Hemodynamic monitors for triaging patients with low ejection fraction

Abstract

The hemodynamic monitor includes an integrated hardware unit with a noninvasive blood pressure sensor and a system processor, a system memory, and a display with a user interface. The system memory includes instructions configured to: adjust, by a pressure controller, the pressure in an inflatable blood pressure bladder to maintain a constant arterial volume of the patient for a period of time, generate arterial pressure waveform data for the patient based on the adjusted pressure in the inflatable blood pressure bladder over the period of time, extract a plurality of signal measurements from the arterial pressure waveform data for the patient, extract input features indicative of an ejection fraction score for the patient from the plurality of signal measurements, and determine the ejection fraction score for the patient based on the extracted input features.

Description

[Technical Field]

[0001] CROSS-REFERENCE TO RELATED APPLICATIONS This application claims the benefit of U.S. Provisional Application No. 63 / 375,843, filed September 15, 2022, entitled "HEMODYNAMIC MONITOR FOR TRIAGING PATIENTS WITH LOW EJECTION FRACTION OR AORTIC STENOSIS," the disclosure of which is incorporated herein by reference in its entirety.

[0002] The present disclosure relates generally to ejection fraction, and more particularly to measuring a patient's ejection fraction to triage the patient for treatment. [Background technology]

[0003] The ejection fraction (EF) is a measure of the amount of blood pumped out of the ventricles with each contraction. The EF basically compares the volume of blood in the ventricles to the volume of blood pumped out of the ventricles. The left ventricular EF is the ejection fraction of the left heart and indicates how effectively the heart pumps blood into the systemic circulatory system. Traditionally, a patient's EF is measured through imaging tests, such as an echocardiogram, a multi-gated (MUGA) scan, or a computed tomography (CT) scan. Other tests used to determine EF include cardiac catheterization and nuclear stress testing. Each of these tests requires highly trained professionals to perform the test and interpret the results. Therefore, patients must visit a cardiologist or other cardiovascular specialist for initial cardiac screening. These tests can be expensive, and it can take days or weeks to inform patients of their EF. A solution is needed to make it easier for patients to undergo EF screening with fewer trips. Preferably, this solution also reduces the amount of time a patient must wait to receive the results of their ejection fraction screening, allowing the patient to seek further testing and / or treatment sooner. Summary of the Invention [Means for solving the problem]

[0004] In one example, a hemodynamic monitor for detecting heart failure is disclosed. The hemodynamic monitor includes a noninvasive blood pressure sensor having an inflatable blood pressure bladder, a pressure controller pneumatically connected to the inflatable blood pressure bladder, and an optical transmitter and an optical receiver electrically connected to the pressure controller. The hemodynamic monitor also includes an integrated hardware unit with a system processor, a system memory, and a display with a user interface. The system memory includes instructions that, when executed by the system processor, are configured to: adjust, by the pressure controller, the pressure in the inflatable blood pressure bladder to keep the patient's arterial volume constant for a period of time based on a feedback signal generated by the optical transmitter and the optical receiver; generate arterial pressure waveform data for the patient based on the adjusted pressure in the inflatable blood pressure bladder over the period of time; extract a plurality of signal measurements from the patient's arterial pressure waveform data; extract input features indicative of the patient's ejection fraction score from the plurality of signal measurements; determine the patient's ejection fraction score based on the extracted input features; generate a first sensory alert signal configured to generate a first sensory alert indicating that the patient's ejection fraction is low when the ejection fraction score exceeds a threshold score, or generate a second sensory alert signal configured to generate a second sensory alert indicating that the patient's ejection fraction is not low when the ejection fraction score is below the threshold score; transmit the first sensory alert signal or the second sensory alert signal to a user interface; and output the first sensory alert or the second sensory alert through the user interface.

[0005] In another example, a hemodynamic monitor for detecting heart failure is disclosed. The hemodynamic monitor includes an arterial blood pressure sensor with a housing, a fluid input port connected to a fluid source via tubing, a catheter-side fluid port connected to a catheter inserted into a patient's arterial system, a pressure transducer in communication with the fluid source through the fluid port, and an I / O cable in electrical communication with the pressure transducer. The hemodynamic monitor also includes an integrated hardware unit with a system processor, system memory, a display with a user interface, and an analog-to-digital converter (ADC). The system memory includes instructions that, when executed by the system processor, are configured to: receive from a pressure transducer an electrical signal based on pressure in the patient's arterial system transmitted through the fluid source over a period of time; convert the electrical signal to a digital signal; generate arterial pressure waveform data for the patient based on the digital signal; extract a plurality of signal measurements from the arterial pressure waveform data; extract input features from the plurality of signal measurements indicative of the patient's ejection fraction score; determine the patient's ejection fraction score based on the extracted input features; generate a first sensory alert signal configured to generate a first sensory alert indicating that the patient's ejection fraction is low when the ejection fraction score exceeds a threshold score, or generate a second sensory alert signal configured to generate a second sensory alert indicating that the patient's ejection fraction is not low when the ejection fraction score is below the threshold score; transmit the first sensory alert signal or the second sensory alert signal to a user interface; and output the first sensory alert or the second sensory alert through the user interface.

[0006] In a further example, a method for triaging a patient for risk of heart failure is disclosed. The method includes receiving, by a hemodynamic monitor, sensed hemodynamic data representing the patient's arterial pressure waveform. The hemodynamic monitor performs waveform analysis of the sensed hemodynamic data to calculate a plurality of signal measurements of the sensed hemodynamic data and extracts input features from the plurality of signal measurements indicative of the patient's ejection fraction. The hemodynamic monitor further determines the patient's ejection fraction based on the input features and outputs the patient's ejection fraction to a display. The hemodynamic monitor alerts the patient or a healthcare professional that the ejection fraction is low when the ejection fraction is 40 percent or less. [Brief explanation of the drawings]

[0007] [Figure 1] FIG. 1 is a perspective view of an exemplary hemodynamic monitor that analyzes a patient's arterial pressure and provides a medical professional with the patient's ejection fraction and heart failure risk score. [Figure 2] FIG. 1 is a perspective view of an exemplary minimally invasive pressure sensor for sensing hemodynamic data representative of a patient's arterial pressure. [Figure 3] FIG. 1 is a perspective view of an exemplary non-invasive sensor for sensing hemodynamic data representative of a patient's arterial blood pressure. [Figure 4] FIG. 1 is a block diagram illustrating an exemplary hemodynamic monitoring system that determines a patient's ejection fraction based on a set of input features derived from signal measurements of the patient's arterial pressure waveform. [Figure 5] FIG. 1 is a schematic diagram of a method for triaging a patient's ejection fraction. [Figure 6] 1 is a diagram of a first clinical data set, a second clinical data set, and a third clinical data set used for data mining and machine training of a hemodynamic monitoring system. [Figure 7] 1 is a flow diagram for extracting a set of input features derived from signal measurements of a patient's arterial pressure waveform for training a machine learning model of a hemodynamic monitoring system. [Figure 8]1 is a graph illustrating an example trajectory of an arterial pressure waveform including example indices corresponding to signal measurements used to extract input features for determining a patient's ejection fraction. DETAILED DESCRIPTION OF THE INVENTION

[0008] As described herein, a hemodynamic monitoring system detects a patient's ejection fraction using the patient's arterial waveform. The hemodynamic monitoring system uses machine learning to extract a set of input features from the patient's arterial pressure. The set of input features is used by the hemodynamic monitoring system to determine the patient's ejection fraction during a visit to a primary care physician's office, in an emergency medical situation, or in any other patient care setting.

[0009] Depending on the ejection fraction measured by the hemodynamic monitoring system, the hemodynamic monitoring system may issue a signal or alarm to a medical professional and / or patient to alert the medical professional and / or patient that the patient's ejection fraction is low and the patient is at high risk for heart failure. Hemodynamic monitoring systems are described in more detail below with reference to Figures 1-8.

[0010] FIG. 1 is a perspective view of a hemodynamic monitor 10 capable of determining a patient's ejection fraction. As shown in FIG. 1 , hemodynamic monitor 10 includes a display 12 that presents a graphical user interface including control elements (e.g., graphical control elements) that enable user interaction with hemodynamic monitor 10 in the example of FIG. 1 . As described further below, hemodynamic monitor 10 may also include multiple input and / or output (I / O) connectors configured for wired connection (e.g., electrical and / or communication connection) with one or more peripheral components, such as one or more hemodynamic sensors. For example, as shown in FIG. 1 , hemodynamic monitor 10 may include I / O connector 14. While the example of FIG. 1 shows five separate I / O connectors 14, it should be understood that in other examples, hemodynamic monitor 10 may include fewer than five I / O connectors or more than five I / O connectors. In still other examples, hemodynamic monitor 10 may not include I / O connector 14 but rather may communicate wirelessly with various peripheral devices.

[0011] As described further below, hemodynamic monitor 10 includes one or more processors and computer-readable memory storing ejection fraction software code executable to determine a patient's ejection fraction measurement based on the patient's sensed hemodynamic data. Hemodynamic monitor 10 can receive sensed hemodynamic data representing the patient's arterial pressure waveform, such as via one or more hemodynamic sensors connected to hemodynamic monitor 10 via I / O connector 14. As described further below, hemodynamic monitor 10 executes the ejection fraction software code to obtain, using the sensed hemodynamic data, one or more vital sign parameters characterizing the patient's vital sign data, as well as multiple ejection fraction profiling parameters (e.g., input features), which may include difference parameters and composite parameters derived from the one or more vital sign parameters.

[0012] 1, hemodynamic monitor 10 can present a graphical user interface on display 12. Display 12 can be a liquid crystal display (LCD), a light emitting diode (LED) display, an organic light emitting diode (OLED) display, or other display device suitable for providing information to a user in a graphical format. In some examples, such as the example of FIG. 1, display 12 can be a touch-sensitive and / or presence-sensitive display device configured to receive user input in the form of gestures, such as touch gestures, scroll gestures, zoom gestures, swipe gestures, or other gesture inputs.

[0013] 2 is a perspective view of a hemodynamic sensor 16 that may be attached to a patient for sensing hemodynamic data representative of the patient's arterial blood pressure. The hemodynamic sensor 16 shown in FIG. 2 is an example of a minimally invasive hemodynamic sensor that may be attached to a patient via, for example, a radial artery catheter inserted in the patient's arm. In another example, the hemodynamic sensor 16 may be attached to a patient via a femoral artery catheter inserted in the patient's leg.

[0014] As shown in FIG. 2 , hemodynamic sensor 16 includes a housing 18, a fluid input port 20, a catheter-side fluid port 22, and an I / O cable 24. Fluid input port 20 is configured to be connected to a fluid source, such as a saline bag or other fluid input source, via tubing or other hydraulic connection. Catheter-side fluid port 22 is configured to be connected to a catheter (e.g., a radial artery catheter or a femoral artery catheter) inserted into a patient's arm (i.e., a radial artery catheter) or the patient's leg (i.e., a femoral artery catheter) via tubing or other hydraulic connection. I / O cable 24 is configured to connect to hemodynamic monitor 10, for example, via one or more of I / O connectors 14 ( FIG. 1 ). Housing 18 of hemodynamic sensor 16 houses one or more pressure transducers, communication circuitry, processing circuitry, and corresponding electrical components for sensing fluid pressure corresponding to the patient's arterial pressure, which is transmitted to hemodynamic monitor 10 ( FIG. 1 ) via I / O cable 24.

[0015] In operation, a column of fluid (e.g., saline) is introduced from a fluid source (e.g., a saline bag) through hemodynamic sensor 16 via fluid input port 20 toward a catheter inserted into a patient and into catheter-side fluid port 22. Arterial pressure is transmitted through the fluid column to a pressure sensor located within housing 16, which senses the pressure of the fluid column. Hemodynamic sensor 16 converts the sensed pressure of the fluid column into an electrical signal via a pressure transducer and outputs a corresponding electrical signal to hemodynamic monitor 10 (FIG. 1) via I / O cable 24. Hemodynamic sensor 16 thus transmits analog sensor data (or a digital representation of the analog sensor data) to hemodynamic monitor 10 (FIG. 1) that represents substantially continuous beat-to-beat monitoring of the patient's arterial pressure.

[0016] FIG. 3 is a perspective view of a hemodynamic sensor 26 for sensing hemodynamic data representative of a patient's arterial pressure. The hemodynamic sensor 26 shown in FIG. 3 is an example of a non-invasive hemodynamic sensor that may be attached to a patient via one or more finger cuffs to sense data representative of the patient's arterial pressure. As shown in FIG. 3, the hemodynamic sensor 26 includes an inflatable finger cuff 28 and a cardiac reference sensor 30. The inflatable finger cuff 28 also includes an optical (e.g., infrared) transmitter and optical receiver electrically connected to a pressure controller (not shown). The optical transmitter and optical receiver can measure the changing volume of the artery below the finger cuff. The optical transmitter and optical receiver may be positioned to transmit and receive light between them through an inflatable blood pressure bladder.

[0017] In operation, the pressure controller continuously adjusts the pressure within the finger cuff to maintain a constant arterial volume of the finger (i.e., the arterial unloaded volume) as measured via the optical transmitter and receiver of the inflatable finger cuff 28. The pressure applied by the pressure controller to continuously maintain the unloaded volume represents the finger's blood pressure and is communicated by the pressure controller to the hemodynamic monitor 10 shown in FIG. 1. The cardiac reference sensor 30 measures the hydrostatic height difference between the height at which the finger is held and the reference height for pressure measurement, which is typically the height of the heart. Thus, the hemodynamic sensor 26 transmits sensor data representing a substantially continuous beat-to-beat monitoring of the patient's arterial pressure waveform.

[0018] 4 is a block diagram of a hemodynamic monitoring system 32 that determines an ejection fraction measurement for a patient 36 based on a set of ejection fraction profiling parameters (also called input features) derived from the arterial pressure of the patient 36. The hemodynamic monitoring system 32 monitors the arterial pressure of the patient 36 and provides the ejection fraction measurement to a medical professional 38. If the patient 36 has a low or borderline ejection fraction measurement, the medical professional 38 can respond to the ejection fraction measurement by recommending treatment for heart failure or cardiomyopathy for the patient 36.

[0019] As shown in FIG. 4 , hemodynamic monitoring system 32 includes hemodynamic monitor 10 and hemodynamic sensor 34. Hemodynamic monitoring system 32 can be implemented in a primary care physician's office during a routine physical or checkup, while in a patient care environment such as an ICU, OR, or any other patient care environment. Hemodynamic monitoring system 32 may even be used and operated at home by patient 36. As shown in FIG. 4 , the patient care environment can include patient 36 and healthcare worker 38 who are trained to utilize hemodynamic monitoring system 32.

[0020] Hemodynamic monitor 10, as described above with respect to FIG. 1, may be an integrated hardware unit including, for example, system processor 40, system memory 42, display 12, analog-to-digital converter (ADC) 44, and digital-to-analog converter (DAC) 46. In other examples, any one or more components and / or described functionality of hemodynamic monitor 10 may be distributed across multiple hardware units. For example, in some examples, display 12 may be a separate display device separate from and operably coupled to hemodynamic monitor 10. While generally shown and described as an integrated hardware unit in the example of FIG. 4, it should be understood that hemodynamic monitor 10 may include any combination of devices and components electrically, communicatively, or otherwise operably connected to achieve the functionality attributed to hemodynamic monitor 10 herein.

[0021] As shown in FIG. 4 , the system memory 42 stores ejection fraction software code 48. The ejection fraction software code 48 includes a first module 50 for extracting and calculating waveform features from the arterial blood pressure of the patient 36, a second module 51 for extracting input features from the waveform features, and a third module 52 for determining an ejection fraction measurement for the patient 36 based on the input features. The display 12 provides a user interface 54, which includes control elements 56 that enable user interaction with the hemodynamic monitor 10 and / or other components of the hemodynamic monitoring system 32. As shown in FIG. 4 , the user interface 54 also provides a sensory alert 58 to alert a medical professional if the patient 36 has a low or borderline ejection fraction. The sensory alert 58 may be implemented as one or more of a visual alert, an audible alert, a tactile alert, or other type of sensory alert. For example, the sensory alert 58 may be invoked as any combination of flashing and / or colored graphics shown by the user interface 54 on the display 12, displaying an ejection fraction measurement along with a heart failure risk score via the user interface 54 on the display 12, an audible warning such as a siren or repeating tone, and a tactile alert configured to vibrate the hemodynamic monitor 10 or cause the hemodynamic monitor 10 to convey a physical stimulus that is perceptible to the medical professional 38 or other user.

[0022] The hemodynamic sensor 34 may be attached to the patient 36 to sense hemodynamic data representative of the patient's 36 arterial pressure waveform. The hemodynamic sensor 34 is operably connected (e.g., electrically and / or communicatively connected via a wired or wireless connection or both) to the hemodynamic monitor 10 to provide the sensed hemodynamic data to the hemodynamic monitor 10. In some examples, the hemodynamic sensor 34 provides the hemodynamic data representative of the patient's 36 arterial pressure waveform to the hemodynamic monitor 10 as an analog signal, which is converted by the ADC 44 into digital hemodynamic data representative of the arterial pressure waveform. In other examples, the hemodynamic sensor 34 may provide the sensed hemodynamic data to the hemodynamic monitor 10 in digital form, in which case the hemodynamic monitor 10 may not include or utilize an ADC 44. In yet another example, hemodynamic sensor 34 may provide hemodynamic data representing the arterial pressure waveform of patient 36 to hemodynamic monitor 10 as an analog signal, which is analyzed in analog form by hemodynamic monitor 10.

[0023] Hemodynamic sensor 34 may be a noninvasive or minimally invasive sensor attached to patient 36. For example, hemodynamic sensor 34 may take the form of minimally noninvasive hemodynamic sensor 16 ( FIG. 2 ), noninvasive hemodynamic sensor 26 ( FIG. 3 ), or other minimally invasive or noninvasive hemodynamic sensor. In some examples, hemodynamic sensor 34 may be noninvasively attached at an extremity of patient 36, such as at the wrist, arm, finger, ankle, toe, or other extremity of patient 36. Hemodynamic sensor 34 may thus take the form of a small, lightweight, and comfortable hemodynamic sensor suitable for extended wear by patient 36 to provide substantially continuous beat-to-beat monitoring of the arterial blood pressure of patient 36 over extended periods of time, such as several minutes or even hours. Although the hemodynamic sensor 34 can observe the arterial pressure of the patient 36 over an extended period of time, the hemodynamic sensor 34 only needs to observe the arterial pressure of the patient 36 for a few minutes (e.g., 5 minutes) to provide the hemodynamic monitor 10 with enough data to determine an ejection fraction measurement for the patient 36.

[0024] In some examples, the hemodynamic sensor 34 may be configured to sense the arterial pressure of the patient 36 in a minimally invasive manner. For example, the hemodynamic sensor 34 may be attached to the patient 36 via a radial artery catheter inserted in the patient's 36 arm. In other examples, the hemodynamic sensor 34 may be attached to the patient 36 via a femoral artery catheter inserted in the patient's 36 leg. Such minimally invasive techniques may also enable the hemodynamic sensor 34 to provide substantially continuous beat-to-beat monitoring of the patient's 36 arterial pressure over an extended period of time, such as several minutes or hours. While the hemodynamic sensor 34 can observe the patient's 36 arterial pressure over an extended period of time, the hemodynamic sensor 34 need only observe the patient's 36 arterial pressure for a few minutes (e.g., 5 minutes) to provide the hemodynamic monitor 10 with sufficient data to determine an ejection fraction measurement for the patient 36.

[0025] The system processor 40 is a hardware processor configured to execute the ejection fraction software code 48, which implements a first module 50, a second module 51, and a third module 52 to generate an ejection fraction measurement for the patient 36. Examples of the system processor 40 may include any one or more of a microprocessor, a controller, a digital signal processor (DSP), an application specific integrated circuit (ASIC), a field programmable gate array (FPGA), or other equivalent discrete or integrated logic circuitry.

[0026] The system memory 42 may be configured to store information within the hemodynamic monitor 10 during operation. In some examples, the system memory 42 is described as a computer-readable storage medium. In some examples, the computer-readable storage medium may include non-transitory media. The term "non-transitory" may indicate that the storage medium is not embodied in a carrier wave or propagated signal. In some examples, the non-transitory storage medium may store data that may change over time (e.g., in RAM or cache). The system memory 42 may include volatile and non-volatile computer-readable memory. Examples of volatile memory may include random access memory (RAM), dynamic random access memory (DRAM), static random access memory (SRAM), and other forms of volatile memory. Examples of non-volatile memory may include, for example, a magnetic hard disk, an optical disk, flash memory, or forms of electrically programmable memory (EPROM) or electrically erasable programmable memory (EEPROM).

[0027] Display 12 may be a liquid crystal display (LCD), a light-emitting diode (LED) display, an organic light-emitting diode (OLED) display, or other display device suitable for presenting information graphically to a user. User interface 54 may include graphical and / or physical control elements that enable user input to interact with hemodynamic monitor 10 and / or other components of hemodynamic monitoring system 32. In some examples, user interface 54 may take the form of a graphical user interface (GUI) that presents graphical control elements presented on a touch-sensitive and / or presence-sensitive display screen of display 12, for example. In such examples, user input may be received in the form of gesture input, such as touch gestures, scroll gestures, zoom gestures, or other gesture input. In some examples, user interface 54 may take the form of and / or include physical control elements, such as physical buttons, keys, knobs, or other physical control elements configured to receive user input to interact with components of hemodynamic monitoring system 32.

[0028] In operation, hemodynamic sensor 34 senses hemodynamic data representing the arterial pressure waveform of patient 36. Hemodynamic sensor 34 provides the hemodynamic data (e.g., as analog sensor data) to hemodynamic monitor 10. ADC 44 converts the analog hemodynamic data into digital hemodynamic data representing the patient's arterial pressure waveform.

[0029] The system processor 40 executes the ejection fraction software code 48 to determine an ejection fraction measurement for the patient 36 using the received hemodynamic data. For example, the system processor 40 can execute a first module 50 to perform waveform analysis of the hemodynamic data to determine a plurality of signal measurements. The plurality of signal measurements includes waveform features characterizing individual cardiac cycles of the patient's arterial pressure waveform and hemodynamic effects. The plurality of signal measurements are discussed in more detail below in the discussion of FIG. 8 . The system processor 40 executes a second module 51 to extract input features from the plurality of signal measurements that determine the ejection fraction measurement for the patient 36. The system processor 40 executes a third module 52 to determine the ejection fraction measurement for the patient 36 based on the input features. The third module 52 can also convert the ejection fraction measurement for the patient 36 into a heart failure score that represents the probability of heart failure for the patient 36.

[0030] If the patient's 36 ejection fraction measurement is 40 percent or less, the system processor 40 invokes the sensory alarm 58 of the user interface 54 to send a first sensory signal to alert the healthcare professional 38 that the patient's 36 ejection fraction measurement is low and the patient's 36 is at high risk for heart failure. The healthcare professional 38 can respond to the patient's 36 low ejection fraction measurement by recommending the patient 36 undergo further testing and investigations to verify the patient's 36 cardiac health. In this manner, the hemodynamic monitor 10 functions as a screening tool that can be used in a primary care physician's office to detect and identify heart failure in the patient 36 during a routine physical exam. Similarly, the hemodynamic monitor 10 can be used at home by the patient 36 for self-screening to determine whether the patient 36 should see a doctor or specialist.

[0031] If the system processor 40 and the third module 52 determine that the patient's 36 ejection fraction measurement is within the range of 41 percent to 49 percent, the system processor 40 invokes the sensory alert 58 of the user interface 54 to send a second sensory signal to alert the medical professional 38 that the patient 36 has a borderline ejection fraction measurement and may be at risk for heart failure in the near future. The medical professional 38 can respond to the patient's 36 borderline ejection fraction measurement by recommending that the patient 36 undergo further testing and investigations to verify the patient's 36 cardiac health. In this manner, the hemodynamic monitor 10 functions as a screening tool that can be used in a primary care physician's office to detect the possibility of future or early-onset heart failure in the patient 36 during a routine physical examination. Similarly, the hemodynamic monitor 10 can be used by the patient 36 at home for self-screening to determine whether the patient 36 needs to see a doctor or specialist.

[0032] If the system processor 40 and the third module 52 determine that the patient's 36 ejection fraction measurement exceeds 50 percent, the system processor 40 invokes a sensory alert 58 on the user interface 54 to send a third sensory signal to notify the medical professional 38 that the patient 36 has a normal ejection fraction measurement and that the patient's 36 risk of heart failure in the near future is low. When the hemodynamic monitor 10 determines that the patient 36 has a normal ejection fraction measurement, further screening or investigation of the patient 36 for ejection fraction is unlikely. Because a healthy heart does not pump more than one-half to two-thirds of the blood volume in the ventricle with each stroke, the patient's 36 ejection fraction measurement should not exceed 70%.

[0033] In some embodiments, the system processor 40 can determine multiple subsets of input features, each subset of input features relating to a different level or range of ejection fraction measurements. For example, the system processor 40 can execute a first module 50 to perform waveform analysis of the hemodynamic data to determine a plurality of signal measurements. The system processor 40 can execute a second module 51 to extract a first subset, a second subset, and a third subset of input features from the plurality of signal measurements of the patient 36. The first subset of input features are input features used by a third module 52 to determine whether the patient 36 has a normal ejection fraction measurement. The second subset of input features are input features used by a third module 52 to determine whether the patient 36 has a low ejection fraction measurement. The third subset of input features are input features used by a third module 52 to determine whether the patient 36 has a borderline ejection fraction measurement. System processor 40 may execute first module 50 to extract a single batch of signal measurements for a given unit of time, which may be used by second module 51 to extract all of a first subset, a second subset, and a third subset of input features for that unit of time. Second module 51 may extract all of the first subset, the second subset, and the third subset of input features simultaneously from the multiple signal measurements. System processor 40 may execute third module 52 to simultaneously calculate probabilities of normal, low, and borderline ejection fraction scores for patient 36.

[0034] The ejection fraction software code 48 of the hemodynamic monitor 10, in some examples, may utilize a multi-classification machine learning model with three labels: normal ejection fraction, low ejection fraction, and borderline ejection fraction. For example, the processor 40 may output the patient's 36 normal ejection fraction score along with both the patient's 36 low ejection fraction score and borderline ejection fraction score to the display 12 (and / or the display of the patient's 36 mobile device) so that all subsets of probabilities are compared together: the probability that the patient's 36 ejection fraction measurement is normal, the probability that the patient's 36 ejection fraction measurement is low, and the probability that the patient's 36 ejection fraction measurement is borderline. With the patient's 36 normal ejection fraction score, low ejection fraction score, and borderline ejection fraction score displayed together on the display 12 of the hemodynamic monitor 10, the medical professional 38 may better understand and confirm whether the patient's 36 ejection fraction measurement is normal, low, or borderline. As discussed below with respect to FIG. 5, the hemodynamic monitor 10 is a fast and efficient tool for screening and triaging patients 36 before referring them for more time-consuming and costly testing.

[0035] FIG. 5 is a perspective view of a hemodynamic monitoring system 32 and a schematic diagram of a method for triaging a patient 36 based on the patient's ejection fraction measurement. As shown in FIG. 5, the hemodynamic monitoring system 32 includes a hemodynamic monitor 10 and a hemodynamic sensor 34. In the embodiment of FIG. 5, the hemodynamic sensor 34 is a non-invasive hemodynamic sensor 26 (described in detail above with reference to FIG. 3) that can be attached to the patient 36 via one or more finger cuffs to sense data representative of the patient's 36's arterial blood pressure. In the embodiment of FIG. 5, the hemodynamic monitor 10 is a small, wearable unit that can be strapped to the patient's 36's arm and connected to the hemodynamic sensor 34 to receive sensed data representative of the patient's 36's arterial blood pressure. The embodiment of the hemodynamic monitoring system 32 of FIG. 5 can operate and function as described above with reference to FIG. 4 to determine the patient's 36's ejection fraction measurement. During a routine medical checkup in a primary care physician's office, hemodynamic monitoring system 32 may be connected to the hand and arm of patient 36 and sensed hemodynamic data of patient 36 may be provided to hemodynamic monitor 10, thereby quickly screening patient 36 for risk of heart failure and triaging patient 36. A few minutes (e.g., 5 minutes) after providing the sensed hemodynamic data of patient 36 to hemodynamic monitor 10, hemodynamic monitor 10 outputs an ejection fraction measurement of patient 36 to display 12. In some embodiments, hemodynamic monitor 10 may also output the ejection fraction measurement of patient 36 to a display on the patient's 36's mobile device.

[0036] The hemodynamic monitor 10 may color-code and / or mark the patient's 36 ejection fraction measurement on the display 12 depending on whether the ejection fraction measurement is normal, borderline, or low. As discussed above with respect to FIG. 4 , a normal ejection fraction measurement is 50 percent or greater, a borderline ejection fraction measurement is within the range of 41 percent to 49 percent, and a low ejection fraction measurement is 40 percent or less. If the hemodynamic monitor 10 determines that the patient's 36 ejection fraction measurement is normal, the hemodynamic monitor 10 may output a positive numeric green score on the display 12. If the hemodynamic monitor 10 determines that the patient's 36 ejection fraction measurement is borderline, the hemodynamic monitor may output a neutral or zero numeric yellow score on the display 12. If the hemodynamic monitor 10 determines that the patient's 36 ejection fraction measurement is low, the hemodynamic monitor 10 may output a negative numeric red score on the display 12.

[0037] The medical professional 38 (shown in FIG. 4 ) in this scenario may be a primary care physician or nurse conducting a routine physical examination of the patient 36. After the hemodynamic monitor 10 observes and processes the patient's 36 sensed hemodynamic data for several minutes, the medical professional 38 outputs the patient's 36 ejection fraction measurement on the display 12. The medical professional 38 can triage the patient 36 based on the patient's 36 ejection fraction measurement. If the patient's 36 ejection fraction measurement is shown on the display 12 as being low, the medical professional 38 can inform the patient 36 that the patient 36 may have heart failure and should immediately seek further testing, investigation, and treatment from a cardiovascular specialist. The medical professional 38 can then refer the patient 36 to the cardiovascular specialist for more detailed investigation, such as an echocardiogram, a multi-gated (MUGA) scan, a computed tomography (CT) scan, a cardiac catheterization, and / or a nuclear stress test. If the patient's 36 ejection fraction measurement on display 12 is borderline, medical professional 38 can inform patient 36 that the patient 36 may be at risk for heart failure and that the patient 36 should seek further testing, investigation, and treatment from a cardiovascular specialist within a reasonable timeframe. If the patient's 36 ejection fraction measurement on display 12 is normal, medical professional 38 can inform patient 36 that the patient's 36 risk for heart failure is low and no additional testing is necessary. As discussed below with respect to FIGS. 6-8 , the machine learning model of hemodynamic monitor 10 can be trained using a clinical dataset to recognize input features in the patient's 36 arterial pressure waveform and use those input features to determine the patient's 36 ejection fraction measurement.

[0038] 6 is a diagram of clinical data 60 used for data mining and machine training of hemodynamic monitor 10 of hemodynamic monitoring system 32. Clinical data 60 includes a first clinical data set 61, a second clinical data set 62, and a third clinical data set 63.

[0039] First clinical dataset 61 includes a collection of arterial pressure waveforms recorded from a first group of individuals who are each confirmed to have a normal ejection fraction measurement of greater than 50 percent. First clinical dataset 61 may be collected from the first group of individuals by an invasive hemodynamic sensor, such as hemodynamic sensor 16 shown in FIG. 2, or by a non-invasive hemodynamic sensor, such as hemodynamic sensor 26 shown in FIG. 3. As each individual in first clinical dataset 61 is connected to a hemodynamic sensor, the hemodynamic sensor records that individual's arterial pressure waveform, and that arterial pressure waveform is tagged with a first label so that the individual's arterial pressure waveform can ultimately be added to first clinical dataset 61. Adding the first label to the arterial pressure waveforms of the individuals in the first group also allows the first clinical dataset 61 to be collected and stored in a common location with the second clinical dataset 62 and the third clinical dataset 63 without the arterial pressure waveforms of the first clinical dataset 61 being lost or confused with the arterial pressure waveforms of the second clinical dataset 62 and the third clinical dataset 63.

[0040] After the arterial pressure waveforms of the first clinical dataset 61 are collected and labeled with a first label, the arterial pressure waveforms of the first clinical dataset 61 are ready for use in data mining and machine training of the hemodynamic monitor 10. The arterial pressure waveforms of the first clinical dataset 61 are data mined and used to machine train the hemodynamic monitor 10 to determine a first subset of input features. As discussed above with respect to FIG. 4, the first subset of input features are input features used by the third module 52 to determine whether the patient 36 has a normal ejection fraction measurement. As discussed further below with respect to FIGS. 7-8, waveform analysis is performed on the first clinical dataset 61 to calculate a plurality of signal measurements, which are then used to calculate a first subset of input features that best detects and measures a normal ejection fraction from the arterial pressure waveform.

[0041] The second clinical dataset 62 includes a collection of arterial pressure waveforms recorded from a second group of individuals each identified as having a low ejection fraction measurement of 40 percent or less. The second clinical dataset 62 may be collected from the second group of individuals by an invasive hemodynamic sensor, such as the hemodynamic sensor 16 shown in FIG. 2, or by a non-invasive hemodynamic sensor, such as the hemodynamic sensor 26 shown in FIG. 3. As each individual in the second clinical dataset 62 is connected to a hemodynamic sensor, the hemodynamic sensor records that individual's arterial pressure waveform, which is tagged with a second label so that the individual's arterial pressure waveform can ultimately be added to the second clinical dataset 62. Adding a second label to the arterial pressure waveforms of individuals in the second group also allows the second clinical dataset 62 to be collected and stored in a common location with the first clinical dataset 61 and the third clinical dataset 63 without the arterial pressure waveforms of the second clinical dataset 62 being lost or confused with the arterial pressure waveforms of the first clinical dataset 61 and the third clinical dataset 63.

[0042] After the arterial pressure waveforms of the second clinical dataset 62 are collected and labeled with the second label, the arterial pressure waveforms of the second clinical dataset 62 are ready for use in data mining and machine training of the hemodynamic monitor 10. The arterial pressure waveforms of the second clinical dataset 62 are data mined and used to machine train the hemodynamic monitor 10 to determine a second subset of input features. As discussed above with respect to FIG. 4, the second subset of input features are input features used by the third module 52 to determine whether the patient 36 has a low ejection fraction measurement. As discussed further below with respect to FIGS. 7-8, waveform analysis is performed on the second clinical dataset 62 to calculate a plurality of signal measurements, which are then used to calculate the second subset of input features that best detect and measure a low ejection fraction from the arterial pressure waveform.

[0043] The third clinical dataset 63 includes a collection of arterial pressure waveforms recorded from a third group of individuals who are each identified as having borderline ejection fraction measurements within the range of 41 percent to 49 percent. The third clinical dataset 63 may be collected from the third group of individuals by an invasive hemodynamic sensor, such as the hemodynamic sensor 16 shown in FIG. 2, or by a non-invasive hemodynamic sensor, such as the hemodynamic sensor 26 shown in FIG. 3. As each individual in the third clinical dataset 63 is connected to a hemodynamic sensor, the hemodynamic sensor records that individual's arterial pressure waveform, which is tagged with a third label so that the individual's arterial pressure waveform can ultimately be added to the third clinical dataset 63. Adding a third label to the arterial pressure waveforms of individuals in the third group also allows the third clinical dataset 63 to be collected and stored in a common location with the first clinical dataset 61 and the second clinical dataset 62 without the arterial pressure waveforms of the third clinical dataset 63 being lost or confused with the arterial pressure waveforms of the first clinical dataset 61 and the second clinical dataset 62.

[0044] After the arterial pressure waveforms of the third clinical dataset 63 have been collected and labeled with the third label, the arterial pressure waveforms of the third clinical dataset 63 are ready for use in data mining and machine training of the hemodynamic monitor 10. The arterial pressure waveforms of the third clinical dataset 63 are data mined and used to machine train the hemodynamic monitor 10 to determine a third subset of input features. As discussed above with respect to FIG. 4, the third subset of input features are input features used by the third module 52 to determine whether the patient 36 has a borderline ejection fraction measurement. As discussed further below with respect to FIGS. 7-8, waveform analysis is performed on the third clinical dataset 63 to calculate a plurality of signal measurements, which are then used to calculate a third subset of input features that best detects and measures the borderline ejection fraction from the arterial pressure waveform.

[0045] FIG. 7 is a flow diagram of a method 70 for data mining the clinical data 60 from FIG. 6 to machine train a machine learning model of the hemodynamic monitor 10. The method 70 of FIG. 7 is also discussed with reference to FIG. 8. The method 70 is applied to each of the first clinical data set 61, the second clinical data set 62, and the third clinical data set 63 to train the hemodynamic monitor to find the input features (including the first, second, and third subsets of input features) previously described with respect to FIGS. 4 and 6. The method 70 is described as being applied to the arterial pressure waveform of the first clinical data set 61.

[0046] To machine train the hemodynamic monitor 10 to identify the first subset of input features illustrated in FIG. 4 , the first subset of input features is first determined by applying method 70 to arterial pressure waveforms of a first clinical data set 61 of the clinical data 60. A first step 72 of method 70 performs waveform analysis of the arterial pressure waveforms collected in the first clinical data set 61 to calculate a plurality of signal measurements for the first clinical data set 61. Performing waveform analysis of the arterial pressure waveforms of the first clinical data set 61 may include identifying individual cardiac cycles in each of the arterial pressure waveforms of the first clinical data set 61. FIG. 8 provides an exemplary graph showing an exemplary trajectory of the arterial pressure waveform with the individual cardiac cycles identified and zoomed in. Next, performing waveform analysis of the arterial pressure waveforms of the first clinical data set 61 may include identifying a dicrotic notch in each of the individual cardiac cycles of each of the arterial pressure waveforms of the first clinical data set 61, similar to the example shown in FIG. 8 . Next, waveform analysis of the arterial pressure waveforms of the first clinical data set 61 includes identifying the systolic rise phase, the systolic decay phase, and the diastolic phase in each individual cardiac cycle of each of the arterial pressure waveforms of the first clinical data set 61, similar to the example shown in FIG.

[0047] Signal measurements are extracted from each of the systolic rise phase, systolic decay phase, and diastolic phase from each individual cardiac cycle of each arterial pressure waveform of the first clinical data set 61. The signal measurements may correspond to hemodynamic effects from each of the systolic rise phase, systolic decay phase, and diastolic phase from each individual cardiac cycle. These hemodynamic effects may include contractility, aortic elasticity, stroke volume, vascular tone, afterload, and the entire cardiac cycle. The signal measurements calculated or extracted by the waveform analysis of the first step 72 of method 70 include mean, maximum, minimum, duration, area, standard deviation, derivative, and / or morphological measurements from each of the systolic rise phase, systolic decay phase, and diastolic phase from each individual cardiac cycle. The signal measurements may also include heart rate, respiratory rate, stroke volume, pulse pressure, pulse pressure variability, stroke volume variability, mean arterial pressure (MAP), systolic pressure (SYS), diastolic pressure (DIA), heart rate variability, cardiac output, peripheral resistance, vascular elasticity, and / or left ventricular contractility extracted from each individual cardiac cycle of each arterial pressure waveform of the first clinical dataset 61.

[0048] After the signal measurements have been determined for the first clinical dataset 61, step 74 of method 70 is performed on the signal measurements of the first clinical dataset 61. Step 74 of method 70 calculates composite measurements between the signal measurements of the first clinical dataset 61. Calculating composite measurements between the signal measurements of the first clinical dataset 61 may include performing steps 76, 78, 80, and 82 shown in FIG. 7 for all signal measurements of the first clinical dataset 61. Step 76 is performed by arbitrarily selecting a subset of signal measurements (e.g., a subset of signal measurements) from the signal measurements of the first clinical dataset 61. Next, as shown in step 78 of FIG. 7, powers of different orders are calculated for each of the subsets of signal measurements among the subsets of signal measurements to generate powers of the subsets of signal measurements. In step 80 of FIG. 7, the powers of the subsets of signal measurements are then multiplied together to generate a product of the powers of the subsets of signal measurements. Step 82 involves performing a receiver operating characteristic (ROC) analysis of the product to obtain a composite measure for the subset of signal measures. Steps 76, 78, 80, and 82 are repeated until all of the composite measures have been calculated between all of the signal measures of the first clinical data set 61. The final step 84 involves selecting the signal measures having the most predictive top composite measures (i.e., composite measures that meet the threshold predictive criteria) as the top signal measures for the first clinical data set 61, labeled as the first subset of input features. Once the first subset of input features has been determined, the hemodynamic monitor 10 is trained or programmed to perform waveform analysis on the arterial pressure waveform of the patient 36 (shown in FIG. 4), extract the first subset of input features from the arterial pressure waveform of the patient 36, and use the first subset of input features to determine whether the patient 36 has a normal ejection fraction measure.

[0049] Just as method 70 was applied to the arterial pressure waveform of first clinical data set 61 to determine a first subset of input features, method 70 is applied to second clinical data set 62 to determine a second subset of input features. Method 70 is also applied to third clinical data set 63 to determine a third subset of input features.

[0050] Discussion of Possible Embodiments The following is a comprehensive description of possible embodiments of the present invention.

[0051] In one example, a method for triaging a patient for risk of heart failure includes receiving, by a hemodynamic monitor, sensed hemodynamic data representing the patient's arterial pressure waveform. The hemodynamic monitor performs waveform analysis of the sensed hemodynamic data to calculate a plurality of signal measurements of the sensed hemodynamic data. The hemodynamic monitor extracts input features indicative of the patient's ejection fraction from the plurality of signal measurements. The hemodynamic monitor determines the patient's ejection fraction based on the input features and outputs the patient's ejection fraction to a display. The hemodynamic monitor alerts the patient that the ejection fraction is low when the ejection fraction is 40 percent or less. The hemodynamic monitor alerts the patient that the ejection fraction is borderline when the ejection fraction is between 41 percent and 49 percent. The hemodynamic monitor informs the patient that the ejection fraction is normal when the ejection fraction is 50 percent or greater.

[0052] The method of the preceding paragraph may optionally include, in addition and / or in the alternative, any one or more of the following features, configurations, and / or additional components listed below.

[0053] a second clinical data set including arterial pressure waveforms from a second group of individuals with a low ejection fraction measurement of 40 percent or less; and a third clinical data set including arterial pressure waveforms from a third group of individuals with a borderline ejection fraction measurement within a range of 41 percent to 49 percent; performing waveform analysis of the arterial pressure waveforms of the first clinical data set, the second clinical data set, and the third clinical data set to calculate a plurality of waveform signal measurements; and determining input features by calculating a composite measure among the plurality of waveform signal measurements, selecting top signal measurements from the plurality of waveform signal measurements having the most predictive composite measure, and labeling the top signal measurements as input features.

[0054] In another example, a system for triaging patients for risk of heart failure includes a hemodynamic sensor that produces hemodynamic data representative of the patient's arterial pressure waveform. The system also includes a user interface with a display for displaying an ejection fraction measurement of the patient to a medical professional. Ejection fraction software code is stored in a system memory of the system. The system includes a processor configured to perform waveform analysis of the hemodynamic data to determine a plurality of signal measurements, extract input features indicative of the patient's ejection fraction measurement from the plurality of signal measurements, determine the patient's ejection fraction measurement based on the input features, and execute the ejection fraction software code to output the ejection fraction measurement to a display of the user interface.

[0055] The system of the preceding paragraph may optionally include, in addition and / or in the alternative, any one or more of the following features, configurations, and / or additional components listed below.

[0056] a third clinical data set including arterial pressure waveforms from a third group of individuals with borderline ejection fraction measurements within a range of 41 to 49 percent; performing waveform analysis of the arterial pressure waveforms of the first clinical data set, the second clinical data set, and the third clinical data set to calculate a plurality of waveform signal measurements; and determining the input features by calculating a composite measure among the plurality of waveform signal measurements, selecting a top signal measure from the plurality of waveform signal measurements having a most predictive composite measure, and labeling the top signal measure as the input feature.

[0057] A further embodiment of the above system, wherein performing waveform analysis of the labeled arterial pressure waveforms of the first clinical data set, the second clinical data set, and the third clinical data set to calculate a plurality of waveform signal measurements comprises identifying individual cardiac cycles in each of the arterial pressure waveforms of the first clinical data set, the second clinical data set, and the third clinical data set; identifying a dicrotic notch in each of the individual cardiac cycles; identifying a systolic rise phase, a systolic decay phase, and a diastolic phase in each of the individual cardiac cycles; and extracting a plurality of waveform signal measurements from each of the systolic rise phase, the systolic decay phase, and the diastolic phase from each of the individual cardiac cycles.

[0058] A further embodiment of the above system, wherein the plurality of waveform signal measurements correspond to hemodynamic effects from each of a systolic upstroke phase, a systolic decay phase, and a diastolic phase from each individual cardiac cycle, the hemodynamic effects comprising contractility, aortic elasticity, stroke volume, vascular tone, afterload, and the entire cardiac cycle.

[0059] A further embodiment of the above system, wherein the plurality of waveform signal measurements comprises a mean, maximum, minimum, duration, area, standard deviation, derivative, and / or morphological measurements from each of the systolic upstroke phase, the systolic decay phase, and the diastolic phase from each individual cardiac cycle.

[0060] A further embodiment of the above system, wherein the plurality of waveform signal measurements comprise heart rate, respiratory rate, stroke volume, pulse pressure, pulse pressure variability, stroke volume variability, mean arterial pressure (MAP), systolic pressure (SYS), diastolic pressure (DIA), heart rate variability, cardiac output, peripheral resistance, vascular elasticity, and / or left ventricular contractility extracted from each individual cardiac cycle.

[0061] a first clinical data set, a second clinical data set, and a third clinical data set; and a second clinical data set, each of which is a waveform signal measurement subset; a third clinical data set, each of which is a waveform signal measurement subset; a second clinical data set, each of which is a waveform signal measurement subset; a third clinical data set, each of which is a waveform signal measurement subset; a fourth clinical data set, each of which is a waveform signal measurement subset; a fourth clinical data set, each of which is a waveform signal measurement subset; a fourth clinical data set, each of which is a waveform signal measurement subset;

[0062] A further embodiment of the above system, wherein the hemodynamic sensor is a non-invasive hemodynamic sensor attachable to an extremity of the patient.

[0063] A further embodiment of the above system, wherein the hemodynamic sensor is a minimally non-invasive aortic catheter-based hemodynamic sensor.

[0064] A further embodiment of the above system, wherein the hemodynamic sensor produces the hemodynamic data as an analog hemodynamic sensor signal representative of the patient's arterial pressure waveform.

[0065] A further embodiment of the above system, further comprising an analog-to-digital converter that converts the analog hemodynamic sensor signal into digital hemodynamic data representative of the patient's arterial pressure waveform.

[0066] In another example, a method for triaging a patient for risk of heart failure is disclosed. The method includes receiving, by a hemodynamic monitor, sensed hemodynamic data representing the patient's arterial pressure waveform. The hemodynamic monitor performs waveform analysis of the sensed hemodynamic data to calculate a plurality of signal measurements of the sensed hemodynamic data. The hemodynamic monitor extracts input features indicative of the patient's ejection fraction from the plurality of signal measurements. The hemodynamic monitor determines the patient's ejection fraction based on the input features and outputs the patient's ejection fraction to a display and / or mobile device. The hemodynamic monitor alerts the patient or a healthcare professional that the ejection fraction is low when the ejection fraction is 40 percent or less.

[0067] The method of the preceding paragraph may optionally include, in addition and / or in the alternative, any one or more of the following features, configurations, and / or additional components listed below.

[0068] A further embodiment of the above method, further comprising the step of alerting the patient or healthcare professional that the ejection fraction is borderline when the ejection fraction is within the range of 41 percent to 49 percent.

[0069] A further embodiment of the above method, further comprising the step of informing the patient or healthcare professional that the ejection fraction is normal when the ejection fraction is 50 percent or greater.

[0070] a first subset of input features by: collecting first clinical data including arterial pressure waveforms from a first group of individuals with normal ejection fraction measurements of 50 percent or greater; labeling each of the arterial pressure waveforms of the first clinical data set with a first label; performing waveform analysis of the labeled arterial pressure waveforms of the first clinical data set to calculate a plurality of waveform signal measurements of the first clinical data set; and determining a first subset of input features by calculating a composite measure among the plurality of waveform signal measurements of the first clinical data set, selecting top signal measures from the plurality of waveform signal measures of the first clinical data set that have the most predictive composite measure, and labeling the top signal measures of the first clinical data set as the first subset of input features.

[0071] a second clinical dataset including arterial pressure waveforms from a second group of individuals with low ejection fraction measurements of 40 percent or less; labeling each of the arterial pressure waveforms of the second clinical dataset with a second label; performing waveform analysis of the labeled arterial pressure waveforms of the second clinical dataset to calculate a plurality of waveform signal measurements of the second clinical dataset; and determining a second subset of input features by calculating a composite measure among the plurality of waveform signal measurements of the second clinical dataset, selecting top signal measurements from the plurality of waveform signal measurements of the second clinical dataset having the most predictive composite measure, and labeling the top signal measurements of the second clinical dataset as the second subset of input features.

[0072] a third clinical dataset including arterial pressure waveforms from a third group of individuals with borderline ejection fraction measurements within a range of 41 percent to 49 percent; labeling each of the arterial pressure waveforms of the third clinical dataset with a third label; performing waveform analysis of the labeled arterial pressure waveforms of the third clinical dataset to calculate a plurality of waveform signal measurements of the third clinical dataset; and determining a third subset of input features by calculating a composite measure among the plurality of waveform signal measurements of the third clinical dataset, selecting top signal measures from the plurality of waveform signal measures of the third clinical dataset having the most predictive composite measure, and labeling the top signal measures of the third clinical dataset as the third subset of input features.

[0073] A further embodiment of the above-described method, wherein the step of performing waveform analysis of the labeled arterial pressure waveforms of the first clinical dataset to calculate a plurality of waveform signal measurements of the first clinical dataset comprises the steps of: identifying individual cardiac cycles in each of the arterial pressure waveforms of the first clinical dataset; identifying a dicrotic notch in each of the individual cardiac cycles in each of the arterial pressure waveforms of the first clinical dataset; identifying a systolic rise phase, a systolic decay phase, and a diastolic phase in each of the individual cardiac cycles in each of the arterial pressure waveforms of the first clinical dataset; and extracting a plurality of waveform signal measurements of the first clinical dataset from each of the systolic rise phase, the systolic decay phase, and the diastolic phase from each of the individual cardiac cycles in each of the arterial pressure waveforms of the first clinical dataset.

[0074] A further embodiment of the above-described method, wherein the plurality of waveform signal measurements of the first clinical dataset correspond to hemodynamic effects from each of a systolic rise phase, a systolic decay phase, and a diastolic phase from each individual cardiac cycle in each of the arterial pressure waveforms of the first clinical dataset, the hemodynamic effects comprising contractility, aortic elasticity, stroke volume, vascular tone, afterload, and the entire cardiac cycle.

[0075] A further embodiment of the above-described method, wherein the plurality of waveform signal measurements of the first clinical data set comprise mean, maximum, minimum, time length, area, standard deviation, derivative, and / or morphological measurements from each of a systolic rise phase, a systolic decay phase, and a diastolic phase from each individual cardiac cycle in each of the arterial pressure waveforms of the first clinical data set.

[0076] A further embodiment of the above-described method, wherein the plurality of waveform signal measurements of the first clinical dataset comprise heart rate, respiratory rate, stroke volume, pulse pressure, pulse pressure variability, stroke volume variability, mean arterial pressure (MAP), systolic pressure (SYS), diastolic pressure (DIA), heart rate variability, cardiac output, peripheral resistance, vascular elasticity, and / or left ventricular contractility extracted from each individual cardiac cycle in each of the arterial pressure waveforms of the first clinical dataset.

[0077] The step of calculating composite measures between the plurality of waveform signal measures of the first clinical data set includes: performing step 1 by arbitrarily selecting a subset of signal measures from the plurality of waveform signal measures of the first clinical data set; performing step 2 by calculating powers of different orders for each of the subset of signal measures from the plurality of waveform signal measures of the first clinical data set to generate powers of the subset of signal measures from the plurality of waveform signal measures of the first clinical data set; and performing step 3 by calculating powers of different orders for each of the subset of signal measures from the plurality of waveform signal measures of the first clinical data set to generate powers of the subset of signal measures from the plurality of waveform signal measures of the first clinical data set. a receiver operating characteristic (ROC) analysis of the products of the powers of the subset of signal measurements from the plurality of waveform signal measurements of the first clinical data set to obtain a composite measurement for the product of the powers of the subset of signal measurements from the plurality of waveform signal measurements of the first clinical data set; and repeating steps 1, 2, 3, and 4 until all of the composite measurements have been calculated between all of the plurality of waveform signal measurements of the first clinical data set.

[0078] A further embodiment of the above-described method, wherein the step of performing waveform analysis of the labeled arterial pressure waveforms of the second clinical dataset to calculate a plurality of waveform signal measurements of the second clinical dataset comprises the steps of: identifying individual cardiac cycles in each of the arterial pressure waveforms of the second clinical dataset; identifying a dicrotic notch in each of the individual cardiac cycles in each of the arterial pressure waveforms of the second clinical dataset; identifying a systolic rise phase, a systolic decay phase, and a diastolic phase in each of the individual cardiac cycles in each of the arterial pressure waveforms of the second clinical dataset; and extracting a plurality of waveform signal measurements of the second clinical dataset from each of the systolic rise phase, the systolic decay phase, and the diastolic phase from each of the individual cardiac cycles in each of the arterial pressure waveforms of the second clinical dataset.

[0079] A further embodiment of the above method, wherein the plurality of waveform signal measurements of the second clinical dataset correspond to hemodynamic effects from each of a systolic rise phase, a systolic decay phase, and a diastolic phase from each individual cardiac cycle in each of the arterial pressure waveforms of the second clinical dataset, the hemodynamic effects comprising contractility, aortic elasticity, stroke volume, vascular tone, afterload, and the entire cardiac cycle.

[0080] A further embodiment of the above-described method, wherein the plurality of waveform signal measurements of the second clinical dataset comprises mean, maximum, minimum, time length, area, standard deviation, derivative, and / or morphological measurements from each of a systolic rise phase, a systolic decay phase, and a diastolic phase from each individual cardiac cycle in each of the arterial pressure waveforms of the second clinical dataset.

[0081] A further embodiment of the above-described method, wherein the plurality of waveform signal measurements of the second clinical dataset comprise heart rate, respiratory rate, stroke volume, pulse pressure, pulse pressure variability, stroke volume variability, mean arterial pressure (MAP), systolic pressure (SYS), diastolic pressure (DIA), heart rate variability, cardiac output, peripheral resistance, vascular elasticity, and / or left ventricular contractility extracted from each individual cardiac cycle in each of the arterial pressure waveforms of the second clinical dataset.

[0082] The step of calculating composite measures between the plurality of waveform signal measures of the second clinical data set includes: performing step 1 by arbitrarily selecting a subset of signal measures from the plurality of waveform signal measures of the second clinical data set; performing step 2 by calculating powers of different orders for each of the subset of signal measures from the plurality of waveform signal measures of the second clinical data set to generate powers of the subset of signal measures from the plurality of waveform signal measures of the second clinical data set; and performing step 3 by calculating powers of different orders for each of the subset of signal measures from the plurality of waveform signal measures of the second clinical data set to generate powers of the subset of signal measures from the plurality of waveform signal measures of the second clinical data set. a first clinical dataset of the first clinical data set, wherein the first clinical dataset is a subset of the signal measurements from the plurality of waveform signal measurements of the second clinical data set; and a second clinical dataset of the second clinical data set is a subset of the signal measurements from the plurality of waveform signal measurements of the second clinical data set.

[0083] A further embodiment of the above-described method, wherein the step of performing waveform analysis of the labeled arterial pressure waveforms of the third clinical dataset to calculate a plurality of waveform signal measurements of the third clinical dataset comprises the steps of: identifying individual cardiac cycles in each of the arterial pressure waveforms of the third clinical dataset; identifying a dicrotic notch in each of the individual cardiac cycles in each of the arterial pressure waveforms of the third clinical dataset; identifying a systolic rise phase, a systolic decay phase, and a diastolic phase in each of the individual cardiac cycles in each of the arterial pressure waveforms of the third clinical dataset; and extracting a plurality of waveform signal measurements of the third clinical dataset from each of the systolic rise phase, the systolic decay phase, and the diastolic phase from each of the individual cardiac cycles in each of the arterial pressure waveforms of the third clinical dataset.

[0084] A further embodiment of the above method, wherein the plurality of waveform signal measurements of the third clinical dataset correspond to hemodynamic effects from each of a systolic rise phase, a systolic decay phase, and a diastolic phase from each individual cardiac cycle in each of the arterial pressure waveforms of the third clinical dataset, the hemodynamic effects comprising contractility, aortic elasticity, stroke volume, vascular tone, afterload, and the entire cardiac cycle.

[0085] A further embodiment of the above-described method, wherein the plurality of waveform signal measurements of the third clinical dataset comprises mean, maximum, minimum, time length, area, standard deviation, derivative, and / or morphological measurements from each of a systolic rise phase, a systolic decay phase, and a diastolic phase from each individual cardiac cycle in each of the arterial pressure waveforms of the third clinical dataset.

[0086] A further embodiment of the above-described method, wherein the plurality of waveform signal measurements of the third clinical dataset comprise heart rate, respiratory rate, stroke volume, pulse pressure, pulse pressure variability, stroke volume variability, mean arterial pressure (MAP), systolic pressure (SYS), diastolic pressure (DIA), heart rate variability, cardiac output, peripheral resistance, vascular elasticity, and / or left ventricular contractility extracted from each individual cardiac cycle in each of the arterial pressure waveforms of the third clinical dataset.

[0087] The step of calculating a composite measure between the plurality of waveform signal measures of the third clinical data set includes: performing step 1 by arbitrarily selecting a subset of signal measures from the plurality of waveform signal measures of the third clinical data set; performing step 2 by calculating powers of different orders for each of the signal measures from the subset of signal measures from the plurality of waveform signal measures of the third clinical data set to generate a product of the powers of the subset of signal measures from the plurality of waveform signal measures of the third clinical data set; and performing step 3 by calculating a power of different orders for each of the signal measures from the subset of signal measures from the plurality of waveform signal measures of the third clinical data set to generate a product of the powers of the subset of signal measures from the plurality of waveform signal measures of the third clinical data set. A further embodiment of the above-described method, comprising: performing step 3 by multiplying together powers of a subset of signal measurements from the plurality of waveform signal measurements of the third clinical dataset; and performing step 4 by performing a receiver operating characteristic (ROC) analysis of the products of powers of the subset of signal measurements from the plurality of waveform signal measurements of the third clinical dataset to obtain a composite measurement for the products of powers of the subset of signal measurements from the plurality of waveform signal measurements of the third clinical dataset; and repeating steps 1, 2, 3, and 4 until all of the composite measurements have been calculated between all of the plurality of waveform signal measurements of the third clinical dataset.

[0088] In another example, a method for training a hemodynamic monitor to determine a patient's ejection fraction is disclosed. The method for training the hemodynamic monitor includes collecting a first clinical dataset including arterial pressure waveforms from a first group of individuals with normal ejection fraction measurements of 50 percent or greater. A second clinical dataset including arterial pressure waveforms from a second group of individuals with low ejection fraction measurements of 40 percent or less is collected. The method further includes collecting a third clinical dataset including arterial pressure waveforms from a third group of individuals with borderline ejection fraction measurements within a range of 41 to 49 percent. Waveform analysis of the arterial pressure waveforms of the first, second, and third clinical datasets is performed to calculate a plurality of waveform signal measurements. Input features are determined by calculating a composite measure among the plurality of waveform signal measurements, selecting a top signal measure from the plurality of waveform signal measurements with the most predictive composite measure, and labeling the top signal measure as an input feature. The input features are stored in a memory of the hemodynamic monitor.

[0089] The method of the preceding paragraph may optionally include, in addition and / or in the alternative, any one or more of the following features, configurations, and / or additional components listed below.

[0090] A further embodiment of the above method, further comprising the steps of connecting a hemodynamic sensor to the hemodynamic monitor and the patient for inputting the patient's sensed arterial pressure waveform into the hemodynamic monitor; extracting, by a processor of the hemodynamic monitor, values ​​of input features of the patient's sensed arterial pressure waveform; determining, by the processor of the hemodynamic monitor, an ejection fraction of the patient based on the values ​​of the input features of the sensed arterial pressure waveform; and outputting the patient's ejection fraction to a display and / or mobile device.

[0091] A further embodiment of the above method, further comprising the step of alerting the patient and / or healthcare provider that the ejection fraction is low when the ejection fraction is less than or equal to 40 percent.

[0092] A further embodiment of the above method, further comprising the step of alerting the patient and / or healthcare provider that the ejection fraction is borderline when the ejection fraction is within the range of 41 percent to 49 percent.

[0093] A further embodiment of the above method, further comprising the step of alerting the patient and / or healthcare provider that the ejection fraction is normal when the ejection fraction is greater than or equal to 50 percent.

[0094] In another example, a hemodynamic monitor for detecting heart failure is disclosed. The hemodynamic monitor includes a noninvasive blood pressure sensor having an inflatable blood pressure bladder, a pressure controller pneumatically connected to the inflatable blood pressure bladder, and an optical transmitter and an optical receiver electrically connected to the pressure controller. The hemodynamic monitor also includes an integrated hardware unit with a system processor, a system memory, and a display with a user interface. The system memory includes instructions that, when executed by the system processor, are configured to: adjust, by the pressure controller, the pressure in the inflatable blood pressure bladder to keep the patient's arterial volume constant for a period of time based on a feedback signal generated by the optical transmitter and the optical receiver; generate arterial pressure waveform data for the patient based on the adjusted pressure in the inflatable blood pressure bladder over the period of time; extract a plurality of signal measurements from the patient's arterial pressure waveform data; extract input features indicative of the patient's ejection fraction score from the plurality of signal measurements; determine the patient's ejection fraction score based on the extracted input features; generate a first sensory alert signal configured to generate a first sensory alert indicating that the patient's ejection fraction is low when the ejection fraction score exceeds a threshold score, or generate a second sensory alert signal configured to generate a second sensory alert indicating that the patient's ejection fraction is not low when the ejection fraction score is below the threshold score; transmit the first sensory alert signal or the second sensory alert signal to a user interface; and output the first sensory alert or the second sensory alert through the user interface.

[0095] The hemodynamic monitor of the preceding paragraph may optionally include, in addition and / or in the alternative, any one or more of the following features, configurations, and / or additional components listed below.

[0096] a second clinical data set including arterial pressure waveforms from a second group of individuals with a low ejection fraction measurement of 40 percent or less; and a third clinical data set including arterial pressure waveforms from a third group of individuals with a borderline ejection fraction measurement within a range of 41 percent to 49 percent; performing waveform analysis of the arterial pressure waveforms of the first clinical data set, the second clinical data set, and the third clinical data set to calculate a plurality of waveform signal measurements; and determining the input features by calculating a composite measurement among the plurality of waveform signal measurements, selecting a top signal measurement from the plurality of waveform signal measurements having a most predictive composite measurement, and labeling the top signal measurement as the input feature.

[0097] A further embodiment of the hemodynamic monitor described above, wherein performing waveform analysis of the arterial pressure waveforms of the first clinical data set, the second clinical data set, and the third clinical data set to calculate a plurality of waveform signal measurements comprises identifying individual cardiac cycles in each of the arterial pressure waveforms of the first clinical data set, the second clinical data set, and the third clinical data set; identifying a dicrotic notch in each of the individual cardiac cycles; identifying a systolic rise phase, a systolic decay phase, and a diastolic phase in each of the individual cardiac cycles; and extracting a plurality of waveform signal measurements from each of the systolic rise phase, the systolic decay phase, and the diastolic phase from each of the individual cardiac cycles.

[0098] A further embodiment of the hemodynamic monitor described above, wherein the plurality of waveform signal measurements correspond to hemodynamic effects from each of a systolic upstroke phase, a systolic decay phase, and a diastolic phase from each individual cardiac cycle, the hemodynamic effects comprising contractility, aortic elasticity, stroke volume, vascular tone, afterload, and the entire cardiac cycle.

[0099] Further embodiments of the hemodynamic monitor described above, wherein the plurality of waveform signal measurements comprise mean, maximum, minimum, duration, area, standard deviation, derivative, and / or morphological measurements from each of the systolic upstroke, systolic decay, and diastolic phases from each individual cardiac cycle; and / or wherein the plurality of waveform signal measurements comprise heart rate, respiratory rate, stroke volume, pulse pressure, pulse pressure variation, stroke volume variation, mean arterial pressure (MAP), systolic pressure (SYS), diastolic pressure (DIA), heart rate variability, cardiac output, peripheral resistance, vascular elasticity, and / or left ventricular contractility extracted from each individual cardiac cycle.

[0100] a first clinical data set, a second clinical data set, and a third clinical data set, the step of calculating a composite measure between the plurality of waveform signal measures of the first clinical data set, the second clinical data set, and the third clinical data set comprising: performing step 1 by arbitrarily selecting a subset of signal measures from the plurality of waveform signal measures; performing step 2 by calculating powers of different orders for each of the subset of signal measures to generate a power of the subset of signal measures; performing step 3 by multiplying the powers of the subset of signal measures together to generate a product of the powers of the subset of signal measures; performing step 4 by performing a receiver operating characteristic (ROC) analysis of the product to obtain a composite measure for the subset of signal measures; and repeating steps 1, 2, 3, and 4 until all of the composite measures have been calculated between all of the plurality of waveform signal measures.

[0101] A further embodiment of the hemodynamic monitor as described above, wherein the input features comprise a first subset and a second subset, and the instructions, when executed by the system processor, are further configured to: extract the first subset and the second subset of input features simultaneously from the plurality of signal measurements; simultaneously determine the patient's normal ejection fraction score from the first subset of input features and the patient's low ejection fraction score from the second subset of input features; and output the patient's normal ejection fraction score and the patient's low ejection fraction score to a display of a user interface.

[0102] a third subset of input features; and a third subset of input features simultaneously from the plurality of signal measurements; and a fourth ...

[0103] In another example, a hemodynamic monitor for detecting heart failure is disclosed. The hemodynamic monitor includes an arterial blood pressure sensor with a housing, a fluid input port connected to a fluid source via tubing, a catheter-side fluid port connected to a catheter inserted into a patient's arterial system, a pressure transducer in communication with the fluid source through the fluid port, and an I / O cable in electrical communication with the pressure transducer. The hemodynamic monitor also includes an integrated hardware unit with a system processor, system memory, a display with a user interface, and an analog-to-digital converter (ADC). The system memory includes instructions that, when executed by the system processor, are configured to: receive from a pressure transducer an electrical signal based on pressure in the patient's arterial system transmitted through the fluid source over a period of time; convert the electrical signal to a digital signal; generate arterial pressure waveform data for the patient based on the digital signal; extract a plurality of signal measurements from the arterial pressure waveform data; extract input features from the plurality of signal measurements indicative of the patient's ejection fraction score; determine the patient's ejection fraction score based on the extracted input features; generate a first sensory alert signal configured to generate a first sensory alert indicating that the patient's ejection fraction is low when the ejection fraction score exceeds a threshold score, or generate a second sensory alert signal configured to generate a second sensory alert indicating that the patient's ejection fraction is not low when the ejection fraction score is below the threshold score; transmit the first sensory alert signal or the second sensory alert signal to a user interface; and output the first sensory alert or the second sensory alert through the user interface.

[0104] The hemodynamic monitor of the preceding paragraph may optionally include, in addition and / or in the alternative, any one or more of the following features, configurations, and / or additional components listed below.

[0105] a second clinical data set including arterial pressure waveforms from a second group of individuals with a low ejection fraction measurement of 40 percent or less; and a third clinical data set including arterial pressure waveforms from a third group of individuals with a borderline ejection fraction measurement within a range of 41 percent to 49 percent; performing waveform analysis of the arterial pressure waveforms of the first clinical data set, the second clinical data set, and the third clinical data set to calculate a plurality of waveform signal measurements; and determining the input features by calculating a composite measurement among the plurality of waveform signal measurements, selecting a top signal measurement from the plurality of waveform signal measurements having a most predictive composite measurement, and labeling the top signal measurement as the input feature.

[0106] A further embodiment of the hemodynamic monitor described above, wherein performing waveform analysis of the arterial pressure waveforms of the first clinical data set, the second clinical data set, and the third clinical data set to calculate a plurality of waveform signal measurements comprises identifying individual cardiac cycles in each of the arterial pressure waveforms of the first clinical data set, the second clinical data set, and the third clinical data set; identifying a dicrotic notch in each of the individual cardiac cycles; identifying a systolic rise phase, a systolic decay phase, and a diastolic phase in each of the individual cardiac cycles; and extracting a plurality of waveform signal measurements from each of the systolic rise phase, the systolic decay phase, and the diastolic phase from each of the individual cardiac cycles.

[0107] A further embodiment of the hemodynamic monitor described above, wherein the plurality of waveform signal measurements correspond to hemodynamic effects from each of a systolic upstroke phase, a systolic decay phase, and a diastolic phase from each individual cardiac cycle, the hemodynamic effects comprising contractility, aortic elasticity, stroke volume, vascular tone, afterload, and the entire cardiac cycle.

[0108] Further embodiments of the hemodynamic monitor described above, wherein the plurality of waveform signal measurements comprise mean, maximum, minimum, duration, area, standard deviation, derivative, and / or morphological measurements from each of the systolic upstroke, systolic decay, and diastolic phases from each individual cardiac cycle; and / or wherein the plurality of waveform signal measurements comprise heart rate, respiratory rate, stroke volume, pulse pressure, pulse pressure variation, stroke volume variation, mean arterial pressure (MAP), systolic pressure (SYS), diastolic pressure (DIA), heart rate variability, cardiac output, peripheral resistance, vascular elasticity, and / or left ventricular contractility extracted from each individual cardiac cycle.

[0109] a first clinical data set, a second clinical data set, and a third clinical data set, the step of calculating a composite measure between the plurality of waveform signal measures of the first clinical data set, the second clinical data set, and the third clinical data set comprising: performing step 1 by arbitrarily selecting a subset of signal measures from the plurality of waveform signal measures; performing step 2 by calculating powers of different orders for each of the subset of signal measures to generate a power of the subset of signal measures; performing step 3 by multiplying the powers of the subset of signal measures together to generate a product of the powers of the subset of signal measures; performing step 4 by performing a receiver operating characteristic (ROC) analysis of the product to obtain a composite measure for the subset of signal measures; and repeating steps 1, 2, 3, and 4 until all of the composite measures have been calculated between all of the plurality of waveform signal measures.

[0110] a first subset of input features and a second subset of input features; and a second subset of input features. The instructions, when executed by a system processor, are further configured to: extract the first subset of input features and the second subset of input features simultaneously from the plurality of signal measurements; simultaneously determine a normal ejection fraction score for the patient from the first subset of input features and a low ejection fraction score for the patient from the second subset of input features; determine an ejection fraction measurement for the patient based on the normal ejection fraction score for the patient and the low ejection fraction score for the patient; and output the ejection fraction measurement to a display of a user interface.

[0111] a third subset of input features; and a third subset of input features simultaneously from the plurality of signal measurements; and a fourth ...

[0112] In another example, a method for triaging a patient for risk of heart failure is disclosed. The method includes receiving, by a hemodynamic monitor, sensed hemodynamic data representing the patient's arterial pressure waveform. The hemodynamic monitor performs waveform analysis of the sensed hemodynamic data to calculate a plurality of signal measurements of the sensed hemodynamic data. The method further includes extracting, by the hemodynamic monitor, input features indicative of the patient's ejection fraction from the plurality of signal measurements. The extracting input features includes extracting a first subset of input features and extracting a second subset of input features simultaneously with the first subset of input features. The method further includes simultaneously determining, by the hemodynamic monitor, a normal ejection fraction score for the patient from the first subset of input features and a low ejection fraction score for the patient from the second subset of input features. The hemodynamic monitor outputs the normal ejection fraction score and the low ejection fraction score for the patient to a display and / or a mobile device.

[0113] The method of the preceding paragraph may optionally include, in addition and / or in the alternative, any one or more of the following features, configurations, and / or additional components listed below.

[0114] A further embodiment of the above method, wherein extracting input features comprises extracting a third subset of input features simultaneously with the first and second subsets of input features, wherein the hemodynamic monitor simultaneously determines the patient's normal ejection fraction score from the first subset of input features, the patient's low ejection fraction score from the second subset of input features, and the patient's borderline ejection fraction score from the third subset of input features, and wherein the hemodynamic monitor outputs the patient's normal ejection fraction score, the patient's low ejection fraction score, and the patient's borderline ejection fraction score to a display and / or mobile device.

[0115] The method further comprises training the hemodynamic monitor to determine the patient's ejection fraction, wherein training the hemodynamic monitor includes collecting a first clinical dataset including arterial pressure waveforms from a first group of individuals with normal ejection fraction measurements of 50 percent or greater; labeling each of the arterial pressure waveforms of the first clinical dataset with a first label; performing waveform analysis of the labeled arterial pressure waveforms of the first clinical dataset to calculate a plurality of waveform signal measurements of the first clinical dataset; determining a first subset of input features by calculating a composite measure among the plurality of waveform signal measurements of the first clinical dataset, selecting top signal measurements from the plurality of waveform signal measurements of the first clinical dataset that have a most predictive composite measure, and labeling the top signal measurements of the first clinical dataset as a first subset of input features; collecting a second clinical dataset including arterial pressure waveforms from a second group of individuals with low ejection fraction measurements of 40 percent or less; labeling each of the arterial pressure waveforms of the second clinical dataset with a second label; performing waveform analysis of the labeled arterial pressure waveforms of the second clinical dataset to calculate a plurality of waveform signal measurements for the dataset; determining a second subset of input features by calculating composite measurements between the plurality of waveform signal measurements of the second clinical dataset and selecting the top signal measurements from the plurality of waveform signal measurements of the second clinical dataset having the most predictive composite measurements and labeling the top signal measurements of the second clinical dataset as the second subset of input features; collecting a third clinical dataset including arterial pressure waveforms from a third group of individuals with borderline ejection fraction measurements within a range of 41 percent to 49 percent; labeling each of the arterial pressure waveforms of the third clinical dataset with a third label; performing waveform analysis of the labeled arterial pressure waveforms of the third clinical dataset to calculate a plurality of waveform signal measurements for the third clinical dataset; calculating composite measurements between the plurality of waveform signal measurements of the third clinical dataset and selecting the top signal measurements from the plurality of waveform signal measurements of the third clinical dataset having the most predictive composite measurements;and determining a third subset of input features by labeling top signal measurements of the third clinical dataset as the third subset of input features.

[0116] A further embodiment of the above-described method, wherein the step of performing waveform analysis of the labeled arterial pressure waveforms of the first clinical dataset to calculate a plurality of waveform signal measurements of the first clinical dataset comprises the steps of: identifying individual cardiac cycles in each of the arterial pressure waveforms of the first clinical dataset; identifying a dicrotic notch in each of the individual cardiac cycles in each of the arterial pressure waveforms of the first clinical dataset; identifying a systolic rise phase, a systolic decay phase, and a diastolic phase in each of the individual cardiac cycles in each of the arterial pressure waveforms of the first clinical dataset; and extracting a plurality of waveform signal measurements of the first clinical dataset from each of the systolic rise phase, the systolic decay phase, and the diastolic phase from each of the individual cardiac cycles in each of the arterial pressure waveforms of the first clinical dataset.

[0117] A further embodiment of the above-described method, wherein the step of performing waveform analysis of the labeled arterial pressure waveforms of the second clinical dataset to calculate a plurality of waveform signal measurements of the second clinical dataset comprises the steps of: identifying individual cardiac cycles in each of the arterial pressure waveforms of the second clinical dataset; identifying a dicrotic notch in each of the individual cardiac cycles in each of the arterial pressure waveforms of the second clinical dataset; identifying a systolic rise phase, a systolic decay phase, and a diastolic phase in each of the individual cardiac cycles in each of the arterial pressure waveforms of the second clinical dataset; and extracting a plurality of waveform signal measurements of the second clinical dataset from each of the systolic rise phase, the systolic decay phase, and the diastolic phase from each of the individual cardiac cycles in each of the arterial pressure waveforms of the second clinical dataset.

[0118] A further embodiment of the above-described method, wherein the step of performing waveform analysis of the labeled arterial pressure waveforms of the third clinical dataset to calculate a plurality of waveform signal measurements of the third clinical dataset comprises the steps of: identifying individual cardiac cycles in each of the arterial pressure waveforms of the third clinical dataset; identifying a dicrotic notch in each of the individual cardiac cycles in each of the arterial pressure waveforms of the third clinical dataset; identifying a systolic rise phase, a systolic decay phase, and a diastolic phase in each of the individual cardiac cycles in each of the arterial pressure waveforms of the third clinical dataset; and extracting a plurality of waveform signal measurements of the third clinical dataset from each of the systolic rise phase, the systolic decay phase, and the diastolic phase from each of the individual cardiac cycles in each of the arterial pressure waveforms of the third clinical dataset.

[0119] The plurality of waveform signal measurements of the first clinical dataset correspond to hemodynamic effects from each of a systolic rise phase, a systolic decay phase, and a diastolic phase from each individual cardiac cycle in each of the arterial pressure waveforms of the first clinical dataset, the hemodynamic effects comprising contractility, aortic elasticity, stroke volume, vascular tone, afterload, and the entire cardiac cycle; and the plurality of waveform signal measurements of the second clinical dataset correspond to hemodynamic effects from each of a systolic rise phase, a systolic decay phase, and a diastolic phase from each individual cardiac cycle in each of the arterial pressure waveforms of the second clinical dataset. a plurality of waveform signal measurements of the third clinical dataset correspond to hemodynamic effects from each of the systolic rise phase, the systolic decay phase, and the diastolic phase from each individual cardiac cycle in each of the arterial pressure waveforms of the third clinical dataset, the hemodynamic effects comprising contractility, aortic elasticity, stroke volume, vascular tone, afterload, and the entire cardiac cycle.

[0120] a plurality of waveform signal measurements of the first clinical dataset comprising a mean, maximum, minimum, time length, area, standard deviation, derivative, and / or morphological measurements from each of a systolic rise phase, a systolic decay phase, and a diastolic phase from each individual cardiac cycle in each of the arterial pressure waveforms of the first clinical dataset; a plurality of waveform signal measurements of the second clinical dataset comprising a mean, maximum, minimum, time length, area, standard deviation, derivative, and / or morphological measurements from each of a systolic rise phase, a systolic decay phase, and a diastolic phase from each individual cardiac cycle in each of the arterial pressure waveforms of the second clinical dataset; and a plurality of waveform signal measurements of the third clinical dataset comprising a mean, maximum, minimum, time length, area, standard deviation, derivative, and / or morphological measurements from each of a systolic rise phase, a systolic decay phase, and a diastolic phase from each individual cardiac cycle in each of the arterial pressure waveforms of the third clinical dataset.

[0121] The plurality of waveform signal measurements of the first clinical dataset comprise heart rate, respiratory rate, stroke volume, pulse pressure, pulse pressure variability, stroke volume variability, mean arterial pressure (MAP), systolic pressure (SYS), diastolic pressure (DIA), heart rate variability, cardiac output, peripheral resistance, vascular elasticity, and / or left ventricular contractility extracted from each individual cardiac cycle in each arterial pressure waveform of the first clinical dataset, and the plurality of waveform signal measurements of the second clinical dataset comprise heart rate, respiratory rate, stroke volume, pulse pressure, pulse pressure variability, stroke volume extracted from each individual cardiac cycle in each arterial pressure waveform of the second clinical dataset. a third clinical dataset comprising: heart rate, respiratory rate, stroke volume, pulse pressure, pulse pressure variability, stroke volume variability, mean arterial pressure (MAP), systolic pressure (SYS), diastolic pressure (DIA), heart rate variability, cardiac output, peripheral resistance, vascular elasticity, and / or left ventricular contractility extracted from each of the individual cardiac cycles in each of the arterial pressure waveforms of the third clinical dataset.

[0122] The step of calculating composite measures between the plurality of waveform signal measures of the first clinical data set includes: performing step 1 by arbitrarily selecting a subset of signal measures from the plurality of waveform signal measures of the first clinical data set; performing step 2 by calculating powers of different orders for each of the subset of signal measures from the plurality of waveform signal measures of the first clinical data set to generate powers of the subset of signal measures from the plurality of waveform signal measures of the first clinical data set; and performing step 3 by calculating powers of different orders for each of the subset of signal measures from the plurality of waveform signal measures of the first clinical data set to generate powers of the subset of signal measures from the plurality of waveform signal measures of the first clinical data set. a receiver operating characteristic (ROC) analysis of the products of the powers of the subset of signal measurements from the plurality of waveform signal measurements of the first clinical data set to obtain a composite measurement for the product of the powers of the subset of signal measurements from the plurality of waveform signal measurements of the first clinical data set; and repeating steps 1, 2, 3, and 4 until all of the composite measurements have been calculated between all of the plurality of waveform signal measurements of the first clinical data set.

[0123] The step of calculating a composite measure between the plurality of waveform signal measures of the second clinical dataset includes: performing step 1 by arbitrarily selecting a subset of signal measures from the plurality of waveform signal measures of the second clinical dataset; performing step 2 by calculating powers of different orders for each of the subset of signal measures from the plurality of waveform signal measures of the second clinical dataset to generate powers of the subset of signal measures from the plurality of waveform signal measures of the second clinical dataset; and performing step 3 by multiplying together the powers of the subset of signal measures from the plurality of waveform signal measures of the second clinical dataset to generate products of the powers of the subset of signal measures from the plurality of waveform signal measures of the second clinical dataset; and performing a receiver operating characteristic (ROC) analysis of the products of the powers of the subset of signal measures from the plurality of waveform signal measures of the second clinical dataset to obtain a composite measure for the products of the powers of the subset of signal measures from the plurality of waveform signal measures of the second clinical dataset. and repeating steps 1, 2, 3, and 4 until all of the composite measurements have been calculated between all of the plurality of waveform signal measurements of the second clinical dataset, wherein the step of calculating composite measurements between the plurality of waveform signal measurements of the third clinical dataset comprises: performing step 1 by arbitrarily selecting a subset of signal measurements from the plurality of waveform signal measurements of the third clinical dataset; performing step 2 by calculating powers of different orders for each of the signal measurements from the subset of signal measurements from the plurality of waveform signal measurements of the third clinical dataset to generate powers of the subset of signal measurements from the plurality of waveform signal measurements of the third clinical dataset; and performing step 3 by multiplying together the powers of the subset of signal measurements from the plurality of waveform signal measurements of the third clinical dataset to generate products of the powers of the subset of signal measurements from the plurality of waveform signal measurements of the third clinical dataset.A further embodiment of the above method, comprising: performing step 4 by performing a receiver operating characteristic (ROC) analysis of products of powers of the subset of signal measurements from the plurality of waveform signal measurements of the third clinical data set to obtain a composite measurement for a product of powers of the subset of signal measurements from the plurality of waveform signal measurements of the third clinical data set; and repeating steps 1, 2, 3, and 4 until all of the composite measurements have been calculated between all of the plurality of waveform signal measurements of the third clinical data set.

[0124] While the invention has been described in terms of exemplary embodiments, it will be understood that various changes may be made and equivalents may be substituted for those elements without departing from the scope of the invention. In addition, many modifications may be made to adapt a particular situation or material to the teachings of the invention without departing from essential scope thereof. Therefore, it is not intended that the invention be limited to the particular embodiments disclosed, but rather that the invention will include all embodiments falling within the scope of the appended claims. [Explanation of symbols]

[0125] 10 Health Monitoring System 12 Display 14 I / O connectors 16 Hemodynamic Sensor 18 Housing 20 Fluid input port 22 Fluid port on catheter side 24 I / O cables 26 Hemodynamic Sensor 28 Inflatable Finger Cuff 30 Cardiac Reference Sensor 34 Hemodynamic Sensor 36 patients 38 Healthcare workers 40 System Processors 42 system memory 44 ADC 46 DAC 48 Aortic Stenosis Software Code 50 First Module 51 Second Module 52 Third Module 54 User Interface 56 Control Elements 58 Perceptual Information 61 First Clinical Data Set 62 Second Clinical Data Set 63 Third Clinical Data Set

Claims

1. 1. A hemodynamic monitor for detecting heart failure, comprising: a non-invasive blood pressure sensor comprising an inflatable blood pressure bladder, a pressure controller pneumatically connected to the inflatable blood pressure bladder, and an optical transmitter and an optical receiver electrically connected to the pressure controller; An integrated hardware unit comprising: a system processor; System memory; a display with a user interface; an integrated hardware unit comprising: Equipped with The system memory comprises instructions that, when executed by the system processor, the pressure controller adjusting the pressure in the inflatable blood pressure bladder to maintain a constant arterial volume of the patient for a period of time based on a feedback signal generated by the optical transmitter and the optical receiver; generating arterial pressure waveform data for the patient based on the adjusted pressure in the inflatable blood pressure bladder over the period of time; extracting a plurality of signal measurements from the arterial pressure waveform data of the patient; extracting input features from the plurality of signal measurements indicative of an ejection fraction score for the patient; determining the ejection fraction score for the patient based on the extracted input features; generating a first sensory alert signal configured to generate a first sensory alert indicating that the patient's ejection fraction is low when the ejection fraction score exceeds a threshold score, or generating a second sensory alert signal configured to generate a second sensory alert indicating that the patient's ejection fraction is not low when the ejection fraction score is below the threshold score; transmitting the first sensory alert signal or the second sensory alert signal to the user interface; outputting the first sensory alert or the second sensory alert through the user interface; a hemodynamic monitor configured to:

2. the input features are determined by machine training, the machine training comprising: collecting a first clinical data set including arterial pressure waveforms from a first group of individuals with normal ejection fraction measurements of 50 percent or greater; collecting a second clinical data set including arterial pressure waveforms from a second group of individuals with low ejection fraction measurements of 40 percent or less; collecting a third clinical data set including arterial pressure waveforms from a third group of individuals with borderline ejection fraction measurements within a range of 41 percent to 49 percent; performing a waveform analysis of the arterial pressure waveforms of the first clinical data set, the second clinical data set, and the third clinical data set to calculate a plurality of waveform signal measurements; determining the input features by calculating a composite measure between the plurality of waveform signal measures, selecting a top signal measure from the plurality of waveform signal measures having a most predictive composite measure, and labeling the top signal measure as the input feature; 10. The hemodynamic monitor of claim 1, comprising:

3. performing a waveform analysis of the arterial pressure waveforms of the first clinical data set, the second clinical data set, and the third clinical data set to calculate the plurality of waveform signal measurements; identifying individual cardiac cycles in each of the arterial pressure waveforms of the first clinical data set, the second clinical data set, and the third clinical data set; identifying a dicrotic notch in each of the individual cardiac cycles; identifying a systolic upstroke phase, a systolic downstroke phase, and a diastolic phase in each of said individual cardiac cycles; extracting the plurality of waveform signal measurements from each of the systolic upstroke phase, the systolic decay phase, and the diastolic phase from each of the individual cardiac cycles; 3. The hemodynamic monitor of claim 2, comprising:

4. 4. The hemodynamic monitor of claim 3, wherein the plurality of waveform signal measurements correspond to hemodynamic effects from each of the systolic upstroke phase, the systolic downstroke phase, and the diastolic downstroke phase from each of the individual cardiac cycles, the hemodynamic effects comprising contractility, aortic elasticity, stroke volume, vascular tone, afterload, and the entire cardiac cycle.

5. the plurality of waveform signal measurements comprise a mean, maximum, minimum, duration, area, standard deviation, derivative, and / or morphological measurement from each of the systolic upstroke phase, the systolic decay phase, and the diastolic phase from each of the individual cardiac cycles; and / or 5. The hemodynamic monitor of claim 4, wherein the plurality of waveform signal measurements comprise heart rate, respiratory rate, stroke volume, pulse pressure, pulse pressure variation, stroke volume variation, mean arterial pressure (MAP), systolic pressure (SYS), diastolic pressure (DIA), heart rate variability, cardiac output, peripheral resistance, vascular elasticity, and / or left ventricular contractility extracted from each of the individual cardiac cycles.

6. calculating the composite measure between the plurality of waveform signal measures of the first clinical data set, the second clinical data set, and the third clinical data set; performing step 1 by arbitrarily selecting a subset of signal measurements from said plurality of waveform signal measurements; performing step 2 by calculating powers of different orders for each of said subsets of signal measurements to generate powers of said subsets of signal measurements; performing step 3 by multiplying together said powers of said subset of signal measurements to generate a product of said powers of said subset of signal measurements; performing step 4 by performing a receiver operating characteristic (ROC) analysis of the product to obtain a composite measurement for the subset of signal measurements; repeating steps 1, 2, 3, and 4 until all of said composite measurements are calculated between all of said plurality of waveform signal measurements; 6. The hemodynamic monitor of claim 5, comprising:

7. the input features comprise a first subset and a second subset, and the instructions, when executed by the system processor, extracting the first and second subsets of input features simultaneously from the plurality of signal measurements; simultaneously determining a normal ejection fraction score for the patient from the first subset of input features and a low ejection fraction score for the patient from the second subset of input features; outputting the normal ejection fraction score of the patient and the low ejection fraction score of the patient to the display of the user interface; 7. The hemodynamic monitor of claim 6, further configured to:

8. the input features comprise a third subset, and the instructions, when executed by the system processor, extracting the first subset, the second subset, and the third subset of the input features simultaneously from the plurality of signal measurements; simultaneously determining the patient's normal ejection fraction score from the first subset of input features, the patient's low ejection fraction score from the second subset of input features, and the patient's borderline ejection fraction score from the third subset of input features; outputting the patient's normal ejection fraction score, the patient's low ejection fraction score, and the patient's borderline ejection fraction score to the display of the user interface; 8. The hemodynamic monitor of claim 7, further configured to:

9. 1. A hemodynamic monitor for detecting heart failure, comprising: an arterial blood pressure sensor comprising a housing, a fluid input port connected to a fluid source via tubing, a catheter fluid port connected to a catheter inserted into a patient's arterial system, a pressure transducer in communication with the fluid source through the fluid port, and an I / O cable in electrical communication with the pressure transducer; An integrated hardware unit comprising: a system processor; System memory; a display having a user interface; Analog-to-digital converter (ADC) and an integrated hardware unit comprising: Equipped with The system memory comprises instructions that, when executed by the system processor, receiving, over a period of time, from the pressure transducer, an electrical signal based on pressure in the arterial system of the patient transmitted through the fluid source; converting the electrical signal into a digital signal; generating arterial pressure waveform data for the patient based on the digital signal; extracting a plurality of signal measurements from the arterial pressure waveform data; extracting input features from the plurality of signal measurements indicative of an ejection fraction score for the patient; determining the ejection fraction score for the patient based on the extracted input features; generating a first sensory alert signal configured to generate a first sensory alert indicating that the patient's ejection fraction is low when the ejection fraction score exceeds a threshold score, or generating a second sensory alert signal configured to generate a second sensory alert indicating that the patient's ejection fraction is not low when the ejection fraction score is below the threshold score; transmitting the first sensory alert signal or the second sensory alert signal to the user interface; outputting the first sensory alert or the second sensory alert through the user interface; a hemodynamic monitor configured to:

10. The input features of the ejection fraction software code are determined by machine training, and the machine training comprises: collecting a first clinical data set including arterial pressure waveforms from a first group of individuals with normal ejection fraction measurements of 50 percent or greater; collecting a second clinical data set including arterial pressure waveforms from a second group of individuals with low ejection fraction measurements of 40 percent or less; collecting a third clinical data set including arterial pressure waveforms from a third group of individuals with borderline ejection fraction measurements within a range of 41 percent to 49 percent; performing a waveform analysis of the arterial pressure waveforms of the first clinical data set, the second clinical data set, and the third clinical data set to calculate a plurality of waveform signal measurements; determining the input features by calculating a composite measure between the plurality of waveform signal measures, selecting a top signal measure from the plurality of waveform signal measures having a most predictive composite measure, and labeling the top signal measure as the input feature; 10. The hemodynamic monitor of claim 9, comprising:

11. performing a waveform analysis of the labeled arterial pressure waveforms of the first clinical data set, the second clinical data set, and the third clinical data set to calculate the plurality of waveform signal measurements; identifying individual cardiac cycles in each of the arterial pressure waveforms of the first clinical data set, the second clinical data set, and the third clinical data set; identifying a dicrotic notch in each of the individual cardiac cycles; identifying a systolic upstroke phase, a systolic downstroke phase, and a diastolic phase in each of said individual cardiac cycles; extracting the plurality of waveform signal measurements from each of the systolic upstroke phase, the systolic decay phase, and the diastolic phase from each of the individual cardiac cycles; 11. The hemodynamic monitor of claim 10, comprising:

12. 12. The hemodynamic monitor of claim 11, wherein the plurality of waveform signal measurements correspond to hemodynamic effects from each of the systolic upstroke phase, the systolic downstroke phase, and the diastolic downstroke phase from each of the individual cardiac cycles, the hemodynamic effects comprising contractility, aortic elasticity, stroke volume, vascular tone, afterload, and the entire cardiac cycle.

13. the plurality of waveform signal measurements comprise a mean, maximum, minimum, duration, area, standard deviation, derivative, and / or morphological measurement from each of the systolic upstroke phase, the systolic decay phase, and the diastolic phase from each of the individual cardiac cycles; and / or 13. The hemodynamic monitor of claim 12, wherein the plurality of waveform signal measurements comprise heart rate, respiratory rate, stroke volume, pulse pressure, pulse pressure variation, stroke volume variation, mean arterial pressure (MAP), systolic pressure (SYS), diastolic pressure (DIA), heart rate variability, cardiac output, peripheral resistance, vascular elasticity, and / or left ventricular contractility extracted from each of the individual cardiac cycles.

14. calculating the composite measure between the plurality of waveform signal measures of the first clinical data set, the second clinical data set, and the third clinical data set; performing step 1 by arbitrarily selecting a subset of signal measurements from said plurality of waveform signal measurements; performing step 2 by calculating powers of different orders for each of said subsets of signal measurements to generate powers of said subsets of signal measurements; performing step 3 by multiplying together said powers of said subset of signal measurements to generate a product of said powers of said subset of signal measurements; performing step 4 by performing a receiver operating characteristic (ROC) analysis of the product to obtain a composite measurement for the subset of signal measurements; repeating steps 1, 2, 3, and 4 until all of said composite measurements are calculated between all of said plurality of waveform signal measurements; 14. The hemodynamic monitor of claim 13, comprising:

15. the input features comprise a first subset and a second subset, and the instructions, when executed by the system processor, extracting the first and second subsets of input features simultaneously from the plurality of signal measurements; simultaneously determining a normal ejection fraction score for the patient from the first subset of input features and a low ejection fraction score for the patient from the second subset of input features; outputting the normal ejection fraction score of the patient and the low ejection fraction score of the patient to the display of the user interface; 15. The hemodynamic monitor of claim 14, further configured to:

16. the input features comprise a third subset, and the instructions, when executed by the system processor, extracting the first subset, the second subset, and the third subset of the input features simultaneously from the plurality of signal measurements; simultaneously determining the patient's normal ejection fraction score from the first subset of input features, the patient's low ejection fraction score from the second subset of input features, and the patient's borderline ejection fraction score from the third subset of input features; outputting the normal ejection fraction score of the patient, the low ejection fraction score of the patient, and the borderline ejection fraction score of the patient to the display of the user interface; 16. The hemodynamic monitor of claim 15, further configured to:

17. 1. A method for triaging a patient for risk of heart failure, comprising: receiving, by a hemodynamic monitor, sensed hemodynamic data representative of an arterial pressure waveform of the patient; the hemodynamic monitor performing waveform analysis of the sensed hemodynamic data to calculate a plurality of signal measurements of the sensed hemodynamic data; the hemodynamic monitor extracting input features from the plurality of signal measurements indicative of an ejection fraction score for the patient; The step of extracting the input feature amount includes: extracting a first subset of the input features; extracting a second subset of the input features simultaneously with the first subset of the input features; and the hemodynamic monitor simultaneously determining a normal ejection fraction score for the patient from the first subset of input features and a low ejection fraction score for the patient from the second subset of input features; outputting the normal ejection fraction score and the low ejection fraction score of the patient to a display and / or mobile device; A method comprising:

18. The step of extracting the input feature amount includes: extracting a third subset of the input features simultaneously with the first and second subsets of the input features; the hemodynamic monitor simultaneously determines the patient's normal ejection fraction score from the first subset of input features, the patient's low ejection fraction score from the second subset of input features, and the patient's borderline ejection fraction score from the third subset of input features; 18. The method of claim 17, wherein the hemodynamic monitor outputs the patient's normal ejection fraction score, the patient's low ejection fraction score, and the patient's borderline ejection fraction score to the display and / or the mobile device.

19. training the hemodynamic monitor to determine the ejection fraction score for the patient, the training of the hemodynamic monitor comprising: collecting a first clinical data set including arterial pressure waveforms from a first group of individuals with normal ejection fraction measurements of 50 percent or greater; labeling each of the arterial pressure waveforms of the first clinical data set with a first label; performing waveform analysis of the labeled arterial pressure waveform of the first clinical data set to calculate a plurality of waveform signal measurements of the first clinical data set; determining the first subset of input features by calculating composite measures between the plurality of waveform signal measures of the first clinical dataset, selecting top signal measures from the plurality of waveform signal measures of the first clinical dataset having the most predictive composite measures, and labeling the top signal measures of the first clinical dataset as the first subset of input features; collecting a second clinical data set including arterial pressure waveforms from a second group of individuals with low ejection fraction measurements of 40 percent or less; labeling each of the arterial pressure waveforms of the second clinical data set with a second label; performing waveform analysis of the labeled arterial pressure waveform of the second clinical data set to calculate a plurality of waveform signal measurements of the second clinical data set; determining the second subset of input features by calculating a composite measure between the plurality of waveform signal measures of the second clinical dataset, selecting top signal measures from the plurality of waveform signal measures of the second clinical dataset having the most predictive composite measure, and labeling the top signal measures of the second clinical dataset as the second subset of input features; collecting a third clinical data set including arterial pressure waveforms from a third group of individuals with borderline ejection fraction measurements within a range of 41 percent to 49 percent; labeling each of the arterial pressure waveforms of the third clinical data set with a third label; performing waveform analysis of the labeled arterial pressure waveform of the third clinical data set to calculate a plurality of waveform signal measurements of the third clinical data set; determining the third subset of input features by calculating composite measures between the plurality of waveform signal measures of the third clinical data set, selecting top signal measures from the plurality of waveform signal measures of the third clinical data set having the most predictive composite measures, and labeling the top signal measures of the third clinical data set as the third subset of input features; 20. The method of claim 18, comprising:

20. performing a waveform analysis of the labeled arterial pressure waveform of the first clinical data set to calculate the plurality of waveform signal measurements of the first clinical data set, identifying individual cardiac cycles in each of the arterial pressure waveforms of the first clinical data set; identifying a dicrotic notch in each of the individual cardiac cycles in each of the arterial pressure waveforms of the first clinical data set; identifying a systolic upstroke phase, a systolic decay phase, and a diastolic phase in each of the individual cardiac cycles in each of the arterial pressure waveforms of the first clinical data set; extracting the plurality of waveform signal measurements of the first clinical data set from each of the systolic upstroke phase, the systolic decay phase, and the diastolic phase from each of the individual cardiac cycles in each of the arterial pressure waveforms of the first clinical data set; 20. The method of claim 19, comprising:

21. performing a waveform analysis of the labeled arterial pressure waveform of the second clinical data set to calculate the plurality of waveform signal measurements of the second clinical data set, identifying individual cardiac cycles in each of the arterial pressure waveforms of the second clinical data set; identifying a dicrotic notch in each of the individual cardiac cycles in each of the arterial pressure waveforms of the second clinical data set; identifying a systolic upstroke phase, a systolic decay phase, and a diastolic phase in each of the individual cardiac cycles in each of the arterial pressure waveforms of the second clinical data set; extracting the plurality of waveform signal measurements of the second clinical data set from each of the systolic upstroke phase, the systolic decay phase, and the diastolic phase from each of the individual cardiac cycles in each of the arterial pressure waveforms of the second clinical data set; 21. The method of claim 20, comprising:

22. performing a waveform analysis of the labeled arterial pressure waveform of the third clinical data set to calculate the plurality of waveform signal measurements of the third clinical data set, identifying individual cardiac cycles in each of the arterial pressure waveforms of the third clinical data set; identifying a dicrotic notch in each of the individual cardiac cycles in each of the arterial pressure waveforms of the third clinical data set; identifying a systolic upstroke phase, a systolic decay phase, and a diastolic phase in each of the individual cardiac cycles in each of the arterial pressure waveforms of the third clinical data set; extracting the plurality of waveform signal measurements of the third clinical data set from each of the systolic upstroke phase, the systolic decay phase, and the diastolic phase from each of the individual cardiac cycles in each of the arterial pressure waveforms of the third clinical data set; 22. The method of claim 21, comprising:

23. the plurality of waveform signal measurements of the first clinical dataset correspond to hemodynamic effects from each of the systolic upstroke phase, the systolic downstroke phase, and the diastolic downstroke phase from each of the individual cardiac cycles in each of the arterial pressure waveforms of the first clinical dataset, the hemodynamic effects comprising contractility, aortic elasticity, stroke volume, vascular tone, afterload, and an entire cardiac cycle; the plurality of waveform signal measurements of the second clinical dataset correspond to hemodynamic effects from each of the systolic upstroke phase, the systolic downstroke phase, and the diastolic downstroke phase from each of the individual cardiac cycles in each of the arterial pressure waveforms of the second clinical dataset, the hemodynamic effects comprising contractility, aortic elasticity, stroke volume, vascular tone, afterload, and an entire cardiac cycle; 23. The method of claim 22, wherein the plurality of waveform signal measurements of the third clinical dataset correspond to hemodynamic effects from each of the systolic upstroke phase, the systolic decay phase, and the diastolic phase from each of the individual cardiac cycles in each of the arterial pressure waveforms of the third clinical dataset, the hemodynamic effects comprising contractility, aortic elasticity, stroke volume, vascular tone, afterload, and an entire cardiac cycle.

24. the plurality of waveform signal measurements of the first clinical data set comprise a mean, a maximum, a minimum, a duration, an area, a standard deviation, a derivative, and / or a morphological measurement from each of the systolic upstroke phase, the systolic decay phase, and the diastolic phase from each of the individual cardiac cycles in each of the arterial pressure waveforms of the first clinical data set; the plurality of waveform signal measurements of the second clinical dataset comprising a mean, a maximum, a minimum, a duration, an area, a standard deviation, a derivative, and / or a morphological measurement from each of the systolic upstroke phase, the systolic decay phase, and the diastolic phase from each of the individual cardiac cycles in each of the arterial pressure waveforms of the second clinical dataset; 24. The method of claim 23, wherein the plurality of waveform signal measurements of the third clinical dataset comprise a mean, a maximum, a minimum, a duration, an area, a standard deviation, a derivative, and / or a morphological measurement from each of the systolic upstroke phase, the systolic decay phase, and the diastolic phase from each of the individual cardiac cycles in each of the arterial pressure waveforms of the third clinical dataset.

25. the plurality of waveform signal measurements of the first clinical data set comprise heart rate, respiratory rate, stroke volume, pulse pressure, pulse pressure variation, stroke volume variation, mean arterial pressure (MAP), systolic pressure (SYS), diastolic pressure (DIA), heart rate variability, cardiac output, peripheral resistance, vascular elasticity, and / or left ventricular contractility extracted from each of the individual cardiac cycles in each of the arterial pressure waveforms of the first clinical data set; the plurality of waveform signal measurements of the second clinical dataset comprising heart rate, respiratory rate, stroke volume, pulse pressure, pulse pressure variation, stroke volume variation, mean arterial pressure (MAP), systolic pressure (SYS), diastolic pressure (DIA), heart rate variability, cardiac output, peripheral resistance, vascular elasticity, and / or left ventricular contractility extracted from each of the individual cardiac cycles in each of the arterial pressure waveforms of the second clinical dataset; 25. The method of claim 24, wherein the plurality of waveform signal measurements of the third clinical data set comprise heart rate, respiratory rate, stroke volume, pulse pressure, pulse pressure variation, stroke volume variation, mean arterial pressure (MAP), systolic pressure (SYS), diastolic pressure (DIA), heart rate variability, cardiac output, peripheral resistance, vascular elasticity, and / or left ventricular contractility extracted from each of the individual cardiac cycles in each of the arterial pressure waveforms of the third clinical data set.

26. calculating the composite measure between the plurality of waveform signal measures of the first clinical data set, performing step 1 by arbitrarily selecting a subset of signal measurements from the plurality of waveform signal measurements of the first clinical data set; performing step 2 by calculating powers of different orders for each of the subset of signal measurements from the plurality of waveform signal measurements of the first clinical data set to generate powers of the subset of signal measurements from the plurality of waveform signal measurements of the first clinical data set; performing step 3 by multiplying together the powers of the subset of signal measurements from the plurality of waveform signal measurements of the first clinical data set to generate a product of the powers of the subset of signal measurements from the plurality of waveform signal measurements of the first clinical data set; performing step 4 by performing a receiver operating characteristic (ROC) analysis of the products of the powers of the subset of signal measurements from the plurality of waveform signal measurements of the first clinical data set to obtain a composite measurement for the products of the powers of the subset of signal measurements from the plurality of waveform signal measurements of the first clinical data set; repeating steps 1, 2, 3, and 4 until all of said composite measurements are calculated between all of said plurality of waveform signal measurements of said first clinical data set; 26. The method of claim 25, comprising:

27. calculating the composite measure between the plurality of waveform signal measures of the second clinical data set, performing step 1 by arbitrarily selecting a subset of signal measurements from the plurality of waveform signal measurements of the second clinical data set; performing step 2 by calculating powers of different orders for each of the subset of signal measurements from the plurality of waveform signal measurements of the second clinical data set to generate powers of the subset of signal measurements from the plurality of waveform signal measurements of the second clinical data set; performing step 3 by multiplying together the powers of the subset of signal measurements from the plurality of waveform signal measurements of the second clinical data set to generate a product of the powers of the subset of signal measurements from the plurality of waveform signal measurements of the second clinical data set; performing step 4 by performing a receiver operating characteristic (ROC) analysis of the products of the powers of the subset of signal measurements from the plurality of waveform signal measurements of the second clinical data set to obtain a composite measurement for the products of the powers of the subset of signal measurements from the plurality of waveform signal measurements of the second clinical data set; repeating steps 1, 2, 3, and 4 until all of said composite measurements are calculated between all of said plurality of waveform signal measurements of said second clinical data set; Equipped with calculating the composite measure between the plurality of waveform signal measures of the third clinical data set, performing step 1 by arbitrarily selecting a subset of signal measurements from the plurality of waveform signal measurements of the third clinical data set; performing step 2 by calculating powers of different orders for each of the signal measurements from the subset of signal measurements from the plurality of waveform signal measurements of the third clinical data set to generate powers of the subset of signal measurements from the plurality of waveform signal measurements of the third clinical data set; performing step 3 by multiplying together the powers of the subset of signal measurements from the plurality of waveform signal measurements of the third clinical data set to generate a product of the powers of the subset of signal measurements from the plurality of waveform signal measurements of the third clinical data set; performing step 4 by performing a receiver operating characteristic (ROC) analysis of the products of the powers of the subset of signal measurements from the plurality of waveform signal measurements of the third clinical data set to obtain a composite measurement for the products of the powers of the subset of signal measurements from the plurality of waveform signal measurements of the third clinical data set; repeating steps 1, 2, 3, and 4 until all of said composite measurements are calculated between all of said plurality of waveform signal measurements of said third clinical data set; 27. The method of claim 26, comprising:

Images