforms are provided in [reference], while Figure 3D in [reference] an inaccurately acquired waveform is shown.

[0076] In Figure 3C it, waveform 300 shows the pressure measurements obtained by a healthcare provider starting at time t 0 and ending at time t f . Region 302 represents the acquisition of pressure and the resulting PCWP waveform when the catheter is in the wedge position, while region 304 represents the acquisition of pressure when the catheter is in the pulmonary artery position (e.g., before and after PCWP). In contrast, Figure 3D shows a situation where waveform 300 has damped measurements in region 306. This damping may be due to improper placement of the catheter or other problems during acquisition. To overcome these challenges, the hemodynamic monitoring systems described herein can utilize one or more computational processes to ensure proper wedge pressure readings. These systems and methods can be deployed in healthcare facilities to allow clinicians (e.g., doctors, internists, nurses, cardiologists, etc.) to acquire PCWP readings in real time and evaluate their quality. In some embodiments, the hemodynamic monitoring system can indicate to the clinician the quality and / or whether the catheter needs to be repositioned and the measurement reacquired.

[0077] Figure 4A An example of a method 400 for performing PCWP measurements is provided in. Generally, the method includes using a pulmonary artery catheter (e.g., a Swan-Ganz catheter) delivered to the pulmonary artery to perform PCWP measurements. The pulmonary artery catheter can include one or more sensors, such as (e.g.) a pressure sensor, a thermal sensor (e.g., for measuring cardiac output via thermodilution), and optical fibers (e.g., photometric or other optical measurements). The pulmonary artery catheter can also include a balloon near the distal end for performing the wedge method within the pulmonary artery. The pulmonary artery catheter can be utilized in conjunction with a hemodynamic monitoring system to measure blood pressure, cardiac output, and various other hemodynamic measurements in real time. The monitoring system can also include one or more computational programs to assist in monitoring hemodynamic parameters and / or performing various measurements (e.g., PCWP). To perform computational tasks, the monitoring system can include a processor, a memory, a display, one or more computational programs stored in the memory and run by the processor, and one or more ports for connecting the various sensors to the system.

[0078] Method 400 guides (402) the pulmonary artery catheter to the pulmonary artery. Any suitable method can be employed to reach the pulmonary artery. The catheter can be inserted into a central vein and navigated to the right atrium of the heart, then through the tricuspid valve and the right ventricle, and then through the pulmonary valve into the pulmonary artery (e.g., see Figures 2A to 2D)。To assist in guiding and positioning, as the catheter floats into the pulmonary artery, the pressure waveform can be monitored. In some embodiments, the catheter can include markers (e.g., fluoroscopic, radiographic, or ultrasonic) at specific distances along the proximal-distal axis, and visualization methods such as (e.g.) fluoroscopy, X-ray, or ultrasound can be utilized.

[0079] Once the pulmonary artery catheter reaches the pulmonary artery, the pressure sensor can measure (404) the pulmonary artery blood pressure. In some embodiments, the blood pressure sensor includes a distal lumen connected to a pressure transducer for measuring blood pressure. The blood pressure measurement can be displayed on and / or recorded by the monitoring system.

[0080] Method 400 also acquires (406) PCWP measurements. In many embodiments, the balloon at or near the distal end of the catheter is inflated to allow the catheter to float into the wedge position. Once the catheter is in the wedge position, the distal lumen connected to the pressure transducer measures the blood pressure at the wedge position, which can be displayed on the monitor. As the pressure sensor measures the blood pressure while in the wedge position, PCWP measurements are obtained. As understood in the art, the timing of the PCWP measurements is obtained according to the patient's respiration and can also depend on whether the patient is breathing spontaneously or via a mechanical ventilator. Generally, PCWP can be obtained at the end of the respiratory cycle (i.e., after expiration is complete). In some embodiments, the PCWP measurements are displayed on the display screen of the monitoring system.

[0081] Method 400 can evaluate the quality of the PCWP measurements. The reliability of the PCWP measurements depends on the acquisition conditions. For example, if the distal lumen is not in the correct position or if the balloon is over-inflated, the PCWP measurements may be inaccurate. The quality can be evaluated by analyzing and comparing the pressure waveforms in the pulmonary artery and the wedge position.

[0082] In some embodiments, a quality rating can be generated. In some embodiments, the quality rating is generated by a computational process performed by a hemodynamic monitoring system. The quality rating can be quantitative (e.g., a score on a scale of 0 - 100) or categorical (e.g., "good" vs. "bad", "adequate" vs. "poor", etc.). In some embodiments, a threshold is utilized to determine if the quality score is low / high or "bad / good". The threshold can be based on clinical data. In some embodiments, a trained computational model is utilized to determine the quality rating. The machine learning computational model can be trained using clinical data. In some embodiments, the quality rating is displayed on a display screen of the hemodynamic monitoring system. In some embodiments, the quality rating is saved in a memory or transmitted to another computing device for storage or downstream analysis. In some embodiments, a low or "bad" quality score is utilized to notify a clinician to recommend repeating the PCWP measurement. In some embodiments, a low or "bad" quality score is utilized to automatically repeat the PCWP measurement.

[0083] Although specific examples of methods for performing PCWP measurements were referenced above Figure 4A those of ordinary skill in the art will understand that the various steps of the method can be performed in a different order and, depending on the various embodiments, certain steps can be optional. Thus, it should be clear that the various steps of the method can be used as appropriate according to the requirements of a particular application. Additionally, any of a variety of methods for performing PCWP measurements that are suitable for the requirements of a given application can be utilized in the various embodiments.

[0084] Figure 4B Examples of a set of actions and processes that a hemodynamic monitoring system can perform to acquire PCWP measurements and evaluate their quality are provided. The hemodynamic monitoring system can perform one or more of the set of actions and processes in any combination. In some embodiments, the set of actions and processes performed by the hemodynamic monitoring system enables automatic acquisition and / or evaluation of PCWP measurements.

[0085] A hemodynamic monitoring system may include a pulmonary artery catheter (e.g., a Swan-Ganz catheter) and a computational processing system. The pulmonary artery catheter may include one or more sensors, such as (e.g.) a pressure sensor, a thermal sensor (e.g., for measuring cardiac output via thermodilution), and an optical fiber (e.g., photometric or other optical measurements). The pulmonary artery catheter may also include a balloon near the distal end for performing the wedge method within the pulmonary artery. Using the pulmonary artery catheter, the hemodynamic monitoring system can measure blood pressure, cardiac output, and various other hemodynamic measurements in real time. The monitoring system may also include one or more computational programs to assist in monitoring hemodynamic parameters and / or performing various measurements (e.g., PCWP). To perform computational tasks, the monitoring system may include a processor, a memory, a display, one or more computational programs stored in the memory and run by the processor, and one or more ports for connecting the various sensors to the system.

[0086] As Figure 4B Depicted, the hemodynamic monitoring system may perform a set of actions (420) and a set of computational processes (440). The pulmonary artery catheter may be positioned within the pulmonary artery and acquire a pressure waveform (300). To measure PCWP, the monitoring system may enter the PCWP mode (422) to initiate the set of actions and computational processes. After entering the PCWP mode, the monitoring system may determine whether the system and the patient are ready for PCWP (422). The determination of readiness may include computational processes to determine whether there are problems with the machine or the current readings. In some embodiments, the computational process analyzes the waveform data for artifacts. The computational process may also run a set of checks, such as (e.g.) ensuring that the pulmonary catheter and the machine are properly connected and that the patient's breathing is normal. After determining that the system and the patient are ready for PCWP measurements, in some embodiments, the monitoring system may provide a signal to the clinician that balloon inflation may begin. In some embodiments, after the determination of readiness, the monitoring system automatically begins the process of balloon inflation.

[0087] The hemodynamic monitoring system inflates the balloon at the distal end of the catheter (424), allowing the catheter to float to the wedge position. By evaluating the hemodynamic parameters derived from the pressure waveform, the hemodynamic system can detect the balloon inflation point (444), thereby giving an indication that the catheter is in the wedge position. In some embodiments, the computational process detects balloon inflation through a probability model. In some embodiments, the computational process uses fuzzy logic or Boolean logic to detect balloon inflation.

[0088] After detecting balloon inflation, in some embodiments, the hemodynamic monitoring system may perform one or more computational processes to determine the respiratory cycle (446). Figure 4BA respiratory cycle (310) and a blood pressure waveform (300) are shown. In some embodiments, the respiratory cycle is extracted using the blood pressure waveform. In some embodiments, a sensor or other device is used in combination with a hemodynamic monitor to detect the respiratory cycle. There are many known techniques for measuring the respiratory rate, such as (for example) measuring air flow (e.g., a flow meter), detecting respiratory sounds (e.g., a microphone), measuring air temperature (e.g., a thermistor), measuring gas humidity (e.g., a capacitive sensor), measuring CO 2 (e.g., an infrared sensor), measuring chest wall movement (e.g., an accelerometer), and cardiac activity (such as an ECG sensor). In some embodiments, when a mechanical ventilator is used, the respiratory rate is determined by the ventilator settings.

[0089] The hemodynamic monitoring system can measure PCVP (426) using a computational process for measuring PCWP (446). In some embodiments, PCWP is measured according to the respiratory cycle. In certain embodiments, PCWP is the pressure measured after completion of the respiratory cycle (i.e., end-expiratory pressure or the average of end-expiratory pressures). In some embodiments, PCWP is the average pressure measured over multiple respiratory cycles (e.g., 1, 2, 3, or 4 respiratory cycles) during the wedge-in period. In some embodiments, PCWP is the average pressure measured over a period of time (e.g., between 2 and 20 seconds) during the wedge-in period.

[0090] After completion of the PCWP measurement, the hemodynamic monitoring system deflates the balloon (428) at the distal end of the catheter, allowing the catheter to retract into the pulmonary artery. In a manner similar to detecting balloon inflation, the monitoring system can include a computational process (448) for detecting the balloon deflation point. Thus, by evaluating hemodynamic parameters derived from the pressure waveform, the hemodynamic system can detect the balloon deflation point, thereby giving an indication that the catheter has retracted from the wedge position. In some embodiments, the computational process detects balloon deflation through a probability model. In some embodiments, the computational process uses fuzzy logic or Boolean logic to detect balloon deflation.

[0091] A hemodynamic monitoring system can perform a quality assessment of PCWP acquisition (450) by analyzing hemodynamic data between an airbag inflation point and an airbag deflation point. In some embodiments, the quality of PCWP acquisition can be performed by analyzing a pressure waveform. In certain embodiments, the quality of PCWP acquisition can be performed by comparing the acquired PAP with the acquired PCWP. In some embodiments, a trained machine learning model is utilized to determine the quality of PCWP acquisition, which can be performed by analysis. Various machine learning models can be used, including (but not limited to) regression models, logistic regression models, neural networks, support vector machines, decision trees, adaboost, random forests, ensemble learning models (combining one or more models), and / or any other machine learning model capable of determining the quality of a pressure reading based on feature data and classification measurements. Clinical data of PCWP acquisition can be utilized to train the machine learning model, where a set of hemodynamic data features are utilized to distinguish low-quality PCWP acquisition data from high-quality PCWP acquisition data. In some embodiments, hemodynamic data features are extracted from PAP and wedge pressure waveforms. The quality rating can be quantitative (e.g., a score on a scale of 0 - 100) or categorical (e.g., "good" vs. "bad", "sufficient" vs. "poor", etc.). In some embodiments, a threshold is utilized to determine whether the quality score is low / high or "bad / good". The threshold can be based on clinical data.

[0092] The hemodynamic monitoring system can report a quality rating of the acquired PCWP and / or PCWP measurements. In some embodiments, the quality of the acquired PCWP and / or PCWP measurements is displayed on a display screen of the monitoring system. In some embodiments, the quality rating is saved in a memory or transmitted to another computing device for storage or downstream analysis. In some embodiments, a low or "bad" quality score is utilized to notify a clinician to recommend repeating the PCWP measurement. In some embodiments, a low or "bad" quality score is utilized to automatically repeat the PCWP measurement. After the PCWP measurement is completed (or the repetition of the PCWP measurement), the monitoring system can exit the PCWP mode (432).

[0093] Although specific examples of a set of hemodynamic monitoring system actions and computational processes for performing PCWP measurements are described above with reference to Figure 4B those skilled in the art will understand that the various actions and processes can be performed in different orders, and according to various embodiments, certain actions and processes can be optional or omitted. Thus, it should be clear that the various actions and processes can be used appropriately according to the requirements of a particular application. Additionally, any of the various actions and processes for performing PCWP measurements suitable for the requirements of a given application can be utilized in various embodiments.

[0094] Figure 5 Examples of computational methods are provided for assessing a pulmonary artery pressure (PAP) waveform for artifacts and alerting when an artifact is detected, which can be performed using a hemodynamic monitoring system. Before starting a procedure to acquire PCWP measurements, it can be determined whether the system and the patient are in appropriate conditions to perform the procedure. This determination can prevent poor PCWP acquisition quality and help ensure high-quality measurements are taken. In some embodiments, the computational method for assessing a pulmonary artery pressure (PAP) waveform for artifacts is used as or within a computational process for determining whether the monitoring system and the patient are ready for PCWP acquisition (e.g., see Figure 4B 442 of

[0095] Computational method 500 can be performed in real time while a PAP waveform is being acquired using a pulmonary artery catheter (e.g., a Swan-Ganz catheter) and before the balloon is inflated to perform a PCWP measurement. Method 500 can assess (502) the PAP for artifacts. Artifacts can be caused by things such as patient movement, measurements obtained during catheter flushing, and measurements indicating a flat line, insufficient damping, excessive damping, and / or any other phenomenon that may result in an abnormal or unnatural reading of the blood pressure waveform.

[0096] The PAP can be assessed for artifacts over a period of time. In some embodiments, the assessment of the PAP for artifacts is performed for a period of time between 1 second and 20 seconds. In some embodiments, the assessment of the PAP for artifacts is performed for a period of approximately 1 second, approximately 2 seconds, approximately 3 seconds, approximately 4 seconds, approximately 5 seconds, approximately 10 seconds, approximately 15 seconds, or approximately 20 seconds. This period of time can be further subdivided into smaller discrete or overlapping intervals for batch analysis.

[0097] In some embodiments, heuristic metrics can be used to determine whether an artifact is present within the PAP waveform. In some embodiments, if the PAP maximum (PAP max ) is greater than a threshold, if the PAP minimum (PAP min ) is less than a threshold, and / or if the PAP maximum minus the PAP minimum (PAP max - PAP min ) is greater than a threshold, then an artifact based on flushing, improper zeroing, or patient movement can be detected. In various embodiments, an artifact is detected when PAP max is greater than 60 mmHg, greater than 80 mmHg, less than 100 mmHg, or greater than 120 mmHg. In various embodiments, when PAP minArtifacts are detected when less than 5 mmHg, less than 0 mmHg, less than -10 mmHg, less than -20 mmHg, or less than -30 mmHg. In various embodiments, when PAP max -PAP min Artifacts are detected when greater than 60 mmHg, greater than 80 mmHg, greater than 100 mmHg, or greater than 120 mmHg.

[0098] In some embodiments, if the PAP maximum value minus the PAP minimum value (PAP max -PAP min ) is less than a threshold, artifacts based on a flat line or excessive damping can be detected. In various embodiments, when PAP max -PAP min Artifacts are detected when less than 5 mmHg, less than 2 mmHg, greater than 1 mmHg, or less than 0.5 mmHg.

[0099] After determining that an artifact exists, method 500 can provide (504) an alert. In some embodiments, the hemodynamic monitoring system displays the alert on a display screen. In some embodiments, when an artifact is detected, the monitoring system prevents the execution of the PCWP procedure.

[0100] Although specific examples of computational methods for evaluating the PAP waveform for artifacts are described above with reference to Figure 5 , those of ordinary skill in the art will understand that the steps of the method can be performed in a different order, and in various embodiments, certain steps may be optional. Thus, it should be clear that the steps of the method can be used appropriately according to the requirements of a particular application. Additionally, in various embodiments, any of the various methods for evaluating the PAP waveform for artifacts suitable for the requirements of a given application can be utilized.

[0101] Figure 6 Examples of computational methods for determining the transition to and / or from the wedge position are provided in Figure 4Bof 444 and 448).

[0102] Calculation method 600 can be executed in real time after deciding to perform PCWP measurement value acquisition and / or after completing PCWP measurement value acquisition. This method can help determine whether the pulmonary artery catheter is properly inflated with the balloon and reaches the wedge position, and / or help determine whether the pulmonary artery catheter is properly deflated with the balloon and returns from the wedge position. Method 600 can extract (602) hemodynamic parameters from the PAP and wedge position waveforms. The PAP and wedge position waveforms can be obtained from a pulmonary artery catheter (such as a Swan-Ganz catheter).

[0103] Any hemodynamic parameter that can distinguish the PAP waveform and the wedge position waveform can be utilized. Examples of hemodynamic parameters that can be utilized include (but are not limited to) blood pressure parameters, phase parameters, morphological features, determinable components, demographic features, and combinations thereof. Examples of blood pressure parameters that can be utilized include (but are not limited to) pulse pressure (systolic blood pressure - diastolic blood pressure), mean pressure, median pressure, diastolic blood pressure, systolic blood pressure, and / or any other obtainable blood pressure parameter. Examples of phase parameters that can be utilized include (but are not limited to) contractility, pulmonary artery compliance, stroke volume, vascular tone, afterload, complete cardiac cycle, and / or any other phase parameter. Morphological features that can be utilized include (but are not limited to) frequency components (fft), skewness, kurtosis, and / or any other obtainable morphological feature. In addition, determinable components include numerical analysis of the waveform, such as percentiles (e.g., 10%, 25%, 50%, 75%, 90% of systolic blood pressure or any other hemodynamic parameter), the number of peaks above the percentile, the number of valleys below the percentile, statistical moments, histograms of the data, Shannon entropy, mean crossing number, area under the curve, singular vector coefficients after principal component analysis, subsampled signals, and / or any other determinable component. Demographic features that can be utilized include (but are not limited to) age, gender, sex, medical history, body mass index, any other relevant feature, and / or any other demographic feature. In some embodiments, the hemodynamic parameters extracted from the PAP waveform and the wedge position waveform include pulse pressure (systolic blood pressure - diastolic blood pressure) and mean pressure.

[0104] Using the extracted hemodynamic parameters, a fuzzy logic score is generated (604) using a fuzzy logic membership function to determine the status of the current pressure measurement value. To generate the fuzzy logic value, the difference between the extracted hemodynamic parameters of the current pressure measurement value and past PAP samples is calculated. This difference is utilized within the fuzzy logic membership function to determine the current status.

[0105] Any fuzzy logic membership function can be utilized that is suitable for assigning membership based on the difference between the wedge position pressure and the extracted hemodynamic parameters of the PAP. Examples of fuzzy logic membership functions that can be utilized include (but are not limited to) triangular, Gaussian, and trapezoidal. Additionally, any method for generating a fuzzy logic membership function can be utilized that is suitable for assigning membership based on the difference between the wedge position pressure and the extracted hemodynamic parameters of the PAP. Generally, clinical data of the wedge position pressure and PAP waveforms can be utilized to identify hemodynamic parameters that provide robust membership classification. An example of generating a fuzzy logic function is shown in Figure 8 and the related description.

[0106] In some embodiments, hemodynamic parameter thresholds are utilized in combination with the fuzzy logic membership function to assign membership. Thus, in addition to assigning fuzzy logic membership, the extracted hemodynamic parameters must also be within the thresholds assigned to a particular state (e.g., transitioning to or recovering from the wedge position pressure). For example, a fuzzy logic membership function based on pulse pressure and mean pressure can be used, in combination with a mean pressure threshold (e.g., mean pressure below the threshold) and a pulse pressure threshold (e.g., pulse pressure below the threshold) to detect the transition to the wedge position (i.e., the balloon inflation point). In another example, a fuzzy logic membership function based on pulse pressure and mean pressure or a mean pressure threshold (e.g., mean pressure above the threshold) or a pulse pressure threshold (e.g., pulse pressure above the threshold) can be used to detect the transition from the wedge position recovery (i.e., the balloon deflation point).

[0107] Based on the results of the fuzzy logic membership function (and hemodynamic parameter thresholds, if utilized), method 600 determines (606) the transition point from the pulmonary artery to the wedge position (i.e., the balloon inflation point) and / or the transition point from the wedge position back to the pulmonary artery (i.e., the balloon deflation point).

[0108] Figure 7A Examples of fuzzy logic membership plots are provided in for plotting the transition to the wedge position and the transition back to the pulmonary artery, where the hemodynamic parameters used to distinguish the positions are pulse pressure and mean pressure. As shown in the membership plot, the relationship between the difference in pulse pressure and the difference in mean pressure is plotted. When the calculated fuzzy logic value falls within a certain region, membership can be assigned to that state.

[0109] Figures 7B to 7E A fuzzy logic membership trapezoidal function is provided that is plotted on top of the pulmonary artery pressure waveform 300 captured by the pulmonary artery catheter. The fuzzy logic membership trapezoidal function is based on the input of the difference between pulse pressure and mean pressure. Figure 7B and 7C each depict successful transitions to and from the wedge position, marked by transitions 702, 704, 706, and 708. Figure 7D and7E Each depicts an unsuccessful transition to and from the wedge position, marked by transitions 710 and 712.

[0110] While specific examples of computational methods for determining transitions to and / or from the wedge position were described above with reference to Figure 6 those skilled in the art will appreciate that the various steps of the method may be performed in a different order and, depending on the various embodiments, certain steps may be optional. Thus, it should be clear that the various steps of the method may be used as appropriate according to the requirements of a particular application. Additionally, any of a variety of methods for determining transitions to and / or from the wedge position suitable for the requirements of a given application may be utilized in the various embodiments.

[0111] Figure 8 Examples of computational methods for generating fuzzy logic membership functions for determining transitions to and / or from the wedge position are provided in. The generated fuzzy logic functions may be used to evaluate in real time the transition of a pulmonary artery catheter (e.g., a Swan-Ganz catheter) from the pulmonary artery to the wedge position (i.e., the balloon inflation point) and / or the transition from the wedge position back to the pulmonary artery (i.e., the balloon deflation point).

[0112] Computational method 800 obtains (802) clinical pulmonary artery and wedge position pressure waveform data, which may be obtained from a PCWP procedure. The waveform data should have a discernible transition between the pulmonary artery and the wedge position.

[0113] Method 800 identifies (804) the PAP region and the wedge pressure region in the waveform data. Generally, the regions selected should be clearly in the PAP state or clearly in the wedge pressure state.

[0114] Method 800 extracts one or more hemodynamic parameters from the PAP region and the wedge pressure region. Any hemodynamic parameter capable of differentiating between the PAP waveform and the wedge position waveform can be utilized. Examples of hemodynamic parameters that can be utilized include (but are not limited to) blood pressure parameters, phase parameters, morphological features, determinable components, demographic features, and combinations thereof. Examples of blood pressure parameters that can be utilized include (but are not limited to) pulse pressure (systolic blood pressure - diastolic blood pressure), mean pressure, median pressure, diastolic blood pressure, systolic blood pressure, and / or any other obtainable blood pressure parameter. Examples of phase parameters that can be utilized include (but are not limited to) contractility, pulmonary artery compliance, stroke volume, vascular tone, afterload, complete cardiac cycle, and / or any other phase parameter. Morphological features that can be utilized include (but are not limited to) frequency components (fft), skewness, kurtosis, and / or any other obtainable morphological feature. Additionally, determinable components include numerical analysis of the waveform, such as percentiles (e.g., 10%, 25%, 50%, 75%, 90% of systolic blood pressure or any other hemodynamic parameter), number of peaks above a percentile, number of valleys below a percentile, statistical moments, histogram of the data, Shannon entropy, mean crossing number, area under the curve, singular vector coefficients after principal component analysis, subsampled signal, and / or any other determinable component. Demographic features that can be utilized include (but are not limited to) age, sex, gender, medical history, body mass index, any other relevant feature, and / or any other demographic feature. In certain embodiments, the hemodynamic parameters extracted from the PAP waveform and the wedge position waveform include pulse pressure (systolic blood pressure - diastolic blood pressure) and mean pressure.

[0115] Parameters for generating the fuzzy logic membership can be selected for various purposes. The parameters can be the selected computational power, ease of measurement, correlation, accuracy, any other relevant parameter and / or metric, and combinations thereof. The correlation can be a correlation with another feature, such as a correlation with pulse pressure and / or any other feature.

[0116] Method 800 generates a fuzzy logic membership function using one or more of the extracted hemodynamic parameters. To generate the fuzzy logic membership function, the difference in the extracted hemodynamic parameters between the wedge pressure region and the PAP region is calculated. This difference is utilized in the fuzzy logic membership function to determine whether the hemodynamic parameter provides a robust determination of the state difference.

[0117] In some embodiments, hemodynamic parameter thresholds are utilized in combination with fuzzy logic membership functions to assign membership degrees. Thus, in addition to assigning fuzzy logic membership degrees, the extracted hemodynamic parameters must also be within the thresholds assigned to a particular state (e.g., transitioning to or recovering from a wedge position pressure). For example, a fuzzy logic membership function based on pulse pressure and mean pressure can be used, in combination with a mean pressure threshold (e.g., mean pressure below the threshold) and a pulse pressure threshold (i.e., pulse pressure below the threshold) to detect the transition to the wedge position (i.e., the balloon inflation point). In another example, a fuzzy logic membership function based on pulse pressure and mean pressure or a mean pressure threshold (e.g., mean pressure above the threshold) or a pulse pressure threshold (e.g., pulse pressure above the threshold) can be used to detect the transition from the wedge position recovery (i.e., the balloon deflation point).

[0118] Figure 9 Plots of pulse pressure (systolic pressure - diastolic pressure) and mean pressure (average signal) derived from clinical pulmonary artery and wedge position pressure waveform data from a large number of patients are provided. As can be seen from the plots, the hemodynamic parameter data derived from the wedge position waveforms are mainly located in the lower left quadrant. To ensure that the derived hemodynamic parameter data are actually derived from the wedge position waveforms, a pulse pressure threshold (e.g., 15 mmHg) and a mean pressure threshold (e.g., 32 mmHg) can be applied.

[0119] Although specific examples of computational methods for generating fuzzy logic membership functions are described above with reference to Figure 6 those of ordinary skill in the art will appreciate that the steps of the method can be performed in a different order and, depending on the various embodiments, certain steps may be optional. Thus, it should be clear that the steps of the method can be used as appropriate according to the requirements of a particular application. Additionally, any of a variety of methods for generating fuzzy logic membership functions suitable for the requirements of a given application can be utilized in the various embodiments.

[0120] According to some embodiments, a variety of other methods can be utilized to separate, classify, and / or otherwise analyze data to identify clusters for detecting transitions to and / or from the wedge position. These methods can include algorithms, machine learning (e.g., artificial intelligence), statistics, and / or any other method for identifying data clusters. Various machine learning models include (but are not limited to) convolutional neural networks (CNNs), support vector machines (SVMs), and / or any other machine learning model sufficient to identify features to classify catheter position.

[0121] Figure 10Examples of computational methods for performing PCWP measurements and evaluating the quality of PCWP acquisitions are provided, which can be performed using a hemodynamic monitoring system. Determining the quality of PCWP acquisitions can help inform clinicians whether the PCWP measurement value is suitable for health determination or whether to repeat the PCWP procedure. Additionally, by using only the portion of the wedge position pressure waveform that is above the quality standard, the evaluation of the quality of wedge position pressure acquisitions can be used to improve PCWP calculations. The quality of PCWP acquisitions can be determined using one or more trained machine learning models and / or various hemodynamic thresholds. In some embodiments, the computational method for performing PCWP acquisitions and evaluating their quality can be combined with a set of other computational processes for performing high-quality automated PCWP measurements (e.g., see Figure 4B of 450).

[0122] Computational method 1000 can be performed in real time to improve and / or evaluate the quality of PCWP measurement acquisitions. Method 1000 segments (1002) the PAP and wedge position waveforms into a plurality of time windows. The PAP and wedge position waveforms can be obtained from a pulmonary artery catheter (e.g., a Swan-Ganz catheter).

[0123] The plurality of time windows can be discrete or overlapping, continuous or discontinuous. The time window can be defined by any non-zero time period up to and including the full time frame of the PAP or wedge position waveform. In some embodiments, the time window length is between 0.5 seconds and 20 seconds. In various embodiments, the time window length is between 0.5 seconds and 1.5 seconds, between 1.0 seconds and 2.0 seconds, between 1.0 seconds and 3.0 seconds, between 2.0 seconds and 4.0 seconds, between 3.0 seconds and 5.0 seconds, between 4.0 seconds and 6.0 seconds, between 5.0 seconds and 7.0 seconds, between 6.0 seconds and 8.0 seconds, between 7.0 seconds and 9.0 seconds, between 8.0 seconds and 10.0 seconds, between 9.0 seconds and 11.0 seconds, between 10.0 seconds and 15.0 seconds, between 12.5 seconds and 17.5 seconds, or between 15.0 seconds and 20.0 seconds. In some embodiments, the time window is defined by a physiological event (e.g., one or more respiratory cycles). In some embodiments, the user (e.g., a clinician) can select the time window length.

[0124] Method 1000 extracts one or more hemodynamic parameters from the PAP region and the wedge pressure region. Any hemodynamic parameter capable of distinguishing between the PAP waveform and the wedge position waveform can be utilized. Examples of hemodynamic parameters that can be utilized include, but are not limited to, blood pressure parameters, phase parameters, morphological features, determinable components, demographic features, and combinations thereof. Examples of blood pressure parameters that can be utilized include, but are not limited to, pulse pressure (systolic blood pressure - diastolic blood pressure), mean pressure, median pressure, diastolic blood pressure, systolic blood pressure, and / or any other obtainable blood pressure parameter. Examples of phase parameters that can be utilized include, but are not limited to, contractility, pulmonary artery compliance, stroke volume, vascular tone, afterload, complete cardiac cycle, and / or any other phase parameter. Morphological features that can be utilized include, but are not limited to, frequency components (fft), skewness, kurtosis, and / or any other obtainable morphological feature. In addition, determinable components include numerical analysis of the waveform, such as percentiles (e.g., 10%, 25%, 50%, 75%, 90% of systolic blood pressure or any other hemodynamic parameter), number of peaks above a percentile, number of valleys below a percentile, statistical moments, histogram of the data, Shannon entropy, mean crossing number, area under the curve, singular vector coefficients after principal component analysis, subsampled signal, and / or any other determinable component. Demographic features that can be utilized include, but are not limited to, age, gender, sex, medical history, body mass index, any other relevant feature, and / or any other demographic feature. In some embodiments, the hemodynamic parameters extracted from the PAP waveform and the wedge position waveform include pulse pressure (systolic blood pressure - diastolic blood pressure) and mean pressure.

[0125] Method 1000 evaluates the quality of the PAP and wedge position pressure waveforms at multiple segmented time windows. Quality can be evaluated by various methods, which can be used in combination. In some embodiments, a trained machine learning model is utilized to evaluate quality. In some embodiments, thresholds for various hemodynamic parameters are utilized to evaluate quality. The quality rating can be quantitative (e.g., a score on a scale of 0 - 100) or categorical (e.g., "good" vs. "bad", "sufficient" vs. "poor", etc.). In some embodiments, thresholds are utilized to determine whether the quality score is low / high or "bad / good". The thresholds can be based on clinical data.

[0126] To evaluate the quality of PAP and wedge position pressure waveforms using a machine learning model, one or more parameters are input into the model as features to produce an assessment of the quality. A variety of machine learning models can be used, including (but not limited to) regression models, logistic regression models, neural networks, convolutional neural networks, support vector machines, decision trees, adaboost, random forests, ensemble learning models (combining one or more models), and / or any other machine learning model capable of determining the quality of a pressure reading based on feature data and classification measurements. Clinical data of PAP and wedge position pressures collected can be utilized to train the machine learning model, where a set of hemodynamic data features are used and high-quality data time windows are utilized to distinguish low-quality data time windows. In some embodiments, hemodynamic data features are extracted from clinical PAP and wedge pressure waveforms and associated with their data windows. The data windows can be distinguished as being derived from PAP or from wedge position waveforms based on their quality, assigned quality scores, or any other quality indicator. The machine learning model can be trained to recognize the quality of PAP and wedge pressure waveform windows. In some embodiments, the quality is determined by the ability to distinguish a window of the wedge pressure waveform from a window of the PAP waveform.

[0127] In some embodiments, the quality is evaluated based on whether a window of the waveform is above or below a threshold for various hemodynamic parameters. For multiple hemodynamic parameters, the PAP waveform and the wedge position waveform are expected to be within a certain range, as determined by clinical data. Additionally, some hemodynamic parameters are expected to have a high difference between the PAP waveform and the wedge position waveform, as determined by clinical data. By utilizing these expected ranges or expected differences between waveforms, the windows of these waveforms can be evaluated for quality. In some embodiments, if a window of a waveform meets one or more criteria, it is determined to be above the quality standard. In some embodiments, if a window of a waveform does not meet one or more criteria, it is determined to be below the quality standard.

[0128] Parameter thresholds can be based on one or more qualitative, quantitative, or semi - quantitative factors that can evaluate the quality of the window. Non - limiting examples of thresholds that can be used to identify parameters can be based on professional experience and / or metrics identified in the literature. Parameter thresholds that can be used (either alone or in combination) as quality metrics for the window include (but are not limited to) an increase / decrease in pressure exceeding a threshold, including (but not limited to) mean pressure and pulse pressure; mean wedge position pressure being lower than pulmonary artery diastolic pressure; an increase / decrease in oxygen saturation exceeding a threshold; an increase in respiration - induced variation in the wedge position compared to the pulmonary artery position; and / or morphological features such as well - defined waves in the pressure (e.g., "a" and "v" waves). Using one or more quality features, a quality score can be generated for the window of the waveform, which can be a categorical, quantitative, or semi - quantitative quality score. Qualitative categorization can be binary (e.g., good or bad) and / or qualitative ranking (e.g., poor, moderate, good).

[0129] In some embodiments, multiple evaluations for quality are combined, which can be considered factors and / or requirements for quality assessment. For example, multiple weighting factors can be used to determine the quality of the window, with each factor combined together to produce an overall determination of quality. Alternatively or additionally, the quality of the window must meet all of the requirements in one or more requirements to meet the quality criteria, and failure to meet any one of the one or more requirements results in failure to meet the criteria. In some embodiments, when a machine - learning model is combined with one or more parameter thresholds of hemodynamics, the thresholds can be applied as a prerequisite for entering the model, integrated with the model, or as a post - model assessment.

[0130] Figure 11A and Figure 11B Examples of evaluating the quality of waveform windows are provided. Specifically, Figure 11A A table of evaluations that can classify the window of the waveform on a binary basis (e.g., true - false, 0 - 1, etc.) is provided. These evaluations utilize the chance (or likelihood) that the PCWP exceeds the PAP position, and the mean PCWP pressure is less than the mean PAP diastolic pressure. Based on the binary decision, the quality can be evaluated. Similarly, Figure 14B Multiple binary requirements are provided based on the extracted parameters, including whether the mean PCWP pressure (PCWP Mean ) is less than the mean PAP diastolic pressure (PAP Diastolic ), and whether the weighted product of the mean PCWP pulse pressure (PCWP PulsePress ) and the mean PCWP pressure (PCWP Mean ) is greater than a decisive value or less than or equal to a decisive value. Although Figure 11BSpecific equations are depicted, but various equations can be generated or altered based on specific parameters used in the equations. Additionally, any equation used to generate a hyperplane, where the weights a and b and the decisive value c can be revealed as the equation describing the hyperplane.

[0131] Figure 12 An assessment of the quality 1202 of the window of waveform 300 superimposed on top of a waveform is provided. The wedge position portion 1204 of waveform 300 can be evaluated as a quantitative or qualitative quality score. As shown, several windows of the wedge position portion 1204 have "good quality"

[0132] Method 1000 performs PCWP measurements. In some embodiments, PCWP is measured according to the respiratory cycle. In certain embodiments, PCWP is the pressure measured after completion of the respiratory cycle (i.e., end-expiratory pressure or the average of end-expiratory pressures). In some embodiments, PCWP is the average pressure measured over multiple respiratory cycles (e.g., 1, 2, 3, or 4 respiratory cycles) during the wedge-in period. In some embodiments, PCWP is the average pressure measured over a period of time (e.g., between 2 and 20 seconds) during the wedge-in period. In some embodiments, PCWP measurements are acquired and then the measurement is evaluated for quality.

[0133] Quality assessment can be utilized to enhance the determination of PCWP. In some embodiments, quality assessment is used to enhance the determination of PCWP measurements. In some embodiments, the quality of multiple windows of the wedge position waveform is determined and if a certain number of windows do not meet the quality criteria, the PCWP measurement is not determined. In some embodiments, only windows that meet the quality criteria are used to determine the PCWP measurement. In some embodiments, windows that do not meet the quality criteria are discarded before determining the PCWP measurement.

[0134] Method 1000 reports the PCWP and / or the quality assessment of the PCWP acquisition. In some embodiments, the quality of the acquired PCWP and / or the PCWP measurement is displayed on the display screen of the monitoring system. In some embodiments, the quality rating is saved in a memory or transmitted to another computing device for storage or downstream analysis. In some embodiments, a low or "poor" quality score is used to notify the clinician to recommend repeating the PCWP measurement. In some embodiments, a low or "poor" quality score is used to automatically repeat the PCWP measurement. After the PCWP measurement is completed (or the repetition of the PCWP measurement), the monitoring system can exit the PCWP mode (432).

[0135] In determining quality, certain systems and / or methods of the present disclosure are designed to identify an optimized combination of input parameters to produce a quality rating of a pressure waveform. The optimization objective can be any one or any combination of reducing labor, reducing cost, reducing risk, increasing reliability, increasing efficacy, reducing side effects, reducing toxicity, and reducing drug resistance and other benefits.

[0136] Blood pressure waveforms have various distinguishable characteristics that can be identified within a single pulsatile waveform (e.g., a single cardiac cycle of systole and diastole) or over several pulsatile waveforms (e.g., several cardiac cycles). Referring to Figure 13A and Figure 13B , general waveform characteristics relative to a single beat are shown. In particular, Figure 13A shows a portion of the waveform representing systole and diastole, as well as the ascending and dicrotic limbs of the pulse wave, within a single beat. Figure 13B shows the granular characteristics of the pulsatile waveform, identifying the upstroke, peak pressure, and descent during systole, as well as the dicrotic notch, diastolic runoff, and end-diastolic pressure.

[0137] In addition to Figure 13A and Figure 13B the characteristics identified in, many systems and / or methods can also identify, measure, or calculate various additional characteristics within the waveform of a single beat. Referring to Figures 14A to 14F , waveform phase characteristics identified from a single beat can include contractility ( Figure 14A ), pulmonary artery compliance ( Figure 14B ), stroke volume ( Figure 14C ), vascular tone ( Figure 14D ), afterload ( Figure 14E ), and / or one or more of the entire cardiac cycle ( Figure 14F ). Figures 14A to 14FEach "zhong" in it represents a characteristic that is the shaded part of the waveform. In addition to individual measurements, various systems and / or methods also determine one or more determinable characteristics. Such determinable characteristics can be determined from one or more in each phase of the pressure waveform, including (but not limited to) the average value, maximum value, minimum value, duration, area, standard deviation, slope, derivative (as the differential of pressure with respect to time (dP / dt)), trend, deviation from the trend, variation, variance, and combinations thereof. Some systems and / or methods also determine one or more morphological characteristics. Such morphological characteristics can be determined from one or more waveform phases, and such morphological features include (but not limited to) frequency components (fft), skewness, kurtosis, any other determinable morphological characteristics within the waveform, and combinations thereof. Using any one or more of the above characteristics (e.g., phase characteristics, determinable characteristics, morphological characteristics, etc.), additional systems and / or methods determine one or more of complexity (e.g., entropy measures such as approximate entropy and / or sample entropy), pressure reflection sensitivity (e.g., via cross-correlation analysis), variability, change, and combinations thereof.

[0138] Based on the pressure waveform, various systems and / or methods further extract one or more of the following parameters: heart rate, respiratory rate, stroke volume, pulse pressure, mean pulmonary artery pressure (mPAP), pulmonary artery systolic pressure (sPAP), pulmonary artery diastolic pressure (dPAP), pulse pressure variation, stroke input volume variation, heart rate variability, cardiac output, pulmonary peripheral resistance, vascular compliance, vascular elasticity, right ventricular contractility (dP / dt).

[0139] Some catheters (e.g., Swan-Ganz catheter) include temperature probes, optical fibers, and / or other components capable of providing additional measurements. Thus, certain systems and / or methods can extract characteristics such as cardiac output, stroke volume, ejection fraction, end-diastolic volume, blood temperature, blood oxygenation, any other measurement values that can be obtained from the catheter, and combinations thereof. It should be noted that the above additional measurement values can be measured directly (e.g., individual measurement) or extracted from the waveform. In addition, as mentioned above, many systems and / or methods extract additional characteristics (e.g., determinable features, morphological features, etc.) from any such additional measurement values.

[0140] Various systems and methods use one or more of the above characteristics and characteristic types (e.g., morphological characteristics, phase characteristics, etc.) to identify wedge transitions and / or the quality of the pressure waveform during the PCWP procedure, as will be described in further detail.

[0141] Systems for Hemodynamic Monitoring and Calculation

[0142] Computing processing systems that perform PCWP measurements and evaluate their quality according to the various methods and processes of the present disclosure typically utilize a processing system including one or more of a CPU, GPU, and / or neural processing engine. As described herein, a computing processing system can record digital arterial pressure in real time and transform it into radial arterial pressure.

[0143] The computing processing system can be housed within a hemodynamic monitoring system with a direct wired connection between the monitor and components including a pulmonary artery catheter (e.g., Swan-Ganz catheter). Alternatively, the computing processing system can be housed separately from the hemodynamic monitoring system and components and receive the acquired pulmonary artery pressure via a wireless connection (e.g., WiFi, cellular, Bluetooth, etc.). The computing processing system can be implemented on any suitable computing device such as (but not limited to) a hemodynamic monitoring system, a tablet computer, and / or a portable computer.

[0144] Figure 15 An exemplary computing processing system that can be used to perform the various methods and processes of the present disclosure is shown. Computing processing system 1500 includes a processor system 1502, an I / O interface 1504, and a memory system 1506. As can be readily understood, the processor system 1502, the I / O interface 1504, and the memory system 1506 can be implemented using any of a variety of components suitable for the requirements of a particular application, including (but not limited to) a CPU, a GPU, an ISP, a DSP, a wireless modem (e.g., WiFi, Bluetooth modem), a serial interface, volatile memory (e.g., DRAM), and / or non-volatile storage (e.g., SRAM and / or NAND flash).

[0145] In the example shown, the memory system is capable of storing various data, applications, and models. It should be understood that the listed data, applications, and models are representative samples of what can be stored in the memory, and various memory systems can store some or all of the various data, applications, and models listed. Additionally, any combination of data, applications, and models can be stored, and in some embodiments, the various data, applications, and / or models are stored temporarily.

[0146] In some embodiments, the memory system 1506 may store, for example, one or more of the following applications: determining that the system is ready for PCWP 1508, detecting balloon inflation and / or deflation points 1510, detecting a respiratory cycle 1512, measuring PCWP 1514, and performing quality assessment of PCWP acquisition 1516. The various applications may be provided as individual processes or collections of processes, each of which may be used to provide automated acquisition and quality assessment of PCWP measurements. Real-time PCWP results and quality ratings 1518 may also optionally be stored on the memory system 1506 and / or displayed on a display screen via the I / O interface 150404.

[0147] Although the above reference Figure 15 A specific computing processing system is described, but it should be readily understood that the computing processes and / or other processes for providing the PCWP measurements and assessing their quality may be implemented on any of a variety of processing devices, including combinations of processing devices. Thus, the computing device should be understood to be not limited to a specific monitoring system, computing processing system, and / or specific applications and models. The computing device may be implemented using any combination of the systems described herein and / or modified versions of the systems described herein to perform the processes, combinations of processes, and / or modified versions of processes described herein.

[0148] The systems and methods of the present disclosure may be used in a hemodynamic monitoring system. Typically, a hemodynamic monitoring system includes a pulmonary artery catheter (PAC; eg, a Swan-Ganz catheter). Figure 16 An example of a hemodynamic monitoring system 1600 to be used to measure PCWP of an individual is provided in . Within the patient is a PAC 1620, which may include one or more sensors, such as (for example) a pressure sensor, a thermal sensor (e.g., for measuring cardiac output via thermodilution), and optical fibers (e.g., photometric or other optical measurements). The PAC 1620 may also include a balloon near the distal end for performing a wedge approach within the pulmonary artery. The PAC 1620 may be connected to the hemodynamic monitoring system 1600. A pump system may be provided to inflate the balloon.

[0149] The hemodynamic monitoring system 1600 may include a computing system such as, for example, a reference Figure 8The described and depicted system. The hemodynamic monitoring system 1600 may include a processor system 1602 and an I / O interface 1604 for inputting and outputting data, such as data communicated between the hemodynamic monitoring system 1600, the sensors of the PAC 1620, and the user interface. The hemodynamic monitoring system 900 may utilize multiple application programs stored in a memory system 1606 and executed by the processor system 1602. Application programs that may be stored in the memory system 1606 include a real-time PAP acquisition application program 1608, a real-time PCWP acquisition application program 1610, and a real-time extraction of hemodynamic parameters for operating the hemodynamic monitoring system 1612.

[0150] While the foregoing has been referenced Figure 16 to describe a specific hemodynamic monitoring system configuration, it should be readily understood that various hemodynamic monitoring systems and / or other medical monitoring for providing hemodynamic monitoring may be implemented in any of a variety of configurations. Accordingly, the various systems and methods described herein should be understood as not being limited to any particular hemodynamic monitoring system, but rather may be implemented using any of a variety of hemodynamic monitoring or medical monitoring systems capable of measuring PAP and PCWP.

[0151] Example 1. A computational method for performing pulmonary capillary wedge pressure measurement, i.e., PCWP measurement, of an individual and evaluating its quality, comprising:

[0152] acquiring a blood pressure waveform of the individual, wherein the waveform includes blood pressure measurements acquired from a pulmonary artery location and blood pressure measurements acquired from a wedge location, and wherein the blood pressure waveform is acquired using a pulmonary artery catheter;

[0153] determining a PCWP measurement value based on the blood pressure measurement from the wedge location using a computational processing system; and

[0154] using the computational processing system, determining a quality assessment for the PCWP measurement value using a machine learning model configured to provide a quality assessment for the PCWP measurement value based on the blood pressure measurement acquired from the pulmonary artery location and the blood pressure measurement acquired from the wedge location.

[0155] Example 2. The method according to Example 1, wherein the PCWP measurement value and the quality assessment are each determined in real time.

[0156] Example 3. The method according to Example 2, further comprising:

[0157] displaying the PCWP measurement value and the quality assessment on a display screen in digital communication with the computational processing system.

[0158] Example 4. The method according to Example 1, 2, or 3, wherein the computing processing system and the pulmonary artery catheter are part of a hemodynamic monitoring system.

[0159] Example 5. The method according to any one of Examples 1 to 4, wherein acquiring the blood pressure waveform comprises:

[0160] inserting the pulmonary artery catheter into the central vein of the individual,

[0161] guiding the pulmonary artery catheter to the pulmonary artery; and

[0162] inflating a balloon at or near the distal end of the pulmonary artery catheter to allow the pulmonary artery catheter to migrate to the wedge position.

[0163] Example 6. The method according to Example 5, further comprising:

[0164] before inflating the balloon, using the computing processing system to perform real-time assessment of the blood pressure measurements acquired from the pulmonary artery position for artifacts.

[0165] Example 7. The method according to Example 6, wherein the artifacts include one or more of the following: measurements obtained during patient movement, measurements obtained during catheter flushing, and measurements indicating a flat line, measurements indicating underdamping, and measurements indicating overdamping.

[0166] Example 8. The method according to Example 6 or 7, wherein if the PAP maximum (PAP max ) is greater than a threshold, if the PAP minimum (PAP min ) is less than a threshold, and / or if the PAP maximum minus the PAP minimum (PAP max - PAP min ) is greater than a threshold, then detection of an artifact based on flushing, improper zeroing, or patient movement is detected; wherein PAP is the blood pressure measurement acquired from the pulmonary artery position.

[0167] Example 9. The method according to Example 6, 7, or 8, wherein if the PAP maximum minus the PAP minimum (PAP max - PAP min ) is less than a threshold, then detection of an artifact based on a flat line or overdamping is detected; wherein PAP is the blood pressure measurement acquired from the pulmonary artery position.

[0168] Example 10. The method according to any one of Examples 5 to 9, wherein the pulmonary artery catheter comprises a lumen configured to measure the blood pressure.

[0169] Example 11. The method according to any one of Examples 1 to 10, wherein the pulmonary artery catheter is a Swan-Ganz catheter.

[0170] Example 12. The method according to any one of Examples 1 to 11, wherein the machine learning model is based on at least one feature selected from the group consisting of: waveform phase features, determinable features, and morphological features, wherein:

[0171] The waveform phase features are selected from: contractility, pulmonary artery compliance, stroke volume, vascular tone, afterload, and a complete cardiac cycle;

[0172] The determinable features are determined based on the waveform phase features and are selected from: phase average, maximum value, minimum value, duration, area, standard deviation, slope, derivative, trend, deviation from the trend, variation, and variance; and

[0173] The morphological features are based on the waveform features and are selected from frequency components, skewness, and kurtosis.

[0174] Example 13. The method according to Example 12, wherein the machine learning model is based on one or more of the following: approximate entropy, sample entropy, baroreflex sensitivity, variability, and change.

[0175] Example 14. The method according to any one of Examples 1 to 13, wherein collecting the PCWP measurement value includes:

[0176] Determining the respiratory cycle of the individual, wherein determining the PCWP measurement value is based on the respiratory cycle of the individual.

[0177] Example 15. The method according to Example 14, wherein the determination of the PCWP includes determining a blood pressure measurement value from the wedge position at the end of exhalation of the respiratory cycle.

[0178] Example 16. The method according to Example 14, wherein the determination of the PCWP includes averaging blood pressure measurement values from the wedge position over one or more respiratory cycles.

[0179] Example 17. The method according to any one of Examples 1 to 16, further comprising:

[0180] Using the computing processing system to detect in real time a transition from a pulmonary artery position to a wedge position or from a wedge position to a pulmonary artery position by the following steps:

[0181] Extracting one or more hemodynamic features from the blood pressure measurement values at the pulmonary position and the blood pressure measurement values at the wedge position; and

[0182] For each of the one or more hemodynamic features, a fuzzy logic membership function is used to determine a fuzzy logic value based on the value change between the blood pressure measurement at the pulmonary location and the blood pressure measurement at the wedge location.

[0183] Example 18. The method according to Example 17, wherein the one or more features include at least one of the following: statistical moments, percentile values of data, histograms of data, diastolic blood pressure, mean pressure, average pressure, systolic blood pressure, pulse pressure, Shannon entropy, number of peaks above the percentile, number of valleys below the percentile, average number of crossings, area under the curve, singular vector coefficients after principal component analysis, subsampled signals, determinable features, waveform phase features, and morphological features.

[0184] Example 19. The method according to Example 18, wherein the morphological features are selected from: frequency components, skewness, and kurtosis; wherein the waveform phase features are selected from: contractility, pulmonary artery compliance, stroke volume, vascular tone, afterload, and a complete cardiac cycle; and wherein the determinable features are selected from: phase average, maximum value, minimum value, duration, area, standard deviation, slope, derivative, trend, deviation from the trend, variation, and variance.

[0185] Example 20. The method according to Example 18 or 19, wherein the one or more features include average pressure and pulse pressure.

[0186] Example 21. The method according to any one of Examples 1 to 20, further comprising:

[0187] Segmenting the blood pressure measurements collected from the pulmonary artery location and the blood pressure measurements collected from the wedge location into time windows; and

[0188] Extracting features of the blood pressure measurements collected from the pulmonary artery location and the blood pressure measurements collected from the wedge location from the time windows;

[0189] wherein the machine learning model is trained to detect whether the extracted features of the time window are derived from the blood pressure measurements collected from the pulmonary artery location or the blood pressure measurements collected from the wedge location; wherein using the computing processing system, determining a quality assessment for the PCWP measurement using the machine learning model includes:

[0190] Inputting the extracted features from the time window into the machine learning model to generate a quality assessment for each time window; wherein the quality assessment is based on whether the extracted features for each time window can be distinguished as being derived from the pulmonary artery location or the wedge location.

[0191] Example 22. The method according to Example 21, wherein the quality assessment is categorical.

[0192] Example 23. The method according to Example 22, wherein the categorization is a qualitative ranking.

[0193] Example 24. The method according to Example 21, 22 or 23, wherein using the computing processing system, determining the quality assessment for the PCWP measurement using a machine learning model includes:

[0194] Determining whether one or more of the extracted features of the time window are above or below a threshold.

[0195] Example 25. The method according to Example 24, wherein the one or more of the extracted features include: PCWP Mean , PAP Diastolic and PCWP PulsePress , where PCWP Mean is the average of the pressure measurements collected at the wedge position; where PAP Diastolic is the diastolic blood pressure in the pulmonary artery position; and where PCWP PulsePress is the pulse pressure in the wedge position.

[0196] Example 26. The method according to Example 25, wherein high quality is indicated when: PCWP Mean < PAP Diastolic and a × PCWP PulsePress + b × PCWP Mean ≤ c, where a, b, and c are decisive values.

[0197] Example 27. The method according to Example 25 or 26, wherein medium quality is indicated when: PCWP Mean < PAP Diastolic and a × PCWP PulsePress + b × PCWP Mean > c, where a, b, and c are decisive values.

[0198] Example 28. The method according to Example 25, 26 or 27, wherein low quality is indicated when: PCWP Mean ≥ PAP Diastolic .

[0199] Example 29. The method according to any one of Examples 21 to 28 further includes:

[0200] Using the computing processing system to assign its quality assessment to each time window.

[0201] Example 30. The method according to any one of Examples 21 to 29 further includes:

[0202] displaying, on a display screen in digital communication with the computing processing system, the quality assessment of one or more time windows.

[0203] Example 31. The method according to any one of Examples 21 to 30, wherein determining a PCWP measurement value using a computing processing system includes: averaging blood pressure measurements taken from a wedge position of a time window determined to have a quality above a threshold.

[0204] Example 32. The method according to any one of Examples 21 to 31, wherein determining a PCWP measurement value using a computing processing system includes:

[0205] excluding blood pressure measurements taken from a wedge position of a time window determined to have a quality below a threshold, and

[0206] averaging the unexcluded blood pressure measurements taken from a wedge position of a time window.

[0207] Example 33. The method according to Example 31 or 32 further includes:

[0208] displaying, on a display screen in digital communication with the computing processing system, the quality assessment of one or more time windows.

[0209] Example 34. A hemodynamic monitoring system for performing pulmonary capillary wedge pressure measurement, i.e., PCWP measurement, and evaluating its quality, comprising:

[0210] a pulmonary artery catheter configured to collect blood pressure measurements taken from a pulmonary artery position and blood pressure measurements taken from a wedge position, wherein the pulmonary artery catheter includes an inflatable balloon; and

[0211] a computing processing system connected to the pulmonary artery catheter, the computing processing system including:

[0212] a processor system;

[0213] a display screen digitally connected to the processor system; and

[0214] a memory system including one or more applications that can direct the processor system to:

[0215] collect blood pressure measurements in a pulmonary artery position;

[0216] inflate the balloon, wherein inflating the balloon allows the catheter to migrate to a wedge position;

[0217] collect blood pressure measurements in the wedge position;

[0218] Generate a blood pressure waveform based on the blood pressure measurement values collected from the pulmonary artery location and the blood pressure measurement values collected from the wedge location;

[0219] Determine a PCWP measurement value based on the blood pressure measurement value from the wedge location;

[0220] Use a machine learning model to determine a quality assessment for the PCWP measurement value, the machine learning model being configured to provide a quality assessment for the PCWP measurement value based on the blood pressure measurement value collected from the pulmonary artery location and the blood pressure measurement value collected from the wedge location; and

[0221] Display on the display screen one or more of the following: the blood pressure waveform, the PCWP measurement value, or the quality assessment for the PCWP measurement value.

[0222] Example 35. The system according to Example 34, wherein the blood pressure waveform is generated in real time and displayed on the display screen.

[0223] Example 36. The system according to Example 34 or 35, wherein the PCWP measurement value is determined in real time and displayed on the display screen.

[0224] Example 37. The system according to Example 34, 35 or 36, wherein the quality assessment for the PCWP measurement value is determined in real time and displayed on the display screen.

[0225] Example 38. The system according to any one of Examples 34 to 37, wherein the pulmonary artery catheter includes a lumen for measuring blood pressure and an inflatable balloon that allows the pulmonary artery catheter to migrate to the wedge position.

[0226] Example 39. The system according to any one of Examples 34 to 38, wherein the catheter is a Swan-Ganz catheter.

[0227] Example 40. The system according to any one of Examples 34 to 39, wherein the one or more applications can direct the processor system:

[0228] Perform a real-time assessment of the blood pressure measurement values collected from the pulmonary artery location for artifacts before inflating the balloon.

[0229] Example 41. The system according to Example 40, wherein the artifacts include one or more of the following: measurements obtained during patient movement, measurements obtained during catheter flushing, and measurements indicating a flat line, measurements indicating underdamping, and measurements indicating overdamping.

[0230] Example 42. The system according to Example 40 or 41, wherein if the PAP maximum value (PAP max ) is greater than a threshold value, if the PAP minimum value (PAP min ) is less than a threshold value, and / or if the PAP maximum value minus the PAP minimum value (PAP max - PAP min ) is greater than a threshold value, then detection of an artifact based on flushing, zeroing impropriety, or patient movement is detected; wherein PAP is the measured blood pressure value collected from the pulmonary artery location.

[0231] Example 43. The system according to Example 40, 41, or 42, wherein if the PAP maximum value minus the PAP minimum value (PAP max - PAP min ) is less than a threshold value, then detection of an artifact based on a flat line or excessive damping is detected; wherein PAP is the measured blood pressure value collected from the pulmonary artery location.

[0232] Example 44. The system according to any one of Examples 34 to 43, wherein the machine learning model is based on at least one feature selected from the group consisting of: waveform phase features, determinable features, and morphological features, wherein:

[0233] The waveform phase features are selected from: contractility, pulmonary artery compliance, stroke volume, vascular tone, afterload, and a complete cardiac cycle;

[0234] The determinable features are determined based on the waveform phase features and are selected from: phase average, maximum value, minimum value, duration, area, standard deviation, slope, derivative, trend, deviation from the trend, variation, and variance; and

[0235] The morphological features are based on the waveform features and are selected from frequency components, skewness, and kurtosis.

[0236] Example 45. The system according to Example 44, wherein the machine learning model is based on one or more of the following: approximate entropy, sample entropy, baroreflex sensitivity, variability, and change.

[0237] Example 46. The system according to any one of Examples 34 to 45, wherein the one or more applications can direct the processor system to:

[0238] Determine a respiratory cycle, wherein the determination of the PCWP measurement is based on the respiratory cycle of the individual.

[0239] Example 47. The system according to Example 46, wherein the one or more applications can direct the processor system to: determine a blood pressure measurement from the wedge position at the end of exhalation of a respiratory cycle to generate the PCWP measurement.

[0240] Example 48. The system according to Example 46, wherein the one or more applications can direct the processor system to: determine blood pressure measurements from the wedge position across one or more respiratory cycles.

[0241] Example 49. The system according to any one of Examples 34 to 48, wherein the one or more applications can direct the processor system to:

[0242] detect in real time a transition from a pulmonary artery position to a wedge position or from a wedge position to a pulmonary artery position by the following steps:

[0243] extract one or more hemodynamic features from the blood pressure measurements at the lung position and the blood pressure measurements at the wedge position; and

[0244] for each of the one or more hemodynamic features, use a fuzzy logic membership function to determine a fuzzy logic value based on a change in value between the blood pressure measurements at the lung position and the blood pressure measurements at the wedge position.

[0245] Example 50. The system according to Example 49, wherein the one or more features include at least one of the following: statistical moments, percentile values of data, histograms of data, diastolic blood pressure, mean pressure, mean arterial pressure, systolic blood pressure, pulse pressure, Shannon entropy, number of peaks above a percentile, number of valleys below a percentile, mean number of crossings, area under the curve, singular vector coefficients after principal component analysis, subsampled signals, determinable features, waveform phase features, and morphological features.

[0246] Example 51. The system according to Example 50, wherein the morphological features are selected from: frequency components, skewness, and kurtosis; wherein the waveform phase features are selected from: contractility, pulmonary artery compliance, stroke volume, vascular tone, afterload, and a complete cardiac cycle; and wherein the determinable features are selected from: phase mean, maximum value, minimum value, duration, area, standard deviation, slope, derivative, trend, deviation from trend, variation, and variance.

[0247] Example 52. The system according to Example 50 or 51, wherein the one or more features include mean pressure and pulse pressure.

[0248] Example 53. The system according to any one of Examples 34 to 52, wherein the one or more applications can direct the processor system to:

[0249] Segment the blood pressure measurements taken from the pulmonary artery location and the blood pressure measurements taken from the wedge location into time windows; and

[0250] Extract features of the blood pressure measurements taken from the pulmonary artery location and the blood pressure measurements taken from the wedge location from the time windows; and

[0251] Input the extracted features from the time windows into the machine learning model to generate a quality assessment for each time window; wherein the quality assessment is based on whether the extracted features for each time window can be distinguished as being derived from the pulmonary artery location or from the wedge location to generate the quality assessment for the PCWP, wherein the machine learning model is trained to detect whether the extracted features of the time window are derived from the blood pressure measurements taken from the pulmonary artery location or from the blood pressure measurements taken from the wedge location.

[0252] Example 54. The system according to Example 53, wherein the quality assessment is categorical.

[0253] Example 55. The system according to Example 54, wherein the categorization is a qualitative ranking.

[0254] Example 56. The system according to Example 53, 54 or 55, wherein the one or more applications can direct the processor system: to determine whether one or more of the extracted features of the time window are above or below a threshold to generate the quality assessment for the PCWP measurement.

[0255] Example 57. The system according to Example 56, wherein the one or more extracted features include: PCWP Mean , PAP Diastolic and PCWP PulsePress , where PCWP Mean is the average of the pressure measurements taken in the wedge location; where PAP Diastolic is the diastolic blood pressure in the pulmonary artery location; and where PCWP PulsePress is the pulse pressure in the wedge location.

[0256] Example 58. The system according to Example 57, wherein high quality is indicated when: PCWP Mean < PAP Diastolic and a × PCWP PulsePress + b × PCWP Mean ≤ c, where a, b and c are decisive values.

[0257] Example 59. The system according to Example 57 or 58, wherein medium quality is indicated when: PCWP Mean <PAP Diastolic and a × PCWP PulsePress + b × PCWP Mean > c, where a, b, and c are decisive values.

[0258] Example 60. The system according to Example 57, 58, or 59, wherein low quality is indicated when: PCWP Mean ≥ PAP Diastolic .

[0259] Example 61. The system according to any one of Examples 53 to 60, wherein the one or more applications can direct the processor system to:

[0260] Assign its quality assessment to each time window.

[0261] Example 62. The system according to any one of Examples 53 to 61, wherein the one or more applications can direct the processor system to:

[0262] Display the quality assessment of one or more time windows on the display screen.

[0263] Example 63. The system according to any one of Examples 53 to 62, wherein the one or more applications can direct the processor system to:

[0264] Average the blood pressure measurements taken from the wedge position for time windows determined to have a quality above the threshold to produce the PCWP measurement.

[0265] Example 64. The system according to any one of Examples 53 to 63, wherein the one or more applications can direct the processor system to:

[0266] Exclude the blood pressure measurements taken from the wedge position for time windows determined to have a quality below the threshold, and

[0267] Average the unexcluded blood pressure measurements taken from the wedge position for the time windows to produce the PCWP measurement.

[0268] Example 65. The system according to any one of Examples 53 to 64, wherein the one or more applications can direct the processor system to:

[0269] Display the quality assessment of one or more time windows on a display screen in digital communication with the computing processing system.

[0270] Example 66. A calculation method for performing pulmonary capillary wedge pressure measurement (PCWP measurement) of an individual, comprising:

[0271] Collecting a blood pressure waveform of an individual using the pulmonary artery catheter, wherein the waveform includes blood pressure measurement values collected from a pulmonary artery location, and wherein the pulmonary artery catheter is connected to a hemodynamic monitoring system;

[0272] While collecting blood pressure measurement values from the pulmonary artery location, using the hemodynamic monitoring system to evaluate the blood pressure measurement values collected from the pulmonary artery location for artifacts;

[0273] After determining that the blood pressure measurement values collected from the pulmonary artery location have no artifacts, using the hemodynamic monitoring system to inflate the balloon at or near the distal end of the pulmonary artery catheter to allow the pulmonary artery catheter to migrate to the wedge position;

[0274] Further collecting a blood pressure waveform of the individual using the pulmonary artery catheter, wherein the waveform includes blood pressure measurement values collected from the wedge position; and

[0275] Using the hemodynamic monitoring system to determine a PCWP measurement value based on the blood pressure measurement values from the wedge position.

[0276] Example 67. The method according to Example 66, wherein the PCWP measurement value is determined in real time.

[0277] Example 68. The method according to Example 67, further comprising:

[0278] Displaying the PCWP measurement value on a display screen of a hemodynamic processing system.

[0279] Example 69. The method according to Example 66, 67 or 68, wherein the artifacts include one or more of the following: measurement values obtained during patient movement, measurement values obtained during catheter flushing, and measurement values indicating a flat line, measurement values indicating insufficient damping, and measurement values indicating excessive damping.

[0280] Example 70. The method according to any one of Examples 66 to 69, wherein if the maximum pulmonary artery pressure (PAP max ) is greater than a threshold, if the minimum pulmonary artery pressure (PAP min ) is less than a threshold, and / or if the difference between the maximum pulmonary artery pressure and the minimum pulmonary artery pressure (PAP max - PAP min ) is greater than a threshold, then detection of artifacts due to flushing, improper zeroing, or patient movement is detected; wherein PAP is the blood pressure measurement value collected from the pulmonary artery location.

[0281] Example 71. The method according to any one of Examples 66 to 70, wherein if the maximum PAP minus the minimum PAP (PAP max - PAP min ) is less than a threshold, detection of an artifact based on a flat line or excessive damping is detected; wherein PAP is the blood pressure measurement value collected from the pulmonary artery location.

[0282] Example 72. The method according to any one of Examples 66 to 71, wherein the pulmonary artery catheter includes a lumen configured to measure the blood pressure.

[0283] Example 73. The method according to any one of Examples 66 to 72, wherein the pulmonary artery catheter is a Swan - Ganz catheter.

[0284] Example 74. A hemodynamic monitoring system for performing pulmonary capillary wedge pressure measurement, i.e., PCWP measurement, comprising:

[0285] A pulmonary artery catheter configured to collect a blood pressure measurement value collected from the pulmonary artery location and a blood pressure measurement value collected from the wedge location, wherein the pulmonary artery catheter includes an inflatable balloon; and

[0286] A computing and processing system connected to the pulmonary artery catheter, the computing and processing system comprising:

[0287] A processor system;

[0288] A display screen digitally connected to the processor system; and

[0289] A memory system including one or more applications that can direct the processor system to:

[0290] Collect a blood pressure measurement value at the pulmonary artery location;

[0291] Evaluate the blood pressure measurement value collected from the pulmonary artery location for artifacts;

[0292] After determining that the blood pressure measurement value collected from the pulmonary artery location has no artifacts, inflate the balloon, wherein inflating the balloon allows the catheter to migrate to the wedge location;

[0293] Collect a blood pressure measurement value at the wedge location;

[0294] Generate a blood pressure waveform based on the blood pressure measurement value collected from the pulmonary artery location and the blood pressure measurement value collected from the wedge location;

[0295] Determine a PCWP measurement value based on the blood pressure measurement value from the wedge location; and

[0296] Display the PCWP measurement value on the display screen.

[0297] Example 75. The system according to Example 74, wherein the PCWP measurement is determined in real time.

[0298] Example 76. The system according to Example 74 or 75, wherein the artifacts include one or more of the following: measurements obtained during patient movement, measurements obtained during catheter flushing, and measurements indicating a flat line, measurements indicating insufficient damping, and measurements indicating excessive damping.

[0299] Example 77. The system according to Example 74, 75, or 76, wherein if the PAP maximum value (PAP max ) is greater than a threshold, if the PAP minimum value (PAP min ) is less than a threshold, and / or if the PAP maximum value minus the PAP minimum value (PAP max - PAP min ) is greater than a threshold, then detection of artifacts based on flushing, improper zeroing, or patient movement is detected; wherein PAP is the blood pressure measurement collected from the pulmonary artery location.

[0300] Example 78. The system according to any one of Examples 74 to 77, wherein if the PAP maximum value minus the PAP minimum value (PAP max - PAP min ) is less than a threshold, then detection of artifacts based on a flat line or excessive damping is detected; wherein PAP is the blood pressure measurement collected from the pulmonary artery location.

[0301] Example 79. The system according to any one of Examples 74 to 78, wherein the pulmonary artery catheter includes a lumen configured to measure the blood pressure.

[0302] Example 80. The system according to any one of Examples 74 to 79, wherein the pulmonary artery catheter is a Swan - Ganz catheter.

[0303] Example 81. A computational method for detecting a transition between a pulmonary artery location and a wedge location, comprising:

[0304] Collecting a blood pressure waveform of an individual using the pulmonary artery catheter, wherein the waveform includes a blood pressure measurement collected from the pulmonary artery location, and wherein the pulmonary artery catheter is connected to a hemodynamic monitoring system;

[0305] Inflating a balloon at or near the distal end of the pulmonary artery catheter using the hemodynamic monitoring system to allow the pulmonary artery catheter to migrate to the wedge location;

[0306] Detecting a transition from the pulmonary artery location to the wedge location using the hemodynamic monitoring system;

[0307] Using the pulmonary artery catheter to further collect a blood pressure waveform of an individual, wherein the waveform includes blood pressure measurements collected from the wedge position;

[0308] Using the hemodynamic monitoring system to deflate the balloon at or near the distal end of the pulmonary artery catheter to allow the pulmonary artery catheter to migrate back to the pulmonary artery position; and

[0309] Using the hemodynamic monitoring system to detect the transition from the wedge position to the pulmonary artery position.

[0310] Example 82. The method according to Example 81, wherein detecting the transition from the pulmonary artery position to the wedge position and detecting the transition from the wedge position to the pulmonary artery position each include:

[0311] Extracting one or more hemodynamic features from the blood pressure measurements in the pulmonary position and the blood pressure measurements in the wedge position; and

[0312] For each of the one or more hemodynamic features, using a fuzzy logic membership function to determine a fuzzy logic value based on the value change between the blood pressure measurements in the pulmonary position and the blood pressure measurements in the wedge position.

[0313] Example 83. The method according to Example 82, wherein the one or more features include at least one of the following: statistical moments, percentile values of data, histograms of data, diastolic blood pressure, mean pressure, mean arterial pressure, systolic blood pressure, pulse pressure, Shannon entropy, number of peaks above the percentile, number of valleys below the percentile, average number of crossings, area under the curve, singular vector coefficients after principal component analysis, subsampled signals, determinable features, waveform phase features, and morphological features.

[0314] Example 84. The method according to Example 82 or 83, wherein the morphological features are selected from: frequency components, skewness, and kurtosis; wherein the waveform phase features are selected from: contractility, pulmonary artery compliance, stroke volume, vascular tone, afterload, and a complete cardiac cycle; and wherein the determinable features are selected from: phase mean, maximum value, minimum value, duration, area, standard deviation, slope, derivative, trend, deviation from the trend, variation, and variance.

[0315] Example 85. The method according to Example 82, 83, or 84, wherein the one or more features include mean pressure and pulse pressure.

[0316] Example 86. The method according to any one of Examples 81 to 85, further comprising:

[0317] Using the hemodynamic monitoring system, a pulmonary capillary wedge pressure measurement value, i.e., a PCWP measurement value, is determined based on the blood pressure measurement value from the wedge position.

[0318] Example 87. The method according to any one of Examples 81 to 86, wherein the pulmonary artery catheter includes a lumen configured to measure the blood pressure.

[0319] Example 88. The method according to any one of Examples 81 to 88, wherein the pulmonary artery catheter is a Swan-Ganz catheter.

[0320] Example 89. A hemodynamic monitoring system for detecting a transition between a pulmonary artery position and a wedge position, comprising:

[0321] A pulmonary artery catheter configured to collect a blood pressure measurement value collected from the pulmonary artery position and a blood pressure measurement value collected from the wedge position, wherein the pulmonary artery catheter includes an inflatable balloon; and

[0322] A computing and processing system connected to the pulmonary artery catheter, the computing and processing system comprising:

[0323] A processor system;

[0324] A display screen digitally connected to the processor system; and

[0325] A memory system including one or more applications that can direct the processor system to:

[0326] Collect a blood pressure measurement value in the pulmonary artery position;

[0327] Inflate the balloon, wherein inflating the balloon allows the catheter to migrate to the wedge position;

[0328] Detect the transition from the pulmonary artery position to the wedge position;

[0329] Collect a blood pressure measurement value in the wedge position;

[0330] Deflate the balloon at or near the distal end of the pulmonary artery catheter to allow the pulmonary artery catheter to migrate back to the pulmonary artery position; and

[0331] Detect the transition from the wedge position to the pulmonary artery position.

[0332] Example 90. The system according to Example 89, wherein the one or more applications can direct the processor system to:

[0333] Extract one or more hemodynamic features from the blood pressure measurement value in the lung position and the blood pressure measurement value in the wedge position; and

[0334] For each of the one or more hemodynamic characteristics, a fuzzy logic membership function is used to determine a fuzzy logic value based on the value change between the blood pressure measurement at the pulmonary position and the blood pressure measurement at the wedge position, where the fuzzy logic value is used to detect the transition from the pulmonary artery position to the wedge position and to detect the transition from the wedge position to the pulmonary artery position.

[0335] Example 91. The system according to Example 90, wherein the one or more characteristics include at least one of the following: statistical moments, percentile values of data, histograms of data, diastolic blood pressure, median pressure, mean pressure, systolic blood pressure, pulse pressure, Shannon entropy, number of peaks above the percentile, number of valleys below the percentile, average number of crossings, area under the curve, singular vector coefficients after principal component analysis, subsampled signals, determinable characteristics, waveform phase characteristics, and morphological characteristics.

[0336] Example 92. The system according to Example 90 or 91, wherein the morphological characteristics are selected from: frequency components, skewness, and kurtosis; wherein the waveform phase characteristics are selected from: contractility, pulmonary artery compliance, stroke volume, vascular tone, afterload, and the complete cardiac cycle; and wherein the determinable characteristics are selected from: phase average, maximum value, minimum value, duration, area, standard deviation, slope, derivative, trend, deviation from the trend, variation, and variance.

[0337] Example 93. The system according to Example 90, 91, or 92, wherein the one or more characteristics include mean pressure and pulse pressure.

[0338] Example 94. The system according to any one of Examples 89 to 93, wherein the one or more applications can direct the processor system to:

[0339] Determine a pulmonary capillary wedge pressure measurement, i.e., a PCWP measurement, based on the blood pressure measurement from the wedge position; and

[0340] Display the pulmonary capillary wedge pressure, i.e., the PCWP, on the display screen.

[0341] Example 95. The system according to any one of Examples 89 to 94, wherein the pulmonary artery catheter includes a lumen configured to measure the blood pressure.

[0342] Example 96. The system according to any one of Examples 89 to 95, wherein the pulmonary artery catheter is a Swan - Ganz catheter.

Claims

1. A computational method for performing pulmonary capillary wedge pressure (PCWP) measurement on an individual and evaluating the quality thereof, the method comprising: acquiring a blood pressure waveform of the individual, wherein the waveform includes blood pressure measurements acquired from a pulmonary artery location and blood pressure measurements acquired from a wedge location, and wherein the blood pressure waveform is acquired using a pulmonary artery catheter; using a computational processing system to determine a PCWP measurement value based on the blood pressure measurements from the wedge location; and using the computational processing system to determine a quality assessment for the PCWP measurement value using a machine learning model, the machine learning model being configured to provide a quality assessment for the PCWP measurement value based on the blood pressure measurements acquired from the pulmonary artery location and the blood pressure measurements acquired from the wedge location.

2. The method according to claim 1, wherein the PCWP measurement value and the quality assessment are each determined in real time.

3. The method according to claim 2, further comprising: displaying the PCWP measurement value and the quality assessment on a display screen in digital communication with the computational processing system.

4. The method according to claim 1, 2 or 3, wherein the computational processing system and the pulmonary artery catheter are part of a hemodynamic monitoring system.

5. The method according to any one of claims 1 to 4, wherein acquiring the blood pressure waveform comprises: inserting the pulmonary artery catheter into the central vein of the individual, guiding the pulmonary artery catheter to the pulmonary artery; and inflating a balloon at or near the distal end of the pulmonary artery catheter to allow the pulmonary artery catheter to migrate to the wedge location.

6. The method according to claim 5, further comprising: before inflating the balloon, using the computational processing system to perform real-time assessment of the blood pressure measurements acquired from the pulmonary artery location for artifacts.

7. The method according to any one of claims 1 to 6, further comprising: using the computational processing system to detect in real time a transition from the pulmonary artery location to the wedge location or from the wedge location to the pulmonary artery location by the following steps: extracting one or more hemodynamic features from the blood pressure measurements in the pulmonary location and the blood pressure measurements in the wedge location; and for each of the one or more hemodynamic features, using a fuzzy logic membership function to determine a fuzzy logic value based on the value change between the blood pressure measurements in the pulmonary location and the blood pressure measurements in the wedge location.

8. The method according to claim 7, wherein the one or more features include mean pressure and pulse pressure.

9. The method according to any one of claims 1 to 8, further comprising: segmenting the blood pressure measurements acquired from the pulmonary artery location and the blood pressure measurements acquired from the wedge location into time windows; and extracting features of the blood pressure measurements acquired from the pulmonary artery location and the blood pressure measurements acquired from the wedge location from the time windows. wherein the machine learning model is trained to detect whether the extracted features of the time window are derived from the blood pressure measurements collected from the pulmonary artery location or the blood pressure measurements collected from the wedge location; wherein using the computing processing system, determining a quality assessment for the PCWP measurement using a machine learning model includes: inputting the extracted features from the time window into the machine learning model to generate a quality assessment for each time window; wherein the quality assessment is based on whether the extracted features for each time window can be distinguished as being derived from the pulmonary artery location or the wedge location.

10. The method according to claim 9, wherein using the computing processing system, determining a quality assessment for the PCWP measurement using a machine learning model includes: determining whether one or more extracted features of a time window are above or below a threshold.

11. The method according to claim 10, wherein the one or more extracted features include: PCWP Mean , PAP Diastolic and PCWP PulsePress , where PCWP Mean is the average of the pressure measurements taken at the wedge position; where PAP Diastolic is the diastolic blood pressure at the pulmonary artery position; and where PCWP PulsePress is the pulse pressure at the wedge position.

12. The method according to claim 11, wherein high quality is indicated when: PCWP Mean <PAP Diastolic and a × PCWP PulsePress + b × PCWP Mean ≤ c, where a, b, and c are decisive values.

13. The method according to claim 11 or 12, wherein medium quality is indicated when: PCWP Mean <PAP Diastolic and a × PCWP PulsePress + b × PCWP Mean > c, where a, b, and c are decisive values.

14. The method according to claim 11, 12 or 13, wherein low quality is indicated when: PCWP Mean ≥ PAP Diastolic .

15. The method according to any one of claims 9 to 14, wherein determining the PCWP measurement using a computing processing system includes: averaging the blood pressure measurements collected from the wedge location for time windows determined to have a quality above the threshold.

16. A hemodynamic monitoring system for performing pulmonary capillary wedge pressure measurement, i.e., PCWP measurement, and evaluating its quality, comprising: a pulmonary artery catheter configured to collect blood pressure measurements from a pulmonary artery location and blood pressure measurements from a wedge location, wherein the pulmonary artery catheter includes an inflatable balloon; and a computing processing system connected to the pulmonary artery catheter, the computing processing system including: a processor system; a display screen digitally connected to the processor system; and a memory system including one or more applications that can direct the processor system to: collect blood pressure measurements in the pulmonary artery location; inflate the balloon, wherein inflating the balloon allows the catheter to migrate to the wedge location; collect blood pressure measurements in the wedge location; generate a blood pressure waveform based on the blood pressure measurements collected from the pulmonary artery location and the blood pressure measurements collected from the wedge location; determine a PCWP measurement based on the blood pressure measurements from the wedge location; determine a quality assessment for the PCWP measurement using a machine learning model, the machine learning model being configured to provide a quality assessment for the PCWP measurement based on the blood pressure measurements collected from the pulmonary artery location and the blood pressure measurements collected from the wedge location; and display on the display screen one or more of: the blood pressure waveform, the PCWP measurement, or the quality assessment for the PCWP measurement.

17. The system according to claim 16, wherein the blood pressure waveform is generated and displayed on the display screen in real time.

18. The system according to claim 16 or 17, wherein the PCWP measurement value is determined and displayed on the display screen in real time.

19. The system according to claim 16, 17 or 18, wherein the quality assessment for the PCWP measurement value is determined and displayed on the display screen in real time.

20. The system according to any one of claims 16 to 19, wherein the one or more applications are capable of instructing the processor system: Before inflating the balloon, to evaluate in real time the blood pressure measurements collected from the pulmonary artery location for artifacts.

21. The system according to any one of claims 16 to 20, wherein the one or more applications are capable of instructing the processor system: To detect in real time the transition from the pulmonary artery location to the wedge location or from the wedge location to the pulmonary artery location by the following steps: Extracting one or more hemodynamic features from the blood pressure measurements in the pulmonary location and the blood pressure measurements in the wedge location; and For each of the one or more hemodynamic features, using a fuzzy logic membership function to determine a fuzzy logic value based on the change in the value between the blood pressure measurements in the pulmonary location and the blood pressure measurements in the wedge location.

22. The system according to claim 21, wherein the one or more features include mean pressure and pulse pressure.

23. The system according to any one of claims 16 to 22, wherein the one or more applications can instruct the processor system: To segment the blood pressure measurements collected from the pulmonary artery location and the blood pressure measurements collected from the wedge location into time windows; and To extract the features of the blood pressure measurements collected from the pulmonary artery location and the blood pressure measurements collected from the wedge location from the time windows ； and To input the extracted features from the time windows into the machine learning model to generate a quality assessment for each time window; wherein the quality assessment is based on whether the extracted features for each time window can be distinguished as being derived from the pulmonary artery location or from the wedge location to generate the quality assessment for the PCWP, wherein the machine learning model is trained to detect whether the extracted features of the time window are derived from the blood pressure measurements collected from the pulmonary artery location or from the wedge location.

24. The system according to claim 23, wherein the one or more applications are capable of instructing the processor system: To determine whether one or more of the extracted features of the time window are above or below a threshold to generate the quality assessment for the PCWP measurement value.

25. The system according to claim 24, wherein the one or more extracted features include: PCWP Mean , PAP Diastolic and PCWP PulsePress , where PCWP Mean is the average of the pressure measurements taken at the wedge position; where PAP Diastolic is the diastolic blood pressure at the pulmonary artery position; and where PCWP PulsePress is the pulse pressure at the wedge position.

26. The system according to claim 25, wherein high quality is indicated in the following cases: PCWP Mean <PAP Diastolic and a × PCWP PulsePress + b × PCWP Mean ≤ c, where a, b, and c are decisive values.

27. The system according to claim 25 or 26, wherein medium quality is indicated in the following cases: PCWP Mean <PAP Diastolic and a × PCWP PulsePress + b × PCWP Mean > c, where a, b, and c are decisive values.

28. The system according to claim 25, 26 or 27, wherein low quality is indicated in the following cases: PCWP Mean ≥ PAP Diastolic .

29. The system according to any one of claims 23 to 28, wherein the one or more applications are capable of instructing the processor system: To display on the display screen the quality assessment of one or more time windows.

30. The system according to any one of claims 23 to 29, wherein the one or more applications are capable of instructing the processor system: To average the blood pressure measurements taken from the wedge position for time windows determined to have a quality above a threshold to produce the PCWP measurement.

Images