Hemodynamic monitor with nociception prediction and detection

Abstract

A hemodynamic monitor with nociception prediction and detection is disclosed. A hemodynamic monitoring system monitors arterial pressure of a patient and provides a warning of a nociceptive event of the patient to medical personnel. The system includes a hemodynamic sensor that generates hemodynamic data representative of a waveform of arterial pressure of the patient. A hardware processor in the system is configured to execute nociception detection software code to perform waveform analysis of the hemodynamic data to determine a plurality of signal measurements. Detection input features are extracted by the processor from the plurality of signal measurements indicative of a nociceptive event of the patient. A nociception score representative of a probability of the nociceptive event of the patient is determined from the detection input features. A sensory alarm is invoked in response to the nociception score satisfying a predetermined detection criterion.

Description

Background Technology

[0001] This disclosure generally relates to hemodynamic monitoring, and more particularly to the use of monitored hemodynamic data to detect and predict a patient’s perception of injury.

[0002] Nociception is the process by which nerve endings called nociceptors detect harmful stimuli and send signals to the central nervous system that are interpreted as pain. Nociception can trigger an automatic response before or during consciousness, so unconscious patients may experience pain during surgery. To prevent patients from waking up from surgery due to pain, medical staff administer analgesics before and / or during surgery. However, because pain thresholds and tolerance vary from patient to patient, and patients are unable to communicate verbally or provide feedback signals while unconscious, it can be difficult to determine the appropriate dosage of analgesics. Administering too little analgesia during surgery can cause patients to wake up in pain after surgery. Administering too much analgesia during surgery can cause nausea, drowsiness, impaired cognitive function, and functional impairment.

[0003] Given the negative consequences of administering too little or too much analgesic to patients, a solution is needed that enables healthcare professionals to detect or predict harm experienced by unconscious patients during surgery. Accurate detection or prediction of a patient's harm during surgery can help healthcare professionals determine the appropriate analgesic dosage to administer, ensuring the patient does not wake up postoperatively due to significant pain and avoiding the administration of excessive analgesics. Summary of the Invention

[0004] In one example, a method for monitoring a patient's arterial pressure and providing alerts to healthcare professionals regarding the patient's current or predicted future pain experiences includes receiving sensed hemodynamic data representing the patient's arterial pressure waveform via a hemodynamic monitor. The method also includes waveform analysis of the sensed hemodynamic data by the hemodynamic monitor to calculate multiple signal measurements of the sensed hemodynamic data. The hemodynamic monitor extracts input features from the multiple signal measurements indicating current pain events and predicting future pain events in the patient. A pain score is determined by the hemodynamic monitor based on the input features. The pain score represents the probability of the patient's current pain event and / or the probability of future pain events in the patient. In response to a predetermined level criterion met by the pain score, the hemodynamic monitor invokes a sensory alarm to generate a sensory signal.

[0005] In another example, a system for monitoring a patient's arterial pressure and providing alerts to healthcare personnel regarding the patient's perceived injury includes a hemodynamic sensor that generates hemodynamic data representing the patient's arterial pressure waveform. The system also includes system memory storing injury detection software code and a user interface with a sensory alarm that provides a sensory signal to alert healthcare personnel to an injury event. A hardware processor in the system is configured to execute the injury detection software code to perform waveform analysis of the hemodynamic data to determine multiple signal measurements. The hardware processor is also configured to execute injury detection software code to extract detection input features from the multiple signal measurements indicating an injury event in the patient. The hardware processor is further configured to execute injury detection software code to determine an injury score, representing the probability of the injury event in the patient, based on the detection input features. The processor is configured to invoke the sensory alarm in the user interface in response to the injury score meeting predetermined detection criteria. Attached Figure Description

[0006] Figure 1 This is a perspective view of an example hemodynamic monitor that analyzes a patient's arterial pressure and provides healthcare professionals with a risk score and warnings for nociceptive events.

[0007] Figure 2 This is a perspective view of an example minimally invasive pressure sensor used to sense hemodynamic data representing a patient's arterial pressure.

[0008] Figure 3 This is a perspective view of an example non-invasive sensor used to sense hemodynamic data representing a patient's arterial pressure.

[0009] Figure 4 This is a block diagram illustrating an example hemodynamic monitoring system. The example hemodynamic monitoring system determines a risk score based on a set of input features derived from signal measurements of the patient's arterial pressure waveform, representing the probability of the patient's current nociceptive events, future nociceptive events, current hemodynamic drug administration events, future hemodynamic drug administration events, and / or stable periods.

[0010] Figure 5 It is a graph of clinical datasets and clinical annotations used for data mining and machine training of hemodynamic monitoring systems.

[0011] Figure 6 This is an explanation from Figure 5 The clinical dataset shows systolic blood pressure and heart rate over time, and plots of nociceptive events and administration of analgesics are presented.

[0012] Figure 7 This is an explanation from Figure 5 The clinical dataset shows the systolic blood pressure and heart rate over time, and plots of the stable period are presented.

[0013] Figure 8 This is an explanation from Figure 5 The clinical dataset shows systolic blood pressure and heart rate over time, and plots of hemodynamic drug administration events and administration of vasopressors are presented.

[0014] Figure 9 This is a flowchart for extracting a set of input features derived from the waveform features of a patient's arterial pressure waveform to train a machine learning model for a hemodynamic monitoring system.

[0015] Figure 10 It is a graph illustrating an example trajectory of the arterial pressure waveform, which includes example markers corresponding to signal measurements used to extract input features for determining a patient's risk score. Detailed Implementation

[0016] As described in this article, the hemodynamic monitoring system implements a predictive model that generates risk scores representing the probability of a patient experiencing a current nociceptive event, the probability of a patient experiencing a future nociceptive event, and the probability of a patient experiencing a stable period. The predictive model of the hemodynamic monitoring system can also optionally generate a risk score representing the probability that a patient is experiencing the effects of a previously administered hemodynamic drug (hereinafter referred to as a "hemodynamic drug administration event") and is not experiencing a current nociceptive event. The predictive model of the hemodynamic monitoring system can also optionally generate a risk score representing the probability that a patient is experiencing a future hemodynamic drug administration event and is not experiencing a future nociceptive event.

[0017] The predictive model of the hemodynamic monitoring system uses machine learning to extract a set of input features from the patient's arterial pressure. During surgery, in settings such as the operating room (OR), intensive care unit (ICU), or other patient care environments, the hemodynamic monitoring system uses the input feature set to generate the aforementioned risk score for the patient. Based on the level of the risk score, the hemodynamic monitoring system can signal or alert healthcare personnel to inform them that the patient is experiencing or is about to experience a nociceptive event. Upon receiving the signal, healthcare personnel can administer analgesics to the patient to alleviate or prevent the nociceptive event. If the risk score determines that the patient is experiencing or is about to experience a hemodynamic pharmacological event, the hemodynamic monitoring system can signal healthcare personnel so that they do not confuse the patient's hemodynamic pharmacological event with a nociceptive event.

[0018] The machine learning model for the predictive model of the hemodynamic monitoring system was trained using a clinical dataset containing arterial pressure waveforms with clinical annotations indicating the administration of analgesics, vasopressors, positive inotropic agents, fluids, and other drugs that alter cardiovascular hemodynamics. (See below for reference.) Figures 1-10 Describe the hemodynamic monitoring system in detail.

[0019] Figure 1 This is a perspective view of the hemodynamic monitor 10, which determines scores representing the probability of a patient's current nociceptive event and / or scores representing the probability of a patient's future nociceptive events. Figure 1 As shown, the hemodynamic monitor 10 includes a display 12, in Figure 1 In the example, the display 12 presents a graphical user interface including control elements (e.g., graphical control elements) that enable a user to interact with the hemodynamic monitor 10. The 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 (e.g., 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 illustrates five separate 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.

[0020] As further described below, the hemodynamic monitor 10 includes one or more processors and a computer-readable storage device storing nociception detection and prediction software code executable to generate scores representing the probability of a patient's current (i.e., present) nociceptive event and / or scores representing the probability of a patient's future nociceptive events. The hemodynamic monitor 10 can receive sensed hemodynamic data representing the patient's arterial pressure waveform, for example via one or more hemodynamic sensors connected to the hemodynamic monitor 10 via I / O connector 14. The hemodynamic monitor 10 executes the nociception prediction software code to use the received hemodynamic data to obtain multiple nociception analysis parameters (e.g., input features), which may include one or more vital sign parameters characterizing the patient's vital sign data, and differential and combinational parameters derived from the one or more vital sign parameters, as further described below.

[0021] like Figure 1 As shown, the hemodynamic monitor 10 can display 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, for example... 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.

[0022] Figure 2 This is a perspective view of hemodynamic sensor 16, which can be attached to a patient to sense hemodynamic data representing the patient's arterial pressure. Figure 2 The hemodynamic sensor 16 shown is an example of a minimally invasive hemodynamic sensor that can be attached to a patient via, for example, a radial artery catheter inserted into the patient's arm. In other examples, the hemodynamic sensor 16 can be attached to the patient via a femoral artery catheter inserted into the patient's leg.

[0023] like Figure 2 As shown, the hemodynamic sensor 16 includes a housing 18, a fluid inlet port 20, a catheter-side fluid port 22, and an I / O cable 24. The fluid inlet port 20 is configured to connect to a fluid source, such as a saline bag or other fluid inlet source, via tubing or other hydraulic connection. The catheter-side fluid port 22 is configured to connect to a catheter (e.g., a radial artery catheter) inserted into a patient's arm (i.e., a radial artery catheter) or a patient's leg (i.e., a femoral artery catheter) via tubing or other hydraulic connection. The I / O cable 24 is configured to connect via, for example, one or more I / O connectors 14 (… Figure 1 The hemodynamic sensor 16 is connected to the hemodynamic monitor 10. The housing 18 of the hemodynamic sensor 16 encapsulates one or more pressure transducers, communication circuitry, processing circuitry, and corresponding electronic components to sense fluid pressure corresponding to the patient's arterial pressure, which is transmitted to the hemodynamic monitor 10 via the I / O cable 24. Figure 1 ).

[0024] During the procedure, a fluid column (e.g., saline solution) is introduced into the catheter-side fluid port 22 from a fluid source (e.g., a saline bag) via a fluid inlet port 20 through a hemodynamic sensor 16. Arterial pressure is transmitted through the fluid column to a pressure sensor located within the housing 16, which senses the pressure of the fluid column. The hemodynamic sensor 16 converts the sensed fluid column pressure into an electrical signal via a pressure transducer and outputs the corresponding electrical signal to the hemodynamic monitor 10 via an I / O cable 24. Figure 1The hemodynamic sensor 16 thus transmits analog sensor data (or a digital representation of the analog sensor data) representing substantially continuous beat-by-beat monitoring of the patient's arterial pressure to the hemodynamic monitor 10. Figure 1 ).

[0025] 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) transmitter and a light receiver electrically connected to the pressure controller (not shown) to measure changes in the arteries beneath the cot in the finger.

[0026] During the procedure, a pressure controller continuously regulates the pressure within the finger cot to maintain a constant volume (i.e., the arterial unloaded capacity) of the middle finger artery, as measured by the light emitter and receiver of the inflatable finger cot 28. The pressure applied by the pressure controller to continuously maintain the unloaded capacity represents the blood pressure in the finger and is transmitted by the pressure controller to… Figure 1 The hemodynamic monitor 10 is shown. A cardiac reference sensor 30 measures the hydrostatic height difference between the level at which the finger is held and a reference level for pressure measurement (typically the heart level). Therefore, the hemodynamic sensor 26 transmits sensor data representing a substantially continuous beat-by-beat monitoring of the patient's arterial pressure waveform.

[0027] Figure 4 This is a block diagram of a hemodynamic monitoring system 32, which determines a harm score—representing the probability of a current or future harmless event in patient 36—based on a set of harm perception analysis parameters (also called input features) derived from the patient 36's arterial pressure. The hemodynamic monitoring system 32 monitors the patient 36's arterial pressure and provides an alert to a healthcare worker 38 when the patient 36's harm score rises above a predetermined threshold. The healthcare worker 38 can respond to the alert by administering an appropriate analgesic to the patient 36 to alleviate the current or future harmless event.

[0028] like Figure 4 As shown, the hemodynamic monitoring system 32 includes a hemodynamic monitor 10 and a hemodynamic sensor 34. The hemodynamic monitoring system 32 can be implemented in patient care environments, such as ICUs, ORs, or other patient care environments. Figure 4 As shown, the patient care environment may include a patient 36 trained to use the hemodynamic monitoring system 32 and a healthcare worker 38.

[0029] As mentioned above Figure 1 The hemodynamic monitor 10 may be, for example, an integrated hardware unit including a system processor 40, a system memory 42, a display 12, an analog-to-digital converter (ADC) 44, and a digital-to-analog converter (DAC) 46. In other examples, any one or more components and / or the described functions of the hemodynamic monitor 10 may be distributed across multiple hardware units. For example, in some examples, the display 12 may be a separate display device that is remote from and operatively coupled to the hemodynamic monitor 10. Generally, although in Figure 4 The examples are shown and described as integrated hardware units, but it should be understood that the hemodynamic monitor 10 may include any combination of devices and components that are electrically, communicatively, or otherwise operatively connected to perform the functions attributed to the hemodynamic monitor 10 herein.

[0030] like Figure 4 As shown, system memory 42 stores nociception software code 48 that forms the predictive model of hemodynamic monitor 10. Nociception software code 48 includes a first module 50 for extracting and calculating waveform features from the patient 36's arterial pressure, a second module 51 for extracting input features from the waveform features, and a third module 52 for calculating the probability of nociception for the patient 36 based on the input features. Display 12 provides a user interface 54, which includes control elements 56 that enable the user to interact with the hemodynamic monitor 10 and / or other components of the hemodynamic monitoring system 32. Figure 4 As shown, the user interface 54 also provides a sensory alarm 58 to provide medical personnel with warnings of the patient 36’s current or predicted future injury events, as further described below. The sensory alarm 58 can be implemented as one or more of a visual alarm, an audible alarm, a tactile alarm, or other types of sensory alarms. For example, the sensory alarm 58 can be invoked as any combination of flashing and / or color graphics displayed on the user interface 54 on the display 12, an injury score displayed on the display 12 via the user interface 54, a warning sound such as a siren or a repetitive tone, and a tactile alarm configured to vibrate the hemodynamic monitor 10 or otherwise deliver a physical impulse perceptible to the medical worker 38 or other user.

[0031] A hemodynamic sensor 34 may be attached to a patient 36 to sense hemodynamic data representing the patient's arterial pressure waveform. The hemodynamic sensor 34 is operatively connected to a hemodynamic monitor 10 (e.g., via a wired or wireless connection, or both electrically and / or communicatively) to provide the sensed hemodynamic data to the hemodynamic monitor 10. In some examples, the hemodynamic sensor 34 provides the hemodynamic data representing the patient's arterial pressure waveform as an analog signal to the hemodynamic monitor 10, which is then converted by an ADC 44 into digital hemodynamic data representing 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 use an ADC 44. In other examples, the hemodynamic sensor 34 may provide the hemodynamic data representing the patient's arterial pressure waveform as an analog signal to the hemodynamic monitor 10, which the hemodynamic monitor 10 analyzes in analog form of the analog signal.

[0032] The hemodynamic sensor 34 can be a non-invasive or minimally invasive sensor attached to the patient 36. For example, the hemodynamic sensor 34 can be a minimally invasive hemodynamic sensor 16. Figure 2 ), non-invasive hemodynamic sensor 26 ( Figure 3 This can take the form of a hemodynamic sensor 34 or other minimally invasive or non-invasive hemodynamic sensors. In some examples, the hemodynamic sensor 34 can be non-invasively attached to the limbs of the patient 36, such as the wrist, arm, fingers, ankle, toes, or other extremities of the patient 36. Therefore, the hemodynamic sensor 34 can take the form of a small, lightweight, and comfortable hemodynamic sensor suitable for prolonged wear by the patient 36 to provide essentially continuous beat-by-beat monitoring of the patient 36's arterial pressure over extended periods of time (e.g., minutes or hours).

[0033] In some examples, the hemodynamic sensor 34 can be configured to sense the 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 inserted into the arm of the patient 36. In other examples, the hemodynamic sensor 34 can be attached to the patient 36 via a femoral artery catheter inserted into the leg of the patient 36. Such minimally invasive techniques can similarly enable the hemodynamic sensor 34 to provide essentially continuous beat-by-beat monitoring of the patient 36's arterial pressure over extended time periods (e.g., minutes or hours).

[0034] System processor 40 is a hardware processor configured to execute nociceptive software code 48, which implements first module 50, second module 51, and third module 52 to generate a nociceptive score representing the probability of a current nociceptive event or the probability of a future nociceptive event for the patient 36. Examples of system processor 40 may include any one or more of a microprocessor, controller, digital signal processor (DSP), application-specific integrated circuit (ASIC), field-programmable gate array (FPGA), or other equivalent discrete or integrated logic circuits.

[0035] System memory 42 can be configured to store information within hemodynamic monitor 10 during surgery. In some examples, system memory 42 is described as a computer-readable storage medium. In some examples, the computer-readable storage medium may include a non-transitory medium. The term "non-transitory" may mean that the storage medium is not embodied in a carrier wave or propagating signal. In some examples, a non-transitory storage medium may store data that may change over time (e.g., in RAM or cache). 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, magnetic hard disks, optical disks, flash memory, or electrically programmable memory (EPROM) or electrically erasable programmable memory (EEPROM).

[0036] 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 devices suitable for providing information to the user in graphical form. User interface 54 may include graphical and / or physical control elements that enable user input to interact with other components of hemodynamic monitor 10 and / or hemodynamic monitoring system 32. In some examples, user interface 54 may take the form of a graphical user interface (GUI) presented on a touch-sensitive and / or presence-sensitive display screen, such as display 12, displaying graphical control elements. In such examples, user input may be received in the form of gesture input (e.g., touch gestures, scroll gestures, zoom gestures, or other gesture inputs). 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.

[0037] During the procedure, hemodynamic sensor 34 senses hemodynamic data representing the arterial pressure waveform of patient 36. Hemodynamic sensor 34 provides 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.

[0038] Nociception software code 48 may include nociception detection software code. System processor 40 executes the nociception detection software code of nociception software code 48 to determine a nociception detection score representing the probability of a current nociception event for patient 36 using received hemodynamic data. For example, system processor 40 may execute a first module 50 to perform waveform analysis of the hemodynamic data to determine multiple signal measurements. System processor 40 executes a second module 51 to extract nociception detection input features from the multiple signal measurements detecting a nociception event for patient 36. System processor 40 executes a third module 52 to determine a nociception detection score representing the probability of a nociception event for patient 36 based on the nociception detection input features. If the nociception detection score meets predetermined detection criteria, system processor 40 invokes a sensory alarm 58 on user interface 54 to send a first sensory signal to alert healthcare worker 38 that patient 36 is currently experiencing a current nociception event. Healthcare worker 38 may respond to the alert by administering an analgesic to patient 36 or by applying any other form of treatment to patient 36 to alleviate the current nociception event.

[0039] The nociception software code 48 may also include nociception prediction software code. System processor 40 executes the nociception prediction software code of the nociception software code 48 to determine a nociception prediction score representing the probability of a future nociception event for patient 36 using received hemodynamic data. For example, system processor 40 may execute a first module 50 to perform waveform analysis of the hemodynamic data to determine multiple signal measurements. System processor 40 executes a second module 51 to extract nociception prediction input features from the multiple signal measurements predicting future nociception events for patient 36. System processor 40 executes a third module 52 to determine a nociception prediction score representing the probability of a future nociception event for patient 36 based on the nociception prediction input features. If the nociception prediction score meets predetermined prediction criteria, system processor 40 invokes a sensory alarm 58 on user interface 54 to send a second sensory signal to warn healthcare worker 38 that patient 36 will soon experience a future nociception event. Healthcare worker 38 may respond to this warning by administering an analgesic to patient 36 or by applying any other form of treatment to patient 36 to mitigate or prevent the onset of the predicted future nociceptive event.

[0040] In addition to detecting current nociceptive events and predicting future nociceptive events, the hemodynamic monitoring system 32 can distinguish when the patient 36 is experiencing a current nociceptive event and when the patient 36 is only responding to a hemodynamic drug previously administered to the patient 36 by a healthcare professional 38 (hereinafter referred to as a hemodynamic drug administration event). A hemodynamic drug administration event is defined as an event in which the patient 36 experiences an increase in heart rate and blood pressure due to the administration of a compound that alters cardiovascular hemodynamics (e.g., analgesics, vasopressors, positive inotropic agents, fluids, and / or other drugs) and is not a nociceptive event for the patient 36. The nociceptive software code 48 includes hemodynamic drug detection software code for detecting the presence of a hemodynamic drug administration event in the patient 36. The system processor 40 executes the hemodynamic drug detection software code of the nociceptive software code 48 to determine a hemodynamic drug detection score representing the probability that the hemodynamic drug administration event caused an increase in the heart rate and blood pressure in the patient 36 using the received hemodynamic data. For example, the system processor 40 may execute a first module 50 to perform waveform analysis on the hemodynamic data to determine multiple signal measurements. System processor 40 executes second module 51 to extract hemodynamic drug administration input features from multiple signal measurements detecting the current effect of a hemodynamic drug administration event in patient 36. System processor 40 executes third module 52 to determine a hemodynamic drug administration score for patient 36 based on the hemodynamic drug administration input features. If the hemodynamic drug administration score meets predetermined hemodynamic testing criteria, system processor 40 invokes sensory alarm 58 on user interface 54 to send a third sensory signal to alert healthcare worker 38 that patient 36 is experiencing a hemodynamic drug administration event, rather than a current nociceptive event. The hemodynamic drug administration score and the third sensory signal help prevent healthcare worker 38 from confusing a hemodynamic drug administration event with a nociceptive event and prevent healthcare worker 38 from unnecessarily administering analgesics to patient 36.

[0041] The nociceptive software code 48 also includes hemodynamic drug prediction software code for detecting the onset of a future hemodynamic drug administration event in patient 36. System processor 40 executes the hemodynamic drug prediction software code of the nociceptive software code 48 to determine a hemodynamic drug prediction score representing the probability that the hemodynamic drug administration event will result in an increase in heart rate and blood pressure in patient 36, using received hemodynamic data. For example, system processor 40 may execute a first module 50 to perform waveform analysis of the hemodynamic data to determine multiple signal measurements. System processor 40 executes a second module 51 to extract hemodynamic drug prediction input features from the multiple signal measurements detecting the onset of a hemodynamic drug administration event in patient 36. System processor 40 executes a third module 52 to determine a hemodynamic drug prediction score for patient 36 based on the hemodynamic drug prediction input features. If the hemodynamic drug prediction score meets predetermined hemodynamic prediction criteria, system processor 40 invokes a sensory alarm 58 on user interface 54 to send a fourth sensory signal to warn healthcare worker 38 that patient 36 will soon experience a hemodynamic drug administration event. Hemodynamic drug prediction scores and fourth sensory signals help prevent healthcare professionals from confusing future hemodynamic drug administration events with future nociceptive events and prevent them from unnecessarily administering analgesics to patients.

[0042] The system memory 42 of the hemodynamic monitor 10 may also include stability detection software code for detecting the stable period of the patient 36. The stable period is defined as the period during which the patient 36 has not experienced a nociceptive event or a hemodynamic drug administration event. The stability detection software code may be a sub-part of the nociceptive software code 48. The system processor 40 executes the stability detection software code to extract stability detection input features from multiple signal measurements. The stability detection software code may use a second module 51 to extract stability detection input features from multiple signal measurements. The stability detection input features detect the stable period of the patient 36. The system processor 40 executes a third module 52 to determine a stability score for the patient 36 based on the stability detection input features. The system processor 40 outputs the stability score of the patient 36 to the user interface 54 of the display 12.

[0043] System processor 40 can execute first module 50 to extract a single batch of multiple signal measurements within a given time unit, and second module 51 can use that single batch of signal measurements to extract all of the nociception detection input features, nociception prediction input features, hemodynamic drug detection input features, hemodynamic drug prediction input features, and stability detection input features within that time unit. Second module 51 can simultaneously extract all of the nociception detection input features, nociception prediction input features, hemodynamic drug detection input features, hemodynamic drug prediction input features, and stability detection input features from multiple signal measurements. System processor 40 can execute third module 52 to simultaneously determine the nociception detection score, nociception prediction score, hemodynamic drug detection score, hemodynamic drug prediction score, and stability score. In some examples, the nociception software code 48 of the hemodynamic monitor 10 can utilize a classification-type machine learning model with binary positive labels against negative labels. In some examples, processor 40 can output both the nociceptive test score and the hemodynamic drug administration score to display 12 to compare and contrast the two probabilities and help healthcare professionals 38 better understand whether the nociceptive event or the hemodynamic drug administration event caused the increase in blood pressure and heart rate in patient 36.

[0044] Alternatively, the nociception software code 48 of the hemodynamic monitor 10 can utilize a multi-class machine learning model with three labels: nociceptive event, hemodynamic pharmacological event, and stable period. For example, the processor 40 can output the nociceptive detection score, the stability score, and the hemodynamic pharmacological detection score together to the display 12, thereby comparing all three probabilities together: the probability that the patient is experiencing a current nociceptive event, the probability that the patient is experiencing a current hemodynamic pharmacological administration event, and the probability that the patient is stable. See below for reference. Figures 5-8 The machine learning model discussed can be trained on the hemodynamic monitor 10 to identify and / or predict these types of events and experiences in the arterial pressure waveform of patient 36 using a clinical dataset containing clinical annotations of arterial pressure waveforms and administration of compounds that alter cardiovascular hemodynamics (e.g., analgesics, vasopressors, positive inotropic agents, fluids, and / or other drugs).

[0045] Figure 5 This is a schematic diagram of a clinical dataset 60 used for data mining and machine training of a hemodynamic monitor 10. The clinical dataset 60 includes a first dataset 61 containing a collection of arterial pressure waveforms recorded from previous patients. The first dataset 61 can be generated by an invasive hemodynamic sensor (e.g., Figure 2 The hemodynamic sensor 16 shown collects the data, or it is collected by a non-invasive hemodynamic sensor (e.g., Figure 3 The hemodynamic sensor 26 shown collects the data. Clinical dataset 60 also includes a second dataset 62 containing logs of instances where compounds that alter cardiovascular hemodynamics (e.g., analgesics, vasopressors, positive inotropic agents, fluids, and / or other drugs) were administered to previous patients in the first dataset 61, while their arterial pressure waveforms were recorded. Healthcare professionals can directly input the dosing information into the same hemodynamic monitor that collected the first dataset 61, thus simultaneously collecting both the first dataset 61 and the second dataset 62. Figures 6-8 As shown, the information in the second dataset 62 is annotated and labeled onto the set of arterial pressure waveforms in the first dataset 61.

[0046] Figure 6 These are graphs showing the changes in systolic blood pressure over time (hereinafter referred to as the "SBP graph") and heart rate over time (hereinafter referred to as the "HR graph"). Before the clinical dataset 60 can be used to train the hemodynamic monitor 10, the SBP graph and HR graph are determined for each arterial pressure waveform collected in the clinical dataset 60. Figure 6 The SBP and HR charts shown are examples of one of the arterial pressure waveforms (not shown) from Clinical Dataset 60. After determining the SBP and HR charts for each arterial pressure waveform collected in Clinical Dataset 60, both the SBP and HR charts were annotated to indicate when a compound altering cardiovascular hemodynamics was administered to a clinical patient. For example, Figure 6 The SBP and HR charts shown include analgesic label 64, which is a vertical bar extending across the SBP and HR charts at the same time point. Figure 6 The label 64 of the analgesic indicates that clinical patients should use analgesics prescribed by the manufacturer. Figure 6 The analgesics were administered during the time periods represented by the SBP and HR charts. After the SBP and HR charts were annotated and labeled to show the clinical patients the drug administration, nociceptive data segment 66 was identified and labeled on the SBP and HR charts.

[0047] like Figure 6 As shown, nociceptive data segments 66 are identified on both the SBP and HR charts by locating time periods in both the SBP and HR charts, wherein the clinical patient's systolic blood pressure increases by at least a threshold amount (e.g., 20% or other threshold amount) compared to the previous time period, and the clinical patient's heart rate also increases by at least a threshold amount (e.g., 20% or other threshold amount) compared to the previous time period, and this increase begins before the infusion of compounds that alter cardiovascular hemodynamics (e.g., analgesics, vasopressors, positive inotropic agents, fluids, and / or other drugs) has begun. Figure 6In the meantime, both the SBP and HR charts increased by more than 20% at point 68, which appears before label 64 of the analgesic, thus indicating... Figure 6 The beginning of segment 66 in the data on the perception of harm. Figure 6 The nociceptive data segment 66 continues until both the SBP and HR charts begin to decline due to timely administration of analgesics to the clinical patient at analgesic label 64. The decline in the SBP and HR charts is indicated by endpoint 70. Both start point 68 and endpoint 70 are marked on the HR and SBP charts, and the time period between start point 68 and endpoint 70 is designated as a nociceptive data segment 66. Using the analgesic label 64 and nociceptive data segment 66 identified on the SBP and HR charts, a nociceptive data segment 66 is generated. Figure 6 The arterial pressure waveforms in the SBP and HR charts can also be annotated and labeled to show when the analgesic label 64 and the nociceptive data segment 66 appear on the arterial pressure waveforms. Once labeled with the analgesic label 64 and the nociceptive data segment 66, the arterial pressure waveforms are ready for data mining and machine training of the hemodynamic monitor 10 to detect nociceptive events. (See below for reference.) Figures 9-10 Further discussion involves performing waveform analysis on a clinical dataset 60 containing nociceptive data segment 66 to compute multiple signal measurements, and then using these signal measurements to compute nociceptive detection input features that optimally detect the probability of the current nociceptive event.

[0048] Figure 6 The predicted data segment 71 can also be used to train the hemodynamic monitor 10 to predict future injury-perceiving events. The predicted data segment 71 can be identified in the clinical dataset 60 by recognizing the preceding time period before the increase in the SBP and HR charts. Figure 6 In the examples, the previous time period occurred before the starting point 68. This previous time period before the starting point 68 is labeled as forecast data segment 71. In some examples, forecast data segment 71 includes a time period that begins 15 minutes before and immediately ends before the injury perception data segment 66. In other examples, forecast data segment 71 may include a longer or shorter time period before the injury perception data segment 66. The forecast data segment 71, identified and labeled on the SBP and HR charts, is used to generate... Figure 6 The arterial pressure waveforms in the SBP and HR charts can also be annotated and labeled to show when the predicted data segment 71 appears on the arterial pressure waveform. Once labeled with the predicted data segment 71, the arterial pressure waveform is ready for data mining and machine training of the hemodynamic monitor 10 to predict nociceptive events. See below for reference. Figures 9-10Further discussion involves performing waveform analysis on the clinical dataset 60 containing the prediction data segment 71 to compute multiple signal measurements, and then using these signal measurements to compute the nociceptive prediction input features that optimally detect the probability of future nociceptive events.

[0049] Figure 7 This is a graph illustrating another SBP and HR chart derived from arterial pressure waveform segments (not shown) from clinical dataset 60. Figure 7 The arterial pressure waveform segments in the SBP and HR charts can be identified as stable data segments 72 and used for data mining and machine training of the hemodynamic monitor 10 to detect when a patient experiences a stable period without perceived harm. If there is no increase greater than a threshold amount (e.g., 20% or other threshold amount) in the SBP chart, no increase greater than a threshold amount (e.g., 20% or other threshold amount) in the HR chart, and no infusion of compounds that alter cardiovascular hemodynamics, then the arterial pressure waveform segment in the clinical dataset 60 is identified as a stable data segment 72. Figure 7 The example SBP chart shown does not include increases greater than 20% between the start point 74 and the end point 76. Figure 7 The HR chart in the example also does not include increases greater than 20% between the starting point 74 and the ending point 76. Figure 7 The HR and SBP charts also do not include any notes or labels indicating the infusion of compounds that alter the cardiovascular hemodynamics of clinical patients between the starting point 74 and the endpoint 76. Given... Figure 7 The aforementioned characteristics of the HR chart and SBP chart in the example, Figure 7 The HR chart and SBP chart are labeled as stable data segment 72 between start point 74 and end point 76. (Generation) Figure 7 The arterial pressure waveform segments (not shown) from the HR and SBP charts are also labeled as stable data segment 72 between the start point 74 and the end point 76. Once labeled with stable data segment 72, the arterial pressure waveform segments are ready for stable data mining and stable machine training of the hemodynamic monitor 10. (See below for reference.) Figures 9-10 Further discussion involves performing waveform analysis on the clinical dataset 60 containing stable data segment 72 to compute multiple signal measurements, and then using these signal measurements to compute stable detection input features that optimally detect the probability of a stable period.

[0050] Figure 8 This is a graph illustrating another SBP and HR chart derived from arterial pressure waveform segments (not shown) from clinical dataset 60. Figure 8The arterial pressure waveform segments in the SBP and HR charts can be identified as hemodynamic drug administration data segment 78 (hereinafter referred to as "HDA data segment 78") and used for data mining and machine training of the hemodynamic monitor 10 to detect when a patient experiences a current hemodynamic drug administration event. If the arterial pressure waveform segment includes the infusion of a compound that alters cardiovascular hemodynamics into a clinical patient and increases by at least a threshold amount (e.g., 20% or other threshold amount) in both the SBP and HR charts after infusion, then the arterial pressure waveform segment in the clinical dataset 60 is identified as HDA data segment 78. Figure 8 In the example, the vasopressor infusion label 80 on the SBP and HR charts indicates that a clinical patient was administered a vasopressor. Shortly after the vasopressor infusion label 80, the SBP chart increases by at least 20% between the start point 82 and the end point 84. The HR chart also increases by at least 20% between the start point 82 and the end point 84, thus indicating that HDA data segment 78 occurs between the start point 82 and the end point 84. Figure 8 The HR chart and SBP chart are labeled as HDA data segment 78 between start point 82 and end point 84. (Generation) Figure 8 The arterial pressure waveform segments (not shown) of the HR and SBP charts are also labeled as HDA data segment 78 between start point 82 and end point 84. Once labeled with HDA data segment 78, the arterial pressure waveform segments are ready for hemodynamic drug detection data mining and machine training for the hemodynamic monitor 10. (See below for reference.) Figures 9-10 Further discussion involves performing waveform analysis on a clinical dataset 60 containing HDA data segment 78 to compute multiple signal measurements, which are then used to compute hemodynamic drug detection input features that optimally detect the probability of the current hemodynamic drug administration event.

[0051] Figure 8 The hemodynamic drug prediction data segment 85 (hereinafter referred to as "HDP data segment 85") can also be used to train the hemodynamic monitor 10 to predict future hemodynamic drug administration events. This can be achieved by identifying... Figure 8 The HDP data segment 85 was identified in clinical dataset 60 before the SBP and HR charts began to increase in the preceding time period. Figure 8 In the examples, the previous time period appears before the starting point 82. The previous time period before the starting point 82 is labeled HDP data segment 85. In some examples, HDP data segment 85 includes a time period that begins 15 minutes before HDA data segment 78 and ends immediately before HDA data segment 78. In other examples, HDP data segment 85 may include a longer or shorter time period before HDA data segment 78. HDP data segment 85, identified and labeled on the SBP and HR charts, is used to generate... Figure 8 The arterial pressure waveforms in the SBP and HR charts can also be annotated and labeled to show when HDP data segment 85 appears on the arterial pressure waveform. Once labeled with HDP data segment 85, the arterial pressure waveform is ready for data mining and machine training of the hemodynamic monitor 10 to predict future hemodynamic drug administration events. See below for reference. Figures 9-10 Further discussion involves performing waveform analysis on a clinical dataset 60 containing HDP data segment 85 to compute multiple signal measurements, which are then used to compute hemodynamic drug prediction input features that optimally detect the probability of future hemodynamic drug administration events.

[0052] Figure 9 It is used for data mining from machine learning models used for training hemodynamic monitor 10. Figures 5-8 The flowchart of method 86 for clinical dataset 60. Figure 9 Method 86 in the middle will also be referenced Figure 10 Simultaneous discussion. Method 86 was applied to each of the nociceptive data segment 66, predictive data segment 71, stable data segment 72, HDA data segment 78, and HDP data segment 85 in the clinical dataset 60 to train the hemodynamic monitor to find previous references. Figure 4 The input features are described. Method 86 will be described as being applied to nociceptive data segment 66 (e.g., ...). Figure 6 (As shown).

[0053] Machine training of hemodynamic monitor 10 to identify Figure 4 The nociception detection input features described herein are first determined by applying method 86 to nociception data segment 66 of clinical dataset 60. The first step 88 of method 86 is to perform waveform analysis on the nociception data segment 66 of arterial waveforms collected in dataset 60 to compute multiple signal measurements of the nociception data segment. Performing waveform analysis on the nociception data segment 66 may include identifying a single cardiac cycle in each arterial pressure waveform of the nociception data segment 66. Figure 10 Example plots are provided illustrating example trajectories of arterial pressure waveforms with identification and magnification of a single cardiac cycle. Next, performing waveform analysis on nociceptive data segment 66 may include identifying dicrotic notches in each single cardiac cycle of each arterial pressure waveform in nociceptive data segment 66, similar to... Figure 10 The example shown. Next, waveform analysis of the nociceptive data segment 66 includes identifying the systolic rise phase, systolic fall phase, and diastolic phase in each individual cardiac cycle of each arterial pressure waveform in the nociceptive data segment 66, similar to... Figure 10 The example shown.

[0054] Signal measurements are extracted from each of the rising systolic phase, falling systolic phase, and diastolic phase of each individual cardiac cycle from each arterial pressure waveform of the nociceptive data segment 66. These signal measurements may correspond to the hemodynamic effects of each of the rising systolic phase, falling systolic phase, and diastolic phase of each individual cardiac cycle. These hemodynamic effects may include contractility, aortic compliance, stroke volume, vascular tone, afterload, and the complete cardiac cycle. Signal measurements calculated or extracted by waveform analysis in the first step 88 of method 86 include the mean, maximum, minimum, duration, area, standard deviation, derivative, and / or morphological measurements of each of the rising systolic phase, falling systolic phase, and diastolic phase of each individual cardiac cycle. Signal measurements may also include heart rate, respiratory rate, stroke volume, pulse pressure, pulse pressure variation, stroke volume variation, mean arterial pressure (MAP), systolic blood pressure (SYS), diastolic blood pressure (DIA), heart rate variability, cardiac output, peripheral resistance, vascular compliance, and / or left ventricular contractility extracted from each individual cardiac cycle of each arterial pressure waveform of the nociceptive data segment 66.

[0055] After determining the signal measurements for the nociceptive data segment 66, step 90 of method 86 is performed on the signal measurements of the nociceptive data segment 66. Step 90 of method 86 calculates the combined measurements between the signal measurements of the nociceptive data segment 66. Calculating the combined measurements between the signal measurements of the nociceptive data segment 66 may include performing all signal measurements of the nociceptive data segment 66... Figure 9 Steps 92, 94, 96, and 98 are shown. Step 92 is performed by arbitrarily selecting three signal measurements from the nociceptive data segment. Next, the order of a different power is calculated for each of the three signal measurements to generate the power of the three signal measurements, as shown. Figure 9 Step 94 is shown. In Figure 9 In step 96, the powers of the three signal measurements are then multiplied to generate a product of powers of the three signal measurements. Step 98 includes performing a receiver operating characteristic (ROC) analysis of the product to derive a combined measurement of the three signal measurements. Steps 92, 94, 96, and 98 are repeated until all combined measurements have been calculated among all signal measurements in the nociceptive data segment 66. The signal measurement of the most predictive top combined measurement (i.e., the combined measurement that meets the threshold prediction criteria) is selected as the top signal measurement of the nociceptive data segment 66 and is labeled as the nociceptive detection input feature. Once the nociceptive detection input feature is determined, the hemodynamic monitor 10 is trained or programmed to detect the arterial pressure waveform of the patient 36 (e.g., Figure 4(As shown) Perform waveform analysis and extract nociceptive detection input features from the arterial pressure waveform of patient 36, and use these nociceptive detection input features to determine the probability that patient 36 is currently experiencing a nociceptive event.

[0056] Similar to how method 86 is applied to nociceptive data segment 66, method 86 is applied to predictive data segment 71, stable data segment 72, HDA data segment 78, and HDP data segment 85 in clinical dataset 60 to determine nociceptive prediction input features, stable detection input features, hemodynamic drug detection input features, and hemodynamic drug prediction input features, respectively.

[0057] While the invention has been described with reference to one or more exemplary embodiments, those skilled in the art will understand that various changes can be made and elements can be substituted with equivalents without departing from the scope of the invention. Furthermore, many modifications can be made to adapt particular situations or materials to the teachings of the invention without departing from the basic scope of the invention. Therefore, the invention is not limited to the specific embodiments disclosed, but rather will include all embodiments falling within the scope of the appended claims.

Claims

1. A method for monitoring a patient's arterial pressure and providing medical personnel with a warning of the patient's current or predicted future sense of harm, the method comprising: Sensed hemodynamic data representing the patient's arterial pressure waveform are received via a hemodynamic monitor; The hemodynamic monitor performs waveform analysis on the sensed hemodynamic data to calculate multiple signal measurements of the sensed hemodynamic data; Extract detection input features for the hemodynamic monitor from the plurality of signal measurements indicating the patient’s current nociceptive event; Predictive input features for the hemodynamic monitor are extracted from the multiple signal measurements used to predict future nociceptive events in the patient. The hemodynamic monitor determines a first harm score representing the probability of the patient's current harm-perceived event based on the detected input features, and a second harm score representing the probability of the patient's future harm-perceived event based on the predicted input features; as well as In response to the first nociceptive score meeting a predetermined level standard, the hemodynamic monitor invokes a sensory alarm to generate a first sensory signal, and In response to the second injury perception score meeting a predetermined level standard, the hemodynamic monitor invokes a sensory alarm to generate a second sensory signal.

2. The method according to claim 1, further comprising: Based on the detection input features and / or the prediction input features, the hemodynamic monitor determines a hemodynamic drug administration score representing the probability of a hemodynamic drug administration event and / or the probability of a future hemodynamic drug administration event for the patient, wherein a hemodynamic drug administration event is defined as an event in which the patient experiences an increase in heart rate and blood pressure due to administration of a compound that alters cardiovascular hemodynamics and is not a nociceptive event for the patient. as well as The hemodynamic drug score is output to the display of the hemodynamic monitor.

3. The method according to claim 2, further comprising: The hemodynamic monitor determines a stability score representing the probability of a stable period based on the detection input characteristics, during which the patient experiences neither a nociceptive event nor a hemodynamic drug administration event; as well as The stability score is output to the display of the hemodynamic monitor.

4. The method according to claim 3, further comprising: Training the hemodynamic monitor to determine the probability of the patient's current nociceptive event, wherein training the hemodynamic monitor includes: Collect clinical datasets, which include arterial pressure waveforms and clinical annotations of compound administration that alters cardiovascular hemodynamics; Identify nociceptive data segments in the clinical dataset, wherein each nociceptive data segment includes: Compared to previous time periods, blood pressure has increased by at least the first threshold. Compared to the previous time period, the heart rate increased by at least a second threshold; and The infusion of the compound that alters cardiovascular hemodynamics was not initiated prior to the aforementioned increase in blood pressure and heart rate; Identify the start and end of the increase in blood pressure and the increase in heart rate; The nociceptive data segment is marked after the onset of the blood pressure increase and the heart rate increase, and during the blood pressure increase and heart rate increase. Waveform analysis is performed on the marked nociceptive data segments to calculate multiple signal measurements of the nociceptive data segments; and The detection input feature is determined by calculating the combined measurements among the plurality of signal measurements and selecting the signal measurement with the most predictive combined measurements from the plurality of signal measurements as belonging to the input feature.

5. The method of claim 4, further comprising training the hemodynamic monitor to determine the probability of the predicted future nociceptive event in the patient by: Before the increase in blood pressure and the increase in heart rate in each of the nociceptive data segments, the preceding time period is identified; The preceding time period for each of the injury perception data segments is marked as a prediction data segment; Perform waveform analysis on the predicted data segment to calculate multiple signal measurements of the predicted data segment; and At least a portion of the predicted input features are determined by calculating combined measurements among the plurality of signal measurements of the predicted data segment and selecting the signal measurement of the most predictive combined measurement from the plurality of signal measurements of the predicted data segment as belonging to the input features.

6. The method according to claim 5, further comprising: Training the hemodynamic monitor to determine the probability of the patient's hemodynamic drug administration event, wherein training the hemodynamic monitor to determine the probability of the patient's hemodynamic drug administration event includes: Identify hemodynamic drug dosing data segments in the clinical dataset, wherein each of the hemodynamic drug dosing data segments includes: Infusion of compounds that alter cardiovascular hemodynamics; The blood pressure rises at least a third threshold after the infusion; and The heart rate increases by at least a fourth threshold after the infusion; The start and end of the increase in blood pressure and heart rate are identified in each of the hemodynamic drug dosing data segments; The hemodynamic drug administration data segment is marked after the onset of the blood pressure rise and the increase in heart rate, and during the blood pressure rise and the increase in heart rate. Waveform analysis was performed on the labeled hemodynamic drug dosing data segments to calculate multiple signal measurements of the hemodynamic drug dosing data segments; and At least a portion of the detection input feature is determined by calculating the combined measurements among the plurality of signal measurements of the hemodynamic drug administration data segment and selecting the signal measurement of the most predictive combined measurement from the plurality of signal measurements of the hemodynamic drug administration data segment as belonging to the input feature.

7. The method according to claim 6, further comprising: Training the hemodynamic monitor to determine the probability of the patient's future hemodynamic drug administration event, wherein training the hemodynamic monitor to determine the probability of the patient's future hemodynamic drug administration event includes: Before the increase in blood pressure and the increase in heart rate in each of the hemodynamic drug dosing data segments, identify the preceding time period; The preceding time period for each of the hemodynamic drug administration data segments is marked as a hemodynamic drug prediction data segment; Waveform analysis is performed on the hemodynamic drug prediction data segment to calculate multiple signal measurements of the hemodynamic drug prediction data segment; and At least a portion of the predicted input features are determined by calculating combined measurements among the plurality of signal measurements in the hemodynamic drug prediction data segment and selecting the signal measurement of the most predictive combined measurement from the plurality of signal measurements in the hemodynamic drug prediction data segment as belonging to the input features.

8. The method according to claim 7, further comprising: Training the hemodynamic monitor to determine the probability of the patient's stable period, wherein training the hemodynamic monitor to determine the probability of the stable period includes: Identify stable data segments in the clinical dataset, wherein each stable data segment includes: Stabilize blood pressure so that it does not increase beyond the first threshold within a set time period; A stable heart rate that does not increase beyond the second threshold during the set time period; and Do not infuse compounds that alter cardiovascular hemodynamics; Identify the start and end of the stable blood pressure and the stable heart rate; The stable data segment is marked with the beginning and end of the stable blood pressure and the stable heart rate; Waveform analysis is performed on the marked stable data segments to calculate multiple stable signal measurements for the stable data segments; and The detection input feature is determined by calculating the combined measurements among the plurality of stable signal measurements and selecting the stable signal measurement that is the most predictive combined measurement from the plurality of stable signal measurements as belonging to the input feature.

9. A system for monitoring a patient's arterial pressure and providing a warning of the patient's pain perception to medical personnel, the system comprising: A hemodynamic sensor that generates hemodynamic data representing the patient's arterial pressure waveform; System memory, which stores the nociception detection software code; The user interface includes sensory alarms that provide sensory signals to warn the medical staff of noxious events in the patient; as well as A hardware processor, configured to execute the nociception detection software code, to: Waveform analysis was performed on the hemodynamic data to determine multiple signal measurements; Extract detection input features from the multiple signal measurements that indicate the patient's nociceptive events; Based on the detected input features, a harm perception score is determined, representing the probability of the harm perception event for the patient. as well as The sensory alarm on the user interface is invoked in response to the injury perception score meeting a predetermined detection criterion. The system memory is configured to store harm perception prediction software code for determining the probability of predicted future harm perception events for the patient, the sensory alarm provides a second sensory signal to alert the medical staff to the predicted future harm perception events, and the hardware processor is configured to execute the harm perception prediction software code to: Extract predictive input features from the multiple signal measurements that predict the patient's future nociceptive events; Based on the predicted input features, a harm perception prediction score is determined, representing the probability of the patient's future harm perception event. The sensory alarm on the user interface is invoked in response to the injury perception prediction score meeting a predetermined prediction criterion.

10. The system of claim 9, wherein the detection input features of the nociception detection software code are determined by a detection machine training method, wherein the detection machine training method comprises: Collect clinical datasets containing arterial pressure waveforms and clinical annotations of compound administration that alters cardiovascular hemodynamics; Identify nociceptive data segments in the clinical dataset, wherein each nociceptive data segment includes: Compared to previous time periods, blood pressure has increased by at least the first threshold. Compared to the previous time period, the heart rate increased by at least a second threshold; and The infusion of the compound that alters cardiovascular hemodynamics was not initiated prior to the aforementioned increase in blood pressure and heart rate; Identify the start and end of the increase in blood pressure and the increase in heart rate; The nociceptive data segment is marked after the onset of the blood pressure increase and the heart rate increase, and during the blood pressure increase and heart rate increase. Waveform analysis is performed on the marked nociceptive data segments to calculate multiple signal measurements of the nociceptive data segments; and The detection input feature is determined by calculating a combination measurement among the plurality of signal measurements of the nociceptive data segment, selecting the top signal measurement of the most predictive combination measurement from the plurality of signal measurements of the nociceptive data segment, and marking the top signal measurement as the detection input feature.

11. The system of claim 10, wherein the predictive input features of the nociceptive prediction software code are determined by predictive machine training, the predictive machine training comprising: Before the increase in blood pressure and heart rate in each of the nociceptive data segments begins, the preceding time period is identified; The preceding time period for each of the injury perception data segments is marked as a prediction data segment; Waveform analysis is performed on the predicted data segment to calculate multiple signal measurements of the predicted data segment; as well as The predictive input feature is determined by calculating the combined measurements among the plurality of signal measurements of the predictive data segment and selecting the signal measurement of the most predictive combined measurement from the plurality of signal measurements of the predictive data segment as the predictive input feature.

12. The system of claim 11, wherein the system memory stores hemodynamic drug administration software code for detecting hemodynamic drug administration events of the patient, the sensory alarm provides a third sensory signal to alert medical personnel of the hemodynamic drug administration event, and the hardware processor is configured to execute the hemodynamic drug administration software code to: Extract hemodynamic drug detection input features from the multiple signal measurements indicating the hemodynamic drug administration event in the patient; Based on the input features of the hemodynamic drug detection, a hemodynamic drug detection score representing the probability of the hemodynamic drug administration event is determined; In response to the hemodynamic drug test score meeting a predetermined prediction criterion, the sensory alarm on the user interface is invoked.

13. The system of claim 12, wherein the hemodynamic drug detection input features of the hemodynamic drug detection software code are determined by hemodynamic drug detection machine training, wherein the hemodynamic drug detection machine training includes: Identify hemodynamic drug dosing data segments in the clinical dataset, wherein each hemodynamic drug dosing data segment includes: Infusion of compounds that alter cardiovascular hemodynamics; The blood pressure rises at least a third threshold after the infusion; and The heart rate increases by at least a fourth threshold after the infusion; In each of the hemodynamic drug dosing data segments, identify the start and end of the increase in blood pressure and the increase in heart rate; The hemodynamic drug administration data segment is marked after the onset of the blood pressure increase and the heart rate increase, and during the blood pressure increase and heart rate increase. Waveform analysis was performed on the marked hemodynamic drug dosing data segments to calculate multiple signal measurements of the hemodynamic drug dosing data segments; and The hemodynamic drug administration data segment input feature is determined by calculating the combined measurements among the plurality of signal measurements of the hemodynamic drug administration data segment and selecting the signal measurement with the most predictive combined measurements from the plurality of signal measurements of the hemodynamic drug administration data segment as the hemodynamic drug administration data segment input feature.

14. The system of claim 13, wherein the system memory stores hemodynamic drug prediction software code for determining the probability of future hemodynamic drug administration events for the patient, the sensory alarm provides a fourth sensory signal to alert the healthcare professional of the future hemodynamic drug administration events, and the hardware processor is configured to execute the hemodynamic drug prediction software code to: Extract hemodynamic drug prediction input features from the multiple signal measurements that predict the patient's future hemodynamic drug administration events; Based on the hemodynamic drug prediction input features, a hemodynamic drug prediction score representing the probability of the patient's future hemodynamic drug administration event is determined; as well as In response to the hemodynamic drug prediction score meeting the predetermined hemodynamic drug prediction criteria, the sensory alarm of the user interface is invoked.

15. The system of claim 14, wherein the hemodynamic drug prediction input features of the hemodynamic drug prediction software code are determined by training a hemodynamic drug prediction machine, wherein the hemodynamic drug prediction machine training includes: Before the elevation of blood pressure and the increase of heart rate in each of the hemodynamic drug dosing data segments, identify the preceding time period; Each preceding time period in the hemodynamic drug administration data segment is labeled as a hemodynamic drug prediction data segment; Waveform analysis is performed on the hemodynamic drug prediction data segment to calculate multiple signal measurements of the hemodynamic drug prediction data segment; as well as The hemodynamic drug prediction input feature is determined by calculating the combined measurements among the plurality of signal measurements in the hemodynamic drug prediction data segment, and selecting the signal measurement with the most predictive combined measurements from the plurality of signal measurements in the hemodynamic drug prediction data segment as the hemodynamic drug prediction input feature.

16. The system of claim 15, wherein the system memory stores stability detection software code for determining the probability of a stable period in the patient, and the hardware processor is configured to execute the stability detection software code to: Extract stability detection input features from the multiple signal measurements indicating the patient's stable period; Based on the stable detection input features, a stability score is determined representing the probability that the patient has not experienced either a nociceptive event or a hemodynamic drug administration event during the stable period. as well as The stable score is output to the display.

17. The system of claim 16, wherein the stability detection input features of the stability detection software code are determined by stability detection machine training, wherein stability detection machine training includes: Identify stable data segments in the clinical dataset, wherein each stable data segment includes: Stabilize blood pressure so that it does not increase beyond the first threshold within a set time period; A stable heart rate that does not increase beyond the second threshold during the set time period; and Do not administer infusions of compounds that alter cardiovascular hemodynamics; Identify the start and end of the stable blood pressure and the stable heart rate; The stable data segment is marked with a start and an end point from the stable blood pressure and the stable heart rate; Waveform analysis is performed on the marked stable data segments to calculate multiple stable signal measurements for the stable data segments; and The stability detection input feature is determined by calculating the combined measurements among the plurality of stable signal measurements and selecting the stable signal measurement with the most predictive combined measurements from the plurality of stable signal measurements as the stability detection input feature.

18. The system of claim 17, wherein waveform analysis of the marked nociceptive data segment to calculate a plurality of signal measurements of the nociceptive data segment comprises: Identify individual cardiac cycles within the arterial pressure waveforms of the clinical dataset; Identify dicrotic notches in each of the single cardiac cycle; In each of the single cardiac cycle, the rising systolic phase, the falling systolic phase, and the diastolic phase are identified; as well as Signal measurements are extracted from each of the rising systolic phase, the falling systolic phase, and the diastolic phase of each of the individual cardiac cycles.

19. The system of claim 18, wherein the signal measurement corresponds to the hemodynamic effects of each of the rising systolic phase, the falling systolic phase, and the diastolic phase of each of the individual cardiac cycles, and wherein the hemodynamic effects include contractility, aortic compliance, stroke volume, vascular tone, afterload, and complete cardiac cycle.

20. The system of claim 19, wherein the signal measurement includes average, maximum, minimum, duration, area, standard deviation, derivative and / or morphological measurements of each of the rising systolic phase, the falling systolic phase and the diastolic phase of each of the individual cardiac cycles.

21. The system of claim 20, wherein the signal measurements include heart rate, respiratory rate, stroke volume, pulse pressure, pulse pressure variation, stroke volume variation, mean arterial pressure (MAP), systolic blood pressure (SYS), diastolic blood pressure (DIA), heart rate variability, cardiac output, peripheral resistance, vascular compliance, and / or left ventricular contractility extracted from each of the individual cardiac cycles.

22. The system of claim 21, wherein calculating the combined measurement among the plurality of signal measurements of the nociceptive data segment comprises: Step one is performed by arbitrarily selecting three signal measurements from the plurality of signal measurements in the nociceptive data segment; Step two is performed by calculating the order of a different power for each of the three signal measurements to generate the powers of the three signal measurements; Step three is performed by multiplying the powers of the three signal measurements to generate a product of the powers of the three signal measurements; Step four is performed by conducting receiver operating characteristic analysis (ROC analysis) on the product to obtain a combined measurement of the three signal measurements; as well as Repeat steps one, two, three, and four until all the combined measurements have been calculated among all the plurality of signal measurements in the injury perception data segment.

23. The system according to any one of claims 9-22, wherein the hemodynamic sensor is a non-invasive hemodynamic sensor that can be attached to the patient's limbs.

24. The system according to any one of claims 9-22, wherein the hemodynamic sensor is a hemodynamic sensor based on a minimally invasive arterial catheter.

25. The system according to any one of claims 9-22, wherein the hemodynamic sensor generates the hemodynamic data as a simulated hemodynamic sensor signal representing the arterial pressure waveform of the patient.

26. The system of claim 25 further includes an analog-to-digital converter that converts the analog hemodynamic sensor signal into digital hemodynamic data representing the arterial pressure waveform of the patient.

Images