Hypotension prediction with feature transformation for adjustable hypotension threshold

Abstract

A hemodynamic monitoring system monitors arterial blood pressure of a patient and provides a warning to medical personnel of a predicted future hypotensive event of the patient. Waveform analysis is performed on sensed hemodynamic data representative of an arterial pressure waveform of the patient to determine a plurality of hypotension analysis parameters predictive of a future hypotensive event of the patient. A set of transformed hypotension analysis parameters is generated based on the hypotension analysis parameters and a standard mean arterial pressure (MAP) threshold for hypotension and mean and standard deviation values of the hypotension analysis parameters at an adjusted MAP threshold for hypotension. A risk score representative of a probability of a future hypotensive event of the patient is determined based on the set of transformed hypotension analysis parameters. A sensory alarm is invoked to produce a sensory signal in response to the risk score satisfying a predetermined risk criterion.

Description

TECHNICAL FIELD

[0001] The present disclosure relates generally to arterial blood pressure monitoring, and more particularly to hypotension prediction with adjustable hypotension threshold. BACKGROUND

[0002] Hypotension, or low blood pressure, can be a serious medical complication and even a harbinger of death for patients undergoing surgery and those in the intensive care unit (ICU) who are acutely or critically ill. The danger associated with the occurrence of hypotension in a patient is due to the potential damage caused by hypotension itself and the many serious potential medical conditions that hypotension occurrence can mean.

[0003] Hypotension in a surgical or critically ill patient is a serious medical condition in itself. For example, in an operating room (OR) setting, hypotension during surgery is associated with increased mortality and organ damage. Even extremely low blood pressure of short duration during surgery is associated with acute kidney injury and myocardial damage. In critically ill patients, the in-hospital mortality rate can almost double for patients who experience hypotension after emergency intubation. For both surgical and critically ill patients, hypotension, if not corrected, impairs organ perfusion, leading to irreversible ischemic damage, neurological deficits, cardiomyopathy, and kidney damage.

[0004] In addition to posing a serious risk to surgical and critically ill patients in itself, hypotension can be a symptom of one or more other serious underlying medical conditions. Examples of potential conditions for which hypotension can be an acute symptom include sepsis, myocardial infarction, arrhythmia, pulmonary embolism, hemorrhage, dehydration, anaphylaxis, acute reaction to medication, hypovolemia, insufficient cardiac output, and vasodilatory shock. Because hypotension is associated with such a variety of serious medical conditions, hypotension is relatively common and is often viewed as one of the first signs of patient deterioration in the OR and ICU.

[0005] Routine patient monitoring for hypotension in OR and ICU settings can include continuous or periodic blood pressure measurements. However, such monitoring, whether continuous or periodic, typically provides only real-time assessment. Thus, hypotension in a surgical or critically ill patient is typically detected only after it has begun to occur, such that remedial measures and interventions are initiated only after the patient has entered a hypotensive state. Even relatively mild levels of hypotension can presage or contribute to cardiac arrest in patients with limited cardiac reserve, despite the fact that extreme hypotension can have potentially devastating medical consequences quite quickly, as noted above.

[0006] In view of the frequency with which hypotension occurrence is observed in OR and ICU settings, and due to the serious and sometimes immediate medical consequences that can result when hypotension occurs, there is a great need for a solution that enables future hypotension events to be predicted in advance of hypotension occurrence. SUMMARY

[0007] In one example, a method for monitoring arterial pressure of a patient and providing a warning to medical personnel of a predicted future hypotensive event of the patient includes receiving, by a hemodynamic monitor, sensed hemodynamic data representative of an arterial pressure waveform of the patient. The method further includes performing, by the hemodynamic monitor, waveform analysis of the hemodynamic data to determine a plurality of hypotension analysis / profiling parameters that predict a future hypotensive event of the patient, and generating, by the hemodynamic monitor, a set of transformed hypotension analysis parameters. Each transformed hypotension analysis parameter is a function of a corresponding hypotension analysis parameter of the plurality of hypotension analysis parameters at a standard mean arterial pressure (MAP) threshold of hypotension, a mean of the corresponding hypotension analysis parameter of the plurality of hypotension analysis parameters at the standard MAP threshold of hypotension, a standard deviation of the corresponding hypotension analysis parameter of the plurality of hypotension analysis parameters at the standard MAP threshold, a mean of the corresponding hypotension analysis parameter of the plurality of hypotension analysis parameters at an adjusted MAP threshold of hypotension, and a standard deviation of the corresponding hypotension analysis parameter of the plurality of hypotension analysis parameters at the adjusted MAP threshold of hypotension. The method further includes determining, by the hemodynamic monitor, a risk score representative of a probability of the future hypotensive event of the patient based on the set of transformed hypotension analysis parameters, and invoking, by the hemodynamic monitor, a sensory alarm to produce a sensory signal in response to the risk score satisfying a predetermined risk criterion.

[0008] In another example, a system for monitoring arterial pressure of a patient and providing a warning to medical personnel of a predicted future hypotensive event of the patient includes a hemodynamic sensor, a system memory, a user interface, and a hardware processor. The hemodynamic sensor generates hemodynamic data representative of an arterial pressure waveform of the patient. The system memory stores hypotension prediction software code including a prediction weighting module. The user interface includes a sensory alarm that provides a sensory signal to warn the medical personnel of a predicted future hypotensive event prior to the patient entering a hypotensive state. The hardware processor is configured to execute the hypotension prediction software code to perform waveform analysis of the hemodynamic data to determine a plurality of hypotension analysis parameters that predict a future hypotensive event of the patient and generate a set of transformed hypotension analysis parameters. Each transformed hypotension analysis parameter is a function of a corresponding hypotension analysis parameter of the plurality of hypotension analysis parameters at a standard mean arterial pressure (MAP) threshold of hypotension, a mean of the corresponding hypotension analysis parameter of the plurality of hypotension analysis parameters at the standard MAP threshold of hypotension, a standard deviation of the corresponding hypotension analysis parameter of the plurality of hypotension analysis parameters at the standard MAP threshold, a mean of the corresponding hypotension analysis parameter of the plurality of hypotension analysis parameters at an adjusted MAP threshold of hypotension, and a standard deviation of the corresponding hypotension analysis parameter of the plurality of hypotension analysis parameters at the adjusted MAP threshold of hypotension. The hardware processor is further configured to determine a risk score representative of a probability of the future hypotensive event of the patient based on the set of transformed hypotension analysis parameters and invoke the sensory alarm of the user interface in response to the risk score satisfying a predetermined risk criterion. BRIEF DESCRIPTION OF DRAWINGS

[0009] Figure 1 is a perspective view of an example hemodynamic monitor that determines a risk score representative of a probability of a future hypotensive event of a patient.

[0010] Figure 2 is a perspective view of an example minimally invasive pressure sensor for sensing hemodynamic data representative of arterial pressure of a patient.

[0011] Figure 3 is a perspective view of an example non-invasive sensor for sensing hemodynamic data representative of arterial pressure of a patient.

[0012] Figure 4 is a block diagram of an example hemodynamic monitoring system that illustrates determining a risk score representative of a probability of a future hypotensive event of a patient based on a set of transformed hypotension analysis parameters.

[0013] Figure 5is a graph illustrating example traces of arterial pressure waveforms, including example markers corresponding to probabilities of future hypotension for a patient.

[0014] Figure 6 is a flowchart illustrating example operations of a hemodynamic monitoring system generating a transformed set of hypotension analysis parameters and determining a risk score representing a probability of a future hypotension event using the transformed set of hypotension analysis parameters. DETAILED DESCRIPTION

[0015] As described herein, a hemodynamic monitoring system implements a predictive risk model that produces a risk score representing a probability of a future hypotension event for a patient. The risk score is determined based on hypotension analysis parameters that predict future hypotension events. A risk coefficient is implemented that is weighted based on a standard (or defined) mean arterial pressure (MAP) threshold for hypotension, such as a pressure of 65 millimeters of mercury (mmHg) or other defined pressure threshold. The selection of the risk coefficient and / or the hypotension analysis parameters can be achieved via training (e.g., offline training) of the predictive risk model using machine learning or other techniques to minimize a cost function representing error of the predictive risk model output to true values of a training subset defining hypotension according to the standard MAP threshold for hypotension.

[0016] According to the techniques of the present disclosure, the hemodynamic monitoring system can utilize an adjustable MAP threshold for hypotension to represent a modified hypotension pressure threshold. Rather than modifying the predictive risk model (via retraining or otherwise) to accommodate the adjustable (e.g., user-defined or otherwise adjusted) MAP threshold, the hemodynamic monitoring system generates a transformed set of hypotension analysis parameters. Each transformed hypotension analysis parameter is a function of a hypotension analysis parameter and mean and standard deviation values of the hypotension analysis parameter at the standard MAP threshold for hypotension and at the adjusted MAP threshold. That is, rather than retraining or otherwise modifying the predictive risk model to determine new risk coefficients based on a modified definition of hypotension (i.e., the adjusted MAP threshold), the hemodynamic monitoring system adjusts the hypotension analysis parameters (or features) extracted from the hemodynamic data to accommodate the adjusted MAP threshold for hypotension. The risk score representing a probability of a future hypotension event for the patient is determined based on the transformed set of hypotension analysis parameters.

[0017] Therefore, hemodynamic monitoring systems implementing the technology of this disclosure can utilize adjustable pressure thresholds for hypotension without requiring retraining or other modifications to the predictive risk model, thereby enabling real-time updates of the hypotension thresholds during procedures in settings such as the operating room (OR), intensive care unit (ICU), or other patient care environments. Consequently, the system can provide a risk score representing the probability of future hypotension in a patient, facilitating timely and effective intervention, while also leveraging the training and / or experience of healthcare personnel to ensure the use of modified hypotension thresholds, thus increasing the system's usability to patient care personnel.

[0018] Figure 1 This is a perspective view of a hemodynamic monitor 10, which uses a risk score to determine the probability of future hypotensive events in a patient. (Example) Figure 1 As shown, the hemodynamic monitor 10 includes a display 12, in Figure 1 In the example, display 12 presents a graphical user interface (GUI) that includes control elements (e.g., graphical control elements) that enable a user to interact with hemodynamic monitor 10. Hemodynamic monitor 10 may also include multiple input and / or output (I / O) connectors configured for wired (e.g., electrical and / or communication connections) connections to one or more peripheral components (such as one or more hemodynamic sensors), as further described below. For example, as Figure 1 As shown, the hemodynamic monitor 10 may include an I / O connector 14. Although Figure 1 The example illustration shows five individual I / O connectors 14; however, it should be understood that in other examples, the hemodynamic monitor 10 may include fewer than five or more I / O connectors. In other examples, the hemodynamic monitor 10 may not include I / O connectors 14 and may instead communicate wirelessly with various peripheral devices.

[0019] As further described below, the hemodynamic monitor 10 includes one or more processors and a computer-readable storage device storing hypotension prediction software code executable to generate a risk score representing the probability of future hypotension events in the patient. For example, the hemodynamic monitor 10 may receive sensed hemodynamic data representing the patient's arterial pressure waveform, such as via one or more hemodynamic sensors connected to the hemodynamic monitor 10 via I / O connector 14. The hemodynamic monitor 10 executes the hypotension prediction software code to obtain multiple hypotension analysis parameters (e.g., characteristics) using the received hemodynamic data, which may include one or more vital sign parameters characterizing the patient's vital sign data, and differential and combined parameters derived from the one or more vital sign parameters, as further described below.

[0020] As described herein, the hemodynamic monitor 10 may also utilize an adjusted MAP threshold for hypotension, which represents the deviation from a standard MAP threshold, from which a coefficient used by the hypotension prediction software code is determined. For example, the hemodynamic monitor 10 may present a graphical control element (e.g., at a graphical user interface presented on display 12) that allows the user to input the adjusted MAP threshold for hypotension, but input received via physical controls (e.g., buttons, knobs, or other physical input controls) is also possible.

[0021] For example, such as Figure 1 As shown, the hemodynamic monitor 10 can present a graphical user interface on the display 12. The 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 devices suitable for providing information to the user in graphical form. In some examples, such as Figure 1 For example, display 12 may 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. Hemodynamic monitor 10 presents control elements that enable the user to input adjusted MAP thresholds, such as absolute pressure (e.g., MAP threshold), deviation values ​​(e.g., deviation from a standard MAP threshold), or other indications that in some examples may be user-defined (such as those defined by a healthcare professional) adjusted MAP thresholds.

[0022] In response to receiving an adjusted MAP threshold, the hemodynamic monitor 10 executes hypotension prediction software code to generate a transformed set of hypotension analysis parameters. As further described below, each transformed hypotension analysis parameter is a function of the corresponding hypotension analysis parameter among multiple hypotension analysis parameters at the standard mean arterial pressure (MAP) threshold for hypotension, the mean of the corresponding hypotension analysis parameter among multiple hypotension analysis parameters at the standard MAP threshold for hypotension, the standard deviation of the corresponding hypotension analysis parameter among multiple hypotension analysis parameters at the standard MAP threshold, the mean of the corresponding hypotension analysis parameter among multiple hypotension analysis parameters at the adjusted MAP threshold for hypotension, and the standard deviation of the corresponding hypotension analysis parameter among multiple hypotension analysis parameters at the adjusted MAP threshold for hypotension.

[0023] The hemodynamic monitor 10 executes the hypotension prediction software code to apply a plurality of risk coefficients to the set of transformed hypotension analysis parameters to produce a weighted combination to produce a risk score representing a probability of a future hypotension event for the patient. As described in further detail below, the plurality of risk coefficients can be determined based on a standard MAP threshold (e.g., 65 mmHg) or other defined pressure threshold. Thus, rather than retraining the prediction model to identify new coefficients based on an adjusted MAP threshold, the hemodynamic monitor 10 executes the hypotension prediction software code to determine the risk score by applying risk coefficients determined based on the standard MAP threshold to the set of transformed hypotension analysis parameters, thereby enabling the model to dynamically adapt to an adjusted MAP threshold that can be based on the training and expertise of medical personnel.

[0024] Figure 2 is a perspective view of a hemodynamic sensor 16 that can be attached to a patient for sensing hemodynamic data representing arterial pressure of the patient. Figure 2 The illustrated hemodynamic sensor 16 is one example of a minimally-invasive hemodynamic sensor that can be attached to a patient via, for example, a radial artery catheter inserted into an arm of the patient. In other examples, the hemodynamic sensor 16 can be attached to a patient via a femoral artery catheter inserted into a leg of the patient.

[0025] As Figure 2 illustrated, the hemodynamic sensor 16 includes a housing 18, a fluid input port 20, a catheter-side fluid port 22, and an I / O cable 24. The 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. The 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 an arm of the patient (i.e., a radial artery catheter) or a leg of the patient (i.e., a femoral artery catheter) via tubing or other hydraulic connection. The I / O cable 24 is configured to be connected to the hemodynamic monitor 10 via one or more of the I / O connectors 14 Figure 1 ) of the hemodynamic monitor 10. The housing 18 of the hemodynamic sensor 16 encloses one or more pressure transducers, communication circuitry, processing circuitry, and corresponding electronic components to sense fluid pressure corresponding to arterial pressure of the patient that is transmitted to the hemodynamic monitor 10 Figure 1 ) via the I / O cable 24.

[0026] In operation, a fluid column (e.g., saline solution) is introduced from a fluid source (e.g., a saline bag) via fluid inlet port 20, through hemodynamic sensor 16, toward a catheter inserted into the patient, reaching 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 the corresponding electrical signal to hemodynamic monitor 10 via I / O cable 24. Figure 1 Therefore, hemodynamic sensor 16 transmits analog sensor data (or a digital representation of analog sensor data) representing substantially continuous beat-by-beat monitoring of the patient's arterial pressure to hemodynamic monitor 10. Figure 1 ).

[0027] Figure 3 This is a perspective view of a hemodynamic sensor 26 used to sense hemodynamic data representing a patient's arterial pressure. Figure 3 The hemodynamic sensor 26 shown is an example of a non-invasive hemodynamic sensor that can be attached to a patient via one or more finger cots to sense data representing the patient's arterial pressure. Figure 3 As shown, the hemodynamic sensor 26 includes an inflatable finger cot 28 and a cardiac reference sensor 30. The inflatable finger cot 28 includes an inflatable blood pressure cuff configured to inflate and deflate under the control of a pressure controller (not shown) pneumatically connected to the inflatable finger cot 28. The inflatable finger cot 28 also includes an optical (e.g., infrared) emitter and an optical receiver electrically connected to the cardiac reference sensor 30 to measure changes in the volume of arteries in the finger.

[0028] During operation, the pressure controller continuously adjusts the pressure within the finger cuff to maintain a constant volume (i.e., the unloaded volume of the artery) in the finger, as measured by the cardiac reference sensor 30 via the optical transmitter and optical receiver of the inflatable finger cuff 28. The pressure applied by the pressure controller to continuously maintain the unloaded volume represents the blood pressure in the finger and is transmitted by the pressure controller to the cardiac reference sensor 30. The cardiac reference sensor 30 converts the pressure signal representing the blood pressure in the finger into hemodynamic data representing the patient's arterial pressure waveform, which is transmitted via, for example, I / O connector 14 (…). Figure 1 Transmitted to hemodynamic monitor 10 ( Figure 1 Therefore, the hemodynamic sensor 26 transmits sensor data representing essentially continuous beat-by-beat monitoring of the patient's arterial pressure.

[0029] Figure 4is a block diagram of a hemodynamic monitoring system 32 that determines a risk score representing a probability of a future hypotension event based on a set of transformed low blood pressure analysis parameters. As shown, Figure 4 The hemodynamic monitoring system 32 includes a hemodynamic monitor 10 and a hemodynamic sensor 34. The hemodynamic monitoring system 32 can be implemented within a patient care environment, such as an ICU, OR, or other patient care environment. As shown, Figure 4 The patient care environment can include a patient 36 and a medical caregiver 38 trained to utilize the hemodynamic monitoring system 32.

[0030] The hemodynamic monitor 10, as described above with reference to Figure 1 The hemodynamic monitor 10 can be, for example, an integrated hardware unit including a system processor 40, a system memory 42, a display 12, an analog-to-digital (ADC) converter 44, and a digital-to-analog (DAC) converter 46. In other examples, any one or more components of the hemodynamic monitor 10 and / or the described functionality can be distributed among multiple hardware units. For example, in some examples, the display 12 can be a separate display device located away from and operably coupled with the hemodynamic monitor 10. Generally, although illustrated and described in the example of Figure 4 as an integrated hardware unit, it will be appreciated that the hemodynamic monitor 10 can include any combination of devices and components electrically, communicatively, or otherwise operably connected to perform the functions attributed herein to the hemodynamic monitor 10.

[0031] As shown, Figure 4 The system memory 42 stores low blood pressure prediction software code 48. The low blood pressure prediction software code 48 includes a prediction weighting module 50, low blood pressure analysis parameters 52, and transformed low blood pressure analysis parameters 53. The display 12 provides a user interface 54 including control elements 56 that enable a user to interact with the hemodynamic monitor 10 and / or other components of the hemodynamic monitoring system 32. As shown, Figure 4 The user interface 54 also provides a sensory alert 58 to provide a medical caregiver with a warning of a predicted future hypotension event for the patient 36, as described further below.

[0032] The hemodynamic sensor 34 can be attached to the patient 36 to sense hemodynamic data representative of an arterial pressure waveform of the patient 36. The hemodynamic sensor 34 is operably connected to the hemodynamic monitor 10 (e.g., via a wired or wireless connection or both electrically and / or communicatively connected) to provide sensed hemodynamic data to the hemodynamic monitor 10. In some examples, the hemodynamic sensor 34 provides hemodynamic data representative of an arterial pressure waveform of the patient 36 to the hemodynamic monitor 10 as an analog signal that is converted to digital hemodynamic data representative of the arterial pressure waveform by the ADC 44. In other examples, the hemodynamic sensor 34 can provide sensed hemodynamic data to the hemodynamic monitor 10 in digital form, in which case the hemodynamic monitor 10 can not include or utilize the ADC 44. In other examples, the hemodynamic sensor 34 can provide hemodynamic data representative of an arterial pressure waveform of the patient 36 to the hemodynamic monitor 10 as an analog signal that is analyzed by the hemodynamic monitor 10 in its analog form.

[0033] The hemodynamic sensor 34 can be a non-invasive or minimally-invasive sensor that is attached to the patient 36. For example, the hemodynamic sensor 34 can take the form of a minimally-invasive hemodynamic sensor 16 Figure 2 ), a non-invasive hemodynamic sensor 26 Figure 3 ), or other minimally-invasive or non-invasive hemodynamic sensor. In some examples, the hemodynamic sensor 34 can be attached non-invasively at an extremity of the patient 36, such as a wrist, arm, finger, ankle, toe, or other extremity of the patient 36. Thus, the hemodynamic sensor 34 can take the form of a small, lightweight, and comfortable hemodynamic sensor that is suitable for being worn by the patient 36 for an extended period of time, such as for minutes or hours, to provide substantially continuous beat-to-beat monitoring of arterial pressure of the patient 36.

[0034] In certain examples, the hemodynamic sensor 34 can be configured to sense arterial pressure of the patient 36 in a minimally-invasive manner. For example, the hemodynamic sensor 34 can be attached to the patient 36 via a radial artery catheter that is inserted into an arm of the patient 36. In other examples, the hemodynamic sensor 34 can be attached to the patient 36 via a femoral artery catheter that is inserted into a leg of the patient 36. Such minimally-invasive techniques can similarly enable the hemodynamic sensor 34 to provide substantially continuous beat-to-beat monitoring of arterial pressure of the patient 36 for an extended period of time, such as for minutes or hours.

[0035] The system processor 40 is configured to execute hypotension prediction software code 48 that utilizes the hypotension analysis parameters 52 and the transformed hypotension analysis parameters 53 to implement a prediction weighting module 50 to produce a risk score representing a probability of a future hypotension event for the patient 36. Examples of the system processor 40 can 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

[0036] The system memory 42 can 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, a computer-readable storage medium can include a non-transitory medium. The term “non-transitory” can indicate that the storage medium is not embodied in a carrier wave or a propagated signal. In certain examples, a non-transitory storage medium can store data that can, over time, change (e.g., in RAM or cache memory). The system memory 42 can include both volatile and non-volatile computer-readable storage media. Examples of volatile computer-readable storage media can 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 computer-readable storage media can include forms of persistent memory, such as magnetic hard disks, optical disks, flash memory, or forms of electrically programmable memory (EPROM) or electrically erasable and programmable (EEPROM) memory.

[0037] The 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 in graphical form to a user. The user interface 54 can include graphical and / or physical control elements that enable user input to interact with other components of the hemodynamic monitor 10 and / or the hemodynamic monitoring system 32. In some examples, the user interface 54 can take the form of a graphical user interface (GUI) that is presented at graphical control elements presented at a touch-sensitive and / or presence-sensitive display screen of, for example, the display 12. In such examples, user input can be received in the form of gesture inputs, such as touch gestures, scroll gestures, zoom gestures, or other gesture inputs. In certain examples, the user interface 54 can 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 the hemodynamic monitoring system 32.

[0038] In operation, the hemodynamic sensor 34 senses hemodynamic data representative of an arterial pressure waveform of the patient 36. The hemodynamic sensor 34 provides the hemodynamic data (e.g., as analog sensor data) to the hemodynamic monitor 10. The ADC 44 converts the analog hemodynamic data to digital hemodynamic data representative of the arterial pressure waveform of the patient.

[0039] The system processor 40 executes the hypotension prediction software code 48 to determine a risk score representative of a probability of a future hypotension event of the patient 36 using the received hemodynamic data. For example, the system processor 40 can execute the hypotension prediction software code 48 to obtain a plurality of hypotension analysis parameters 52 using the received hemodynamic data. The hypotension analysis parameters 52 can include one or more vital sign parameters characterizing vital sign data of the patient 36, as well as differential and combined parameters derived from the one or more vital sign parameters, as further described below. As further described below, the system processor 40 also executes the hypotension prediction software code 48 to generate a set of transformed hypotension analysis parameters 53 based on an adjusted MAP threshold for hypotension.

[0040] The prediction weighting module 50 of the hypotension prediction software code 48 determines a risk score corresponding to a probability of a future hypotension event of the patient 36 based on a weighted combination of the transformed hypotension analysis parameters 53. That is, the prediction weighting module 50 applies a plurality of risk coefficients stored in the system memory 42 to the transformed hypotension analysis parameters 53 to produce a weighted combination that results in the risk score.

[0041] The risk coefficients can be determined via a training operation (e.g., offline training) using machine learning or other techniques to minimize a cost function representative of an error of the risk score from true values of a training subset (e.g., a set of data from a plurality of patients) defining hypotension according to a standard MAP threshold for hypotension. That is, the risk coefficients utilized by the prediction weighting module 50 can be selected via the training operation to minimize an error of the predicted risk score determined by the hypotension prediction software code 48 as a prediction of a future hypotension event using the hypotension analysis parameters 52. The error of the predicted risk score predicting a future hypotension event can be evaluated with respect to a positive training data subset and a negative training data subset defining the occurrence of hypotension with respect to a standard (e.g., defined) MAP threshold, such as 65 mmHg or other pressure threshold.

[0042] As described herein, the hemodynamic monitor 10 can receive an adjusted MAP threshold for hypotension, such as a user-defined MAP threshold, via the control element 56 of the user interface 54. The adjusted MAP threshold can represent a deviation from a standard MAP threshold by which a risk coefficient used by the prediction weighting module 50 is determined. The adjusted MAP threshold provided by, for example, the medical professional 38 can take the form of an absolute pressure (e.g., MAP threshold), a deviation value (e.g., a deviation from a standard MAP threshold), or other indication of an adjusted MAP threshold.

[0043] The hypotension prediction software code 48 generates a set of transformed hypotension analysis parameters 53 using the hypotension analysis parameters 52 in response to receiving the adjusted MAP threshold. As further described below, each of the transformed hypotension analysis parameters 53 can be a function of a corresponding hypotension analysis parameter 52 (i.e., a standard MAP threshold for hypotension), a mean of the corresponding hypotension analysis parameter 52 at the standard MAP threshold for hypotension, a standard deviation of the corresponding hypotension analysis parameter 52 at the standard MAP threshold, a mean of the corresponding hypotension analysis parameter 52 at the adjusted MAP threshold for hypotension, and a standard deviation of the corresponding hypotension analysis parameter 52 at the adjusted MAP threshold for hypotension.

[0044] The system processor 40 executes the hypotension prediction software code 48 to determine a prediction risk score as a weighted combination of the transformed hypotension analysis parameters 53 using the risk coefficient determined based on the standard MAP threshold. Thus, the hypotension prediction software code 48 dynamically adapts to use the transformed hypotension analysis parameters 53 to determine the risk score without requiring retraining or other modification of the prediction model to identify new coefficients.

[0045] The system processor 40 also executes hypotension prediction software code 48 to invoke a sensory alert 58 via the user interface 54 in response to determining that the risk score satisfies a predetermined risk criterion, as further described below. For example, the hypotension prediction software code 48 can invoke the sensory alert 58 to warn of a predicted upcoming hypotension event, e.g., in one to five minutes in the future, or up to about thirty minutes in the future. The sensory alert 58 can be implemented as one or more of a visual alert, an audible alert, a haptic alert, or other type of sensory alert. For example, the sensory alert 58 can be invoked as any combination of a flashing and / or colored graphic shown by the user interface 54 on the display 12, a risk score display via the user interface 54 on the display 12, a warning sound such as a siren or repeating tone, and a haptic alert configured to cause the hemodynamic monitor 10 to vibrate or otherwise deliver a physical impulse perceivable by the medical professional 38 or other user.

[0046] Accordingly, the hemodynamic monitor 10 provides a warning to the medical professional of a predicted upcoming hypotension event for the patient 36, enabling timely and effective intervention to prevent the predicted upcoming hypotension event. Moreover, rather than retraining the predictive risk model to determine new risk coefficients based on an adjusted MAP threshold, the hemodynamic monitor 10 implementing the techniques of the present disclosure generates a transformed set of hypotension analysis parameters (or features) for determining a risk score via application of unmodified risk coefficients. Accordingly, the hemodynamic monitor 10 enables real-time updating of the MAP threshold defining hypotension by the medical professional, thereby increasing the usability of the hemodynamic monitor 10 via dynamic adaptation to, e.g., user-defined changes that can be based on the training and expertise of the attending medical professional to predict upcoming hypotension events for the patient 36.

[0047] Figure 5 is a graph illustrating an example trace of an arterial pressure waveform 60 corresponding to hemodynamic data sensed by the hemodynamic sensor 34 and received by the hemodynamic monitor 10. As Figure 5 shown, the hemodynamic waveform 60 (e.g., via digital hemodynamic data representation) can include various markers that predict an upcoming hypotension event for the patient 36. Figure 5 Example markers 62, 64, 66, and 68 are illustrated that respectively correspond to a start of a heartbeat for the patient 36 (marker 62), a maximum systolic pressure marking an end of systolic rise (marker 64), a presence of a dicrotic notch marking an end of systolic decay (marker 66), and a diastole of the heartbeat (marker 68). In Figure 5 An example slope "m" of the adjusted arterial pressure waveform 60 is also shown in the graph, but it should be understood that the slope "m" represents only a plurality of slopes that can be determined at a plurality of locations along the arterial pressure waveform 60.

[0048] Additional markers that predict future hypotension of the patient 36 can be extracted from the hemodynamic waveform 60 by the hypotension prediction software code 48 based on behavior of the hemodynamic waveform 60 in various intervals, such as in the interval from the maximum systolic pressure at marker 64 to the diastolic period at marker 66, and the interval from the beginning of the heartbeat at marker 62 to the diastolic period at marker 66. The behavior of the arterial pressure waveform 60 during the following intervals (1) systolic rise 62-64, 2) systolic decay 64-66, 3) systolic period 62-66, 4) diastolic period 66-68, 5) interval 64-68, and 6) heartbeat interval 62-68 can be determined by the hypotension prediction software code 48 by determining the area under the curve of the hemodynamic waveform 60 and the standard deviation of the hemodynamic waveform 60 in each of the intervals 1-6. The respective areas and standard deviations determined for the intervals 1-6 can be used as additional markers that predict future hypotension of the patient 36.

[0049] Figure 6 is a flowchart illustrating example operations of generating a transformed set of hypotension analysis parameters and determining a risk score representing a probability of a future hypotension event using the transformed set of hypotension analysis parameters. For purposes of clarity and ease of discussion, the example operations are described below in the context of the hemodynamic monitoring system 32 of Figure 4 The adjusted MAP threshold for hypotension is received by the hemodynamic monitor 10 (step 70). For example, the hemodynamic monitor 10 can receive the adjusted MAP threshold for hypotension provided by, for example, the healthcare professional 38 via the control element 56 of the user interface 54. The hemodynamic monitor 10 receives sensed hemodynamic data representing an arterial pressure waveform of the patient 36 (step 72). For example, the hemodynamic monitor 10 can receive an analog hemodynamic sensor signal representing an arterial pressure waveform of the patient 36 from the hemodynamic sensor 34.

[0050] The hemodynamic monitor 10 performs a waveform analysis of the hemodynamic data to determine a plurality of hypotension analysis parameters that predict future hypotension events of the patient 36 (step 74). For example, the hemodynamic monitor 10 can execute the hypotension prediction software code 48 to perform the waveform analysis of the hemodynamic data to obtain the hypotension analysis parameters 52 that predict future hypotension of the patient 36. The hypotension analysis parameters 52 can include one or more of vital sign parameters characterizing vital sign data of the patient 36, differential parameters derived from the vital sign parameters, and combination parameters representing combinations of one or more of the vital sign parameters and the differential parameters.

[0051] The hemodynamic monitor 10 performs a waveform analysis of the hemodynamic data to determine a plurality of hypotension analysis parameters that predict future hypotension events of the patient 36 (step 74). For example, the hemodynamic monitor 10 can execute the hypotension prediction software code 48 to perform the waveform analysis of the hemodynamic data to obtain the hypotension analysis parameters 52 that predict future hypotension of the patient 36. The hypotension analysis parameters 52 can include one or more of vital sign parameters characterizing vital sign data of the patient 36, differential parameters derived from the vital sign parameters, and combination parameters representing combinations of one or more of the vital sign parameters and the differential parameters.

[0052] Vital sign parameters characterizing the vital sign data can include, for example, stroke volume, heart rate, respiration, myocardial contractility, mean arterial pressure, baroreflex sensitivity metric, hemodynamic complexity measure, frequency domain hemodynamic features, or other vital sign parameters. Baroreflex sensitivity metrics quantify the relationship between complementary physiological processes. For example, a decrease in blood pressure in a healthy patient is typically compensated by an increase in heart rate and / or an increase in peripheral resistance. Baroreflex sensitivity metrics, which can be included in one or more vital sign parameters characterizing the vital sign data, correspond to the degree to which the patient 36 responds appropriately to normal physiological changes.

[0053] Hemodynamic complexity measures quantify the amount of regularity in cardiac measurements over time, as well as entropy, e.g., the unpredictability of fluctuations in cardiac measurements over time. For example, unpredictable cardiac fluctuations are a normal phenomenon associated with health. Very low entropy, i.e., a high degree of regularity in cardiac measurements over time and the substantial absence of unpredictable fluctuations, can be an important warning sign of an impending hypotension event. Frequency domain hemodynamic features quantify various measures of cardiac performance as a function of frequency rather than time.

[0054] The hypotension prediction software code 48 can also determine derivative parameters based on one or more vital sign parameters characterizing the vital sign data of the patient 36. The hypotension prediction software code 48 can derive derivative parameters from one or more vital sign parameters by determining the change in the one or more vital sign parameters with respect to time, with respect to frequency, or with respect to other parameters from among the one or more vital sign parameters. Thus, each of the one or more vital sign parameters can yield one, two, or several derivative parameters included in the hypotension analysis parameters 52.

[0055] For example, a derivative parameter, stroke volume variation (SVV), can be derived based on the change in the parameter stroke volume (SV) as a function of time and / or as a function of sampling frequency. Similarly, a change in mean arterial pressure (AMAP) can be derived as a derivative parameter with respect to time and / or sampling frequency. As another example, a change in MAP with respect to time can be derived by subtracting the average value of MAP over the past five minutes, ten minutes, or other duration, from the current value of MAP.

[0056] The hypotension prediction software code 48 can use one or more of the vital sign parameters and the derived differential parameters to generate combined parameters included in the hypotension analysis parameters 52. For example, the one or more vital sign parameters and the differential parameters can be used to generate the combined parameters by generating power combinations of a subset of the one or more vital sign parameters and the differential parameters. For example, each combined parameter can be generated as a power combination of three parameters, which can be randomly or purposefully selected from among the one or more vital sign parameters and / or the differential parameters characterizing the vital sign data. Each of the three parameters selected from among the one or more vital sign parameters and / or the differential parameters can be raised to an exponential power, and can be multiplied or added to the other two parameters similarly raised to an exponential power. The exponential power to which each of the three parameters selected from among the one or more vital sign parameters and / or the differential parameters is raised can be, but need not be, the same exponential power.

[0057] In some examples, the generation of the combined parameters can be performed using a predetermined and limited range of integer exponents. For example, the exponents used to generate the combined parameters can be integer exponents selected from among negative two, negative one, zero, one, and two (-2, -1, 0, 1, 2). Thus, in some examples, each combined parameter can be expressed according to the following equation:

[0058]

[0059] where Y is one of the one or more vital sign parameters or one of the differential parameters, n is any integer greater than 2, and each of a, b, and c can be any one of -2, -1, 0, 1, and 2. In some examples, Equation 1 above can be applied to substantially all possible power combinations of the one or more vital sign parameters, the differential parameters, and the one or more vital sign parameters, the differential parameters are subject to the predetermined constraints described above (i.e., the value of n is any integer greater than 2, and each of a, b, and c is selected from the group consisting of -2, -1, 0, 1, and 2).

[0060] The hypotension analysis parameters 52 include the one or more vital sign parameters, the differential parameters, and the combined parameters characterizing the vital sign data. Examples of the hypotension analysis parameters 52 can include, but are not limited to, various combinations of one or more of the following parameters v1, v2,... v 19

[0061] v1 = CWI, cardiac work index indexed to body surface area of the patient 36;

[0062] v2 = MAPavg, mean average arterial pressure;

[0063] ​v3 = AMAPavg, change in average mean arterial pressure MAPavg when compared to initial state;

[0064] v4 = avgSysDec, average pressure at the decay portion of the systole;

[0065] v5 = ASys, change in systolic pressure when compared to initial value;

[0066] v6 = ppAreaNor, normalized area under the adjusted arterial pressure waveform;

[0067] v7 = biasDia, bias of diastolic slope;

[0068] v8 = CW, cardiac work;

[0069] v9 = mapDnlocArea, area under the adjusted arterial pressure waveform between the first instance of MAP and the dicrotic notch;

[0070] v 10 = SWcomb, stroke work;

[0071] v 11 = ppArea, area under the adjusted arterial pressure waveform;

[0072] v 12 = decAreaNor, normalized area of the decay phase;

[0073] v 13 = slopeSys, slope of the systole;

[0074] v 14 = Cwk, Winkessel compliance;

[0075] v 15 = sys_rise_area_nor, normalized area under the systolic rise phase;

[0076] v 16 = pulsepres, pulse pressure;

[0077] v 17 = avg_sys, average pressure of the systole;

[0078] v 18 = dpdt2, maximum of the second derivative of the adjusted arterial pressure waveform; and

[0079] v 19 = dpdt, maximum of the first derivative of the adjusted arterial pressure waveform.

[0080] Accordingly, the hypotension prediction software code 48 determines the hypotension analysis parameters 52 by identifying one or more vital sign parameters characterizing the vital sign data based on the hemodynamic data, obtaining a differential parameter based on the one or more vital sign parameters, and generating a combined parameter using the one or more vital sign parameters and the differential parameter.

[0081] The hemodynamic monitor 10 generates a set of transformed hypotension analysis parameters (step 76). For example, the hemodynamic monitor 10 can execute the hypotension prediction software code 48 to determine the transformed hypotension analysis parameters 53. Each of the transformed hypotension analysis parameters 53 can be a function of a corresponding one of the hypotension analysis parameters 52 at a standard MAP threshold for hypotension (e.g., 65 mmHg or other defined pressure threshold), a mean of the corresponding one of the hypotension analysis parameters 52 at the standard MAP threshold, a standard deviation of the corresponding one of the hypotension analysis parameters 52 at the standard MAP threshold, a mean of the corresponding one of the hypotension analysis parameters 52 at an adjusted MAP threshold for hypotension, and a standard deviation of the corresponding one of the hypotension analysis parameters 52 at the adjusted MAP threshold for hypotension.

[0082] In some examples, the hemodynamic monitor 10 executes the hypotension prediction software code 48 to determine the transformed hypotension analysis parameters 53 according to the following equations:

[0083]

[0084] where:

[0085] v kθ is one of the transformed hypotension analysis parameters 53 associated with a corresponding one of the hypotension analysis parameters 52;

[0086] σ kθ is a standard deviation of the corresponding one of the hypotension analysis parameters 52 at an adjusted MAP threshold for hypotension;

[0087] σ k is a standard deviation of the corresponding one of the hypotension analysis parameters 52 at a standard MAP threshold for hypotension;

[0088] v k is the corresponding one of the hypotension analysis parameters 52 at the standard MAP threshold for hypotension;

[0089] μ kθ is a mean of the corresponding one of the hypotension analysis parameters 52 at the adjusted MAP threshold for hypotension; and

[0090] μ k is an average of the corresponding hypotension analysis parameter in the hypotension analysis parameters 52 at the standard MAP threshold of hypotension.

[0091] Accordingly, the hemodynamic monitor 10 can execute the hypotension prediction software code 48 to determine transformed hypotension analysis parameters 53 according to Equation 2 above, which can be represented as the set v 1θ ,v 2θ ,…v 19θ . Each of the transformed hypotension analysis parameters 53 (i.e., each of the set v 1θ ,v 2θ ,…v 19θ ) represents one transformed hypotension analysis parameter of the hypotension analysis parameters 52 that is adapted to a new MAP threshold of hypotension.

[0092] A risk score representing a probability of a future hypotension event for the patient 36 is determined based on the transformed hypotension analysis parameters 53 (step 78). For example, the system processor 40 can execute the hypotension prediction software code 48 to cause the prediction weighting module 50 to determine the risk score as a weighted combination of the transformed hypotension analysis parameters 53. The prediction weighting module 50 can determine the weighted combination of the transformed hypotension analysis parameters 53 by applying a plurality of risk coefficients to the transformed hypotension analysis parameters 53, including the vital sign parameters characterizing the vital sign data of the patient 36, the differential parameters derived from the vital sign parameters, and the combination parameters.

[0093] The plurality of risk coefficients applied by the prediction weighting module 50 can be determined (e.g., via offline training) relative to the standard MAP threshold of hypotension. Accordingly, because the hypotension prediction software code 48 determines the risk score based on the transformed hypotension analysis parameters 53 derived from the hemodynamic data sensed by the hemodynamic sensors attached to the patient 36, it should be noted that the hypotension prediction software code 48 determines the risk score for the patient 36 without needing direct comparison to hypotension in other patients and without needing direct reference to a hypotension database that can store information about hypotension in patients other than the patient 36.

[0094] In some examples, the prediction weighting module 50 determines the risk score representing a probability of a future hypotension event for the patient 36 according to the following equation:

[0095] R = 1 / (1 + e -A ) (Equation 3)

[0096] where R is the risk score, and A is represented as:

[0097]

[0098] wherein

[0099] v 1θ ,v 2θ ,…v 19θ is a transformed set of hypotension analysis parameters 53;

[0100] v 1θ = transformed CWI, cardiac work index indexed by body surface area of patient 36 included in hypotension analysis parameters;

[0101] v 2θ = transformed MAPavg, mean average arterial pressure;

[0102] v 3θ = transformed AMAPavg, change in mean average arterial pressure MAPavg when compared to initial state;

[0103] v 4θ = transformed avgSysDec, average pressure at decay portion of systole;

[0104] v 5θ = transformed ASys, change in systolic pressure when compared to initial value;

[0105] v 6θ = transformed ppAreaNor, normalized area under adjusted arterial pressure waveform;

[0106] v 7θ = transformed biasDia, bias of diastolic slope;

[0107] v 8θ = transformed CW, cardiac work;

[0108] v 9θ = transformed mapDnlocArea area under adjusted arterial pressure waveform between first instance of MAP and dicrotic notch;

[0109] v 10θ = transformed SWcomb, stroke work;

[0110] v 11θ = transformed ppArea, area under adjusted arterial pressure waveform;

[0111] v 12θ = transformed decAreaNor, normalized area of decay phase;

[0112] v 13θ= transformed slopeSys, slope of systole

[0113] v 14θ = transformed Cwk, Winkessel compliance

[0114] v 15θ = transformed sys_rise_area_nor, normalized area under systolic rise

[0115] v 16θ = transformed pulsepres, pulse pressure

[0116] v 17θ = transformed avg_sys, average pressure of systole

[0117] v 18θ = transformed dpdt2, maximum of second derivative of adjusted arterial pressure waveform

[0118] v 19θ = transformed dpdt, maximum of first derivative of adjusted arterial pressure waveform, and

[0119] c0, c1,..., c 11 are risk coefficients determined with respect to a standard MAP threshold for hypotension.

[0120] In some examples, the risk score R can be expressed as a score as represented by Equation 3 above. In other examples, the risk score can be converted to a percentage risk score between 0% and 100%.

[0121] The hemodynamic monitor 10 invokes a sensory alarm in response to the risk score satisfying a predetermined risk criterion (step 80). For example, the hypotension prediction software code 48 can invoke the sensory alarm 58 of the user interface 54 in response to determining that the risk score R determined according to Equation 3 above satisfies a predetermined risk criterion. In some examples, the output of the hypotension prediction software code 48 can be processed using the DAC 46 to convert the digital signal to an analog signal for presentation at the display 12 via the user interface 54.

[0122] The predetermined risk criteria can be based on the value of a risk score, the trend of the risk score over a period of time, or both. For example, where the risk score is expressed as a percentage between 0 and 100, hypotension prediction software code 48 can (e.g., immediately) invoke a sensory alarm 58 in response to determining that the risk score exceeds a first predetermined threshold (such as 85%). In some examples, hypotension prediction software code 48 can invoke a sensory alarm 58 in response to determining that the risk score meets a second predetermined threshold throughout a first predetermined time period. In such examples, the second predetermined threshold may be lower than the first predetermined threshold.

[0123] Therefore, hypotension prediction software code 48 can, for example, immediately invoke sensory alarm 58 in response to determining that a risk score exceeds a first predetermined threshold (e.g., 85%). Hypotension prediction software code 48 can also invoke sensory alarm 58 in response to determining that a risk score exceeds a second predetermined threshold (e.g., 80%) less than the first predetermined threshold within a first predetermined time period (e.g., ten to thirty seconds), during which the risk score is continuously greater than the second predetermined threshold (80%) and less than the first predetermined threshold (e.g., 85%). In some examples, hypotension prediction software code 48 can invoke sensory alarm 58 in response to determining that a risk score is greater than a third predetermined threshold less than the second predetermined threshold within a second predetermined time period (e.g., one minute or more). In other examples, hypotension prediction software code 48 can invoke sensory alarm 58 in response to determining that a risk score exceeds a fourth predetermined threshold (e.g., 75%) a certain number of times (e.g., two, three, or other times) within a third predetermined time period (e.g., one minute, two minutes, or other time periods).

[0124] Although not in Figure 6 The exemplary operation is illustrated, but in some examples, the hemodynamic monitor 10 may use hypotension prediction software code 48 to identify the most probable cause of a predicted future hypotension event in patient 36. For example, based on the identified markers, hypotension prediction software code 48 may identify poor vascular tone, low blood volume, reduced cardiac contractility, or other most probable causes of a predicted future hypotension event in patient 36.

[0125] In some examples, the hemodynamic monitor 10 can recommend medical interventions to prevent predicted future hypotensive events in patient 36, such as by identifying recommended medical interventions corresponding to the most probable causes of the predicted future hypotensive events in patient 36. For example, regarding the most probable cause of poor vascular tone, the hemodynamic monitor 10 can recommend a medical intervention of administering a vasoconstrictor. Regarding the most probable cause of hypovolemia, the hemodynamic monitor 10 can recommend a medical intervention of administering saline or whole blood, for example.

[0126] Accordingly, the hemodynamic monitor 10 implementing the techniques of the present disclosure provides a risk score that predicts a future hypotension event for the patient 36, thereby enabling timely and effective intervention to prevent the hypotension event before the patient 36 enters a hypotensive state. Moreover, by enabling adjustment to the defined hypotension threshold without requiring retraining of the predictive risk model, the techniques described herein increase the usability of the hemodynamic monitoring system 32 to accommodate, for example, the training and experience of medical personnel.

[0127] While the application has been described with reference to example embodiments(s), it will be understood by those skilled in the art that various changes can be made and equivalents can be substituted for elements thereof without departing from the scope of the application. In addition, many modifications can be made to adapt a particular situation or material to the teachings of the application without departing from the central scope thereof. Therefore, it is intended that the application not be limited to the particular embodiment(s) disclosed, but that the application will include all embodiments falling within the scope of the appended claims.

Claims

1. A method for monitoring arterial pressure of a patient and providing a warning to medical personnel of a predicted future hypotensive event of the patient, the method comprising: receiving, by a hemodynamic monitor, via a user interface of the hemodynamic monitor, an adjusted mean arterial pressure threshold for hypotension, adjusted MAP threshold; receiving, by the hemodynamic monitor, sensed hemodynamic data representative of an arterial pressure waveform of the patient; performing, by the hemodynamic monitor, a waveform analysis of the hemodynamic data to determine a plurality of hypotension analysis parameters predictive of a future hypotensive event of the patient; generating, by the hemodynamic monitor, a set of transformed hypotension analysis parameters, each transformed hypotension analysis parameter being a function of a corresponding one of the plurality of hypotension analysis parameters at a standard mean arterial pressure threshold for hypotension, standard MAP threshold, a mean of the corresponding one of the plurality of hypotension analysis parameters at the standard MAP threshold, a standard deviation of the corresponding one of the plurality of hypotension analysis parameters at the standard MAP threshold, a mean of the corresponding one of the plurality of hypotension analysis parameters at the adjusted MAP threshold for hypotension, and a standard deviation of the corresponding one of the plurality of hypotension analysis parameters at the adjusted MAP threshold for hypotension; determining, by the hemodynamic monitor, a risk score representative of a probability of the future hypotensive event of the patient based on the set of transformed hypotension analysis parameters; and invoking, by the hemodynamic monitor, a sensory alarm to produce a sensory signal in response to the risk score satisfying a predetermined risk criterion.

2. The method of claim 1, wherein generating the set of transformed hypotension analysis parameters comprises transforming each of the plurality of hypotension analysis parameters according to the following equation: wherein v kθ is one of the set of transformed hypotension analysis parameters associated with the corresponding one of the plurality of hypotension analysis parameters; where σ kθ is the standard deviation of the corresponding one of the plurality of hypotension analysis parameters at the adjusted MAP threshold of hypotension; where σ k is a standard deviation of the corresponding one of the plurality of hypotension analysis parameters at a standard MAP threshold of hypotension; wherein v k is the corresponding one of the plurality of hypotension analysis parameters at a standard MAP threshold of hypotension; where μ kθ is the average of the corresponding one of the plurality of hypotension analysis parameters at the adjusted MAP threshold of hypotension; and where μ k is the mean of the corresponding one of the plurality of hypotension analysis parameters at the standard MAP threshold of hypotension.

3. The method of claim 1, wherein determining the risk score representative of a probability of the future hypotensive event of the patient comprises applying a plurality of risk coefficients to the set of transformed hypotension analysis parameters to determine the risk score.

4. The method of claim 3, wherein the plurality of risk coefficients are determined based on the standard MAP threshold.

5. The method of claim 1, wherein performing the waveform analysis of the hemodynamic data to determine the plurality of hypotension analysis parameters predictive of the future hypotensive event of the patient comprises: performing the waveform analysis of the hemodynamic data to obtain vital sign parameters from the hemodynamic data; deriving differential parameters based on one or more of the vital sign parameters; and generating a combined parameter using one or more of the vital sign parameters and / or one or more of the differential parameters; wherein the plurality of hypotension analysis parameters comprises one or more of the vital sign parameters, the differential parameters, and the combined parameter.

6. The method of claim 5, wherein the waveform analysis of the hemodynamic data comprises: wherein the vital sign parameters include one or more of stroke volume, heart rate, respiration, and cardiac contractility.

7. The method of claim 5, wherein deriving the differential parameter based on one or more of the vital sign parameters includes deriving the differential parameter to represent a change in the one or more of the vital sign parameters over time, over frequency, or over other vital sign parameters.

8. The method of claim 5, wherein generating the combined parameter includes generating the combined parameter as a combination of vital sign parameters, a combination of differential parameters, or a combination of at least one vital sign parameter and at least one differential parameter.

9. A system for monitoring arterial pressure of a patient and providing a warning to medical personnel of a predicted future hypotensive event of the patient, the system comprising: a hemodynamic sensor producing hemodynamic data representing an arterial pressure waveform of the patient; a system memory storing hypotension prediction software code including a prediction weighting module; a user interface including a sensory alarm providing a sensory signal to warn the medical personnel of a predicted future hypotensive event prior to the patient entering a hypotensive state, and a control element configured to enable a user to input an adjusted MAP threshold for hypotension; and a hardware processor configured to execute the hypotension prediction software code to: perform waveform analysis of the hemodynamic data to determine a plurality of hypotension analysis parameters that predict a future hypotensive event of the patient; generate a set of transformed hypotension analysis parameters, each transformed hypotension analysis parameter being a function of a corresponding one of the plurality of hypotension analysis parameters at a standard mean arterial pressure threshold for hypotension, a mean value of the corresponding one of the plurality of hypotension analysis parameters at the standard MAP threshold for hypotension, a standard deviation of the corresponding one of the plurality of hypotension analysis parameters at the standard MAP threshold, a mean value of the corresponding one of the plurality of hypotension analysis parameters at an adjusted MAP threshold for hypotension, and a standard deviation of the corresponding one of the plurality of hypotension analysis parameters at the adjusted MAP threshold for hypotension; determine a risk score representing a probability of the future hypotensive event of the patient based on the set of transformed hypotension analysis parameters; and invoke the sensory alarm of the user interface in response to the risk score satisfying a predetermined risk criterion.

10. The system of claim 9, wherein the hardware processor is configured to generate the set of transformed hypotension analysis parameters according to the following equation: wherein v kθ is one of the set of transformed hypotension analysis parameters associated with the corresponding one of the plurality of hypotension analysis parameters; where σ kθ is the standard deviation of the corresponding one of the plurality of hypotension analysis parameters at the adjusted MAP threshold of hypotension; where σ k is a standard deviation of the corresponding one of the plurality of hypotension analysis parameters at a standard MAP threshold of hypotension; wherein v k is the corresponding one of the plurality of hypotension analysis parameters at a standard MAP threshold of hypotension; where μ kθ is the average of the corresponding one of the plurality of hypotension analysis parameters at the adjusted MAP threshold of hypotension; and where μ k is the average of the corresponding one of the plurality of hypotension analysis parameters at the standard MAP threshold of hypotension.

11. The system of claim 9, wherein the hardware processor is configured to execute the hypotension prediction software code to determine the risk score representing a probability of the future hypotension event of the patient by applying a plurality of risk coefficients to the plurality of hypotension analysis parameters using the prediction weighting module to determine the risk score.

12. The system of claim 11, wherein the plurality of risk coefficients are determined based on the standard MAP threshold.

13. The system of claim 9, wherein the hardware processor is configured to execute the hypotension prediction software code to determine the plurality of hypotension analysis parameters that predict the future hypotension event of the patient by executing the hypotension prediction software code to: perform the waveform analysis of the hemodynamic data to obtain vital sign parameters from the adjusted hemodynamic data; derive differential parameters based on one or more of the vital sign parameters; and generate a combined parameter using one or more of the vital sign parameters and / or one or more of the differential parameters; wherein the plurality of hypotension analysis parameters include one or more of the vital sign parameters, the differential parameters, and the combined parameter.

14. The system of claim 13, wherein the vital sign parameters include one or more of stroke volume, heart rate, respiration, and cardiac contractility.

15. The system of claim 14, wherein the hardware processor is configured to execute the hypotension prediction software code to derive the differential parameters based on one or more of the vital sign parameters by deriving the differential parameters to represent a change in one or more of the vital sign parameters over time, over frequency, or over other vital sign parameters.

16. The system of claim 14, wherein the hardware processor is configured to execute the hypotension prediction software code to generate the combined parameter by generating the combined parameter as a combination of vital sign parameters, a combination of differential parameters, or a combination of at least one vital sign parameter and at least one differential parameter.

17. The system of claim 9, wherein the hemodynamic sensor is a non-invasive hemodynamic sensor that is attachable to a limb of the patient.

18. The system of claim 9, wherein the hemodynamic sensor is a minimally-invasive arterial catheter-based hemodynamic sensor.

19. The system of claim 9, wherein the hemodynamic sensor produces the hemodynamic data as an analog hemodynamic sensor signal representing an arterial pressure waveform of the patient.

20. The system of claim 19, further comprising: an analog-to-digital converter that converts the analog hemodynamic sensor signal to digital hemodynamic data representing an arterial pressure waveform of the patient.

Images