US20220330842A1 - Non-invasive blood pressure measurement system - Google Patents

Abstract

A system for non-invasively determining an indication of an individual's blood pressure is described. In certain embodiments, the system calculates pulse wave transit time using two acoustic sensors. The system can include a first acoustic sensor configured to monitor heart sounds of the patient corresponding to ventricular systole and diastole and a second acoustic sensor configured to monitor arterial pulse sounds at an arterial location remote from the heart. The system can advantageously calculate a arterial pulse wave transit time (PWTT) that does not include the pre-ejection period time delay. In certain embodiments, the system further includes a processor that calculates the arterial PWTT obtained from the acoustic sensors. The system can use this arterial PWTT to determine whether to trigger an occlusive cuff measurement.

Description

RELATED APPLICATIONS
• This application claims priority under 35 U.S.C. § 119(e) to U.S. Provisional Application No. 61/366,862, filed Jul. 22, 2010, entitled “System for Triggering a Non-Invasive Blood Pressure Device” and to U.S. Provisional Application No. 61/469,511, filed Mar. 30, 2011, entitled “Non-Invasive Blood Pressure Measurement System,” the disclosures of which are hereby incorporated by reference in their entirety.
BACKGROUND
• Prolonged reduction or loss of blood pressure in a patient severely limits the amount of tissue perfusion of the patient and therefore causes damage to or death of the tissue. Although some tissues can tolerate hypoperfusion for long periods of time, the brain, heart and kidneys are very sensitive to a reduction in blood flow. Thus, during and after medical procedures and at other times, blood pressure is a frequently monitored vital sign. Blood pressure can be affected by the type of medical procedure performed and by physiological factors such as the body's reaction to the medical procedure. Moreover, blood pressure is often manipulated and controlled using various medications. Medical procedures, physiological factors, and medications can cause the blood pressure of a patient to change rapidly.
• The traditional method of measuring blood pressure is with a stethoscope, occlusive cuff, and pressure manometer. Blood pressure cuff instruments make only a spot-check measurement, so repetitive interval measurements are often used to trend patient status. More frequent intervals improve vigilance at the expense of patient discomfort, possible patient injury (e.g., due to occlusion of blood vessels), and excessive battery consumption.
SUMMARY
• In certain embodiments, a method of monitoring blood pressure of a patient includes receiving a physiological electrical signal from an electrical sensor coupled with a patient. The physiological electrical signal can reflect electrical activity of the patient's heart. The method may also include receiving a cardiac ejection signal from a second sensor coupled with the patient. This cardiac ejection signal can reflect a cardiac ejection event associated with ejection of blood from the patient's heart. In addition, the method may include receiving an arterial pulse signal from a third sensor coupled with a limb of the patient. The method can also include determining an arterial pulse wave transit time (PWTT) that compensates for a pre-ejection period of a cardiac cycle associated with the patient's heart, based at least partly on the physiological electrical signal, the cardiac ejection signal, and the arterial pulse signal. Moreover, the method may include triggering an occlusive blood pressure cuff to obtain a blood pressure measurement from the patient responsive to a change in the arterial PWTT.
• For purposes of summarizing the disclosure, certain aspects, advantages and novel features of the inventions have been described herein. It is to be understood that not necessarily all such advantages can be achieved in accordance with any particular embodiment of the inventions disclosed herein. Thus, the inventions disclosed herein can be embodied or carried out in a manner that achieves or optimizes one advantage or group of advantages as taught herein without necessarily achieving other advantages as can be taught or suggested herein.
BRIEF DESCRIPTION OF THE DRAWINGS
• Throughout the drawings, reference numbers can be re-used to indicate correspondence between referenced elements. The drawings are provided to illustrate embodiments of the inventions described herein and not to limit the scope thereof.
• FIG. 1 illustrates an embodiment of a parameter calculation system;
• FIGS. 2A and 2B illustrate plots of plethysmograph and electrocardiograph (ECG) waveforms that can be used to calculate pulse wave transit time (PWTT);
• FIGS. 2C and 2D illustrate plots of acoustic waveforms that can be used to calculate PWTT;
• FIG. 3 illustrates another embodiment of a blood pressure monitoring system;
• FIG. 4A illustrates a plot of acoustic and ECG waveforms;
• FIG. 4B illustrates a plot of bioimpedance and ECG waveforms;
• FIGS. 5A and 5B illustrate embodiments of blood pressure monitoring systems coupled to a patient;
• FIG. 6 illustrates example positioning locations for the acoustic sensors that can be used in the various systems and methods described herein;
• FIG. 7 illustrates example positioning locations for acoustic, electrocardiograph (ECG), optical and bioimpedance sensors that can be used in the various systems and methods described herein;
• FIG. 8 illustrates an example acoustic sensor that can be used in the various systems described herein;
• FIGS. 9A through 9F illustrate embodiments of calculating arterial PWTT;
• FIG. 10 illustrates an embodiment of a process for triggering an occlusive blood pressure measurement;
• FIG. 11 illustrates plots of PWTT and heart rate waveforms;
• FIG. 12 illustrates an embodiment of a dynamic PWTT averaging system.
• FIG. 13 illustrates an embodiment of front end circuitry that can be used in the parameter calculation systems described herein to reduce phase impact on the calculations of PWTT;
• FIGS. 14A and 14B illustrate an embodiment of a process for calibrating PWTT measurements based on an individualized patient calibration factor.
DETAILED DESCRIPTION
• The propagation time of an arterial pulse wave from the heart to an extremity is related to blood pressure. Currently available blood pressure monitoring systems estimate this propagation time by detecting a time difference between points on an electrocardiograph (ECG) waveform and a photoplethysmograph waveform. This estimated propagation time is sometimes referred to as pulse wave transit time (PWTT) or time difference of arrival (TDOA). Currently available blood pressure monitoring systems trigger an automatic occlusive cuff to take a blood pressure measurement based on detected changes in PWTT. When the PWTT has not changed substantially, the blood pressure monitoring system usually does not trigger an occlusive blood pressure measurement. As a result, such a system automatically adjusts the frequency of occlusive blood pressure measurements to obtain better data while potentially reducing discomfort for the patient.
• A drawback with existing blood pressure systems is that PWTT as calculated by these systems is not always an accurate indicator of blood pressure or changes in blood pressure. One reason for this inaccuracy is that currently-available systems do not account for a patient's pre-ejection period (PEP) when computing PWTT. The PEP can include the difference in time between an electrical trigger that initiates ventricular contraction (e.g., as detected by an ECG sensor) and the actual ejection of blood from the ventricles into the aorta. Accordingly, the calculated PWTT does not accurately represent the actual propagation time of the arterial pulse from the heart to an extremity, which can result in inaccuracy in the blood pressure variability measurements.
• Another reason for the inaccuracy of existing blood pressure systems is that the photoplethysmograph waveform is typically obtained from an optical sensor coupled to a finger of the patient. Studies have shown that pulse wave velocity slows greatly at the transition from the arteries to the smaller vessels and capillaries, adding considerable delay to the arterial PWTT. This time delay can account for up to 50% of the arterial PWTT. The use of an acoustic sensor positioned over an artery to monitor arterial pulse instead of capillary flow can advantageously remove the effect of the artery-to-capillary transition delay component of the arterial pulse wave transit time measurements.
• This disclosure describes, among other features, a system for non-invasively determining an indication of an individual's blood pressure. In certain embodiments, the system dynamically accounts for a patient's PEP when calculating PWTT. The system can include an optical sensor that obtains plethysmograph information from a patient. The system can further include an electrical sensor, which can be any sensor that obtains information relating to the electrical activity of the patient's heart (such as an ECG sensor). In addition, the system can include another sensor, such as an acoustic sensor or a bioimpedance sensor, which can obtain information about cardiac ejections from the patient's heart. In certain embodiments, the system further includes a processor that calculates PWTT compensated for PEP using the information obtained from the optical, electrical, acoustic, and/or bioimpedance sensors. The system can use this compensated PWTT to determine whether to trigger an occlusive cuff measurement.
• In some embodiments, the system determines an arterial pulse wave transit time measurement from features of two acoustic waveforms. The system can include an acoustic heart sounds sensor that obtains heart sound information from a patient. The system can further include an acoustic pulse sensor, which can be placed at a location remote from the patient's heart at which peripheral arterial pulse pressure wave vibrations can be monitored (such as at a patient's wrist or neck). In addition, the system can include one or more other sensors, such as a second acoustic pulse sensor, ECG sensors, optical sensors, and/or bioimpedance sensors. In certain embodiments, the system further includes a processor that calculates arterial PWTT using the information obtained from the acoustic and/or other sensors. The system can use the arterial PWTT measurements to estimate changes in blood pressure and to aid in determining whether to trigger an occlusive cuff measurement.
• In some embodiments, the system also compensates PWTT data for noise. For example, the system can use a noise reference signal to reduce noise in the PWTT data. The noise reference signal can be derived from the patient's heart rate. The system can reduce noise in the PWTT, for example, by adaptively reducing the noise based on the noise reference signal or by dynamically adjusting an averaging time used to average the PWTT data, among other possible techniques.
• Moreover, in certain embodiments, the non-invasive blood pressure measurement system also calibrates PWTT data based on an individualized patient calibration factor. A calibration function or curve can be determined that maps PWTT measurements to blood pressure values. The slope of the calibration curve can be determined experimentally and can vary greatly from patient to patient. In some embodiments, the system determines an individualized, or personalized, patient calibration factor based on the determined slope. The patient calibration factor can then be used to interpret subsequent PWTT measurements to estimate changes in blood pressure. The individualized patient calibration can advantageously reduce the occurrence of unnecessary blood pressure cuff measurements and/or false alarms.
System Overview
• FIG. 1 illustrates an embodiment of aparameter calculation system100. In certain embodiments, theparameter calculation system100 non-invasively obtains an indication of changes in a patient's blood pressure. Theparameter calculation system100 can use the measured blood pressure changes to trigger ablood pressure cuff108 to obtain an occlusive blood pressure measurement. Advantageously, in certain embodiments, theparameter calculation system100 uses the arterial pulse wave transit time, which accounts for a patient's pre-ejection period (PEP) when calculating changes in blood pressure. Theparameter calculation system100 can therefore more accurately determine when an occlusive blood pressure measurement is appropriate.
• In the depicted embodiment, theparameter calculation system100 includes aparameter calculator110 and adisplay120. Theparameter calculator110 can include hardware (such as one or more processors), software, and/or firmware for measuring a physiological parameter such as blood pressure. Inputs to theparameter calculator110 can include, among others, optical sensor data provided by anoptical sensor102, acoustic sensor data provided by one or moreacoustic sensors104, and/or additional sensor data provided by one or moreadditional sensors106. Theoptical sensor102 can be a pulse oximetry sensor, a co-oximetry sensor, or the like. Theacoustic sensors104 can be biological sound sensors. The biological sounds may include heart, breathing, and/or digestive system sounds, in addition to many other physiological phenomena. Theadditional sensors106 can include any sensing device that provides physiological data to theparameter calculator110. For example, theadditional sensors106 can include an electrical sensor configured to provide an ECG signal, an acoustic sensor, and/or a bioimpedance sensor, or any other sensing device.
• Theoptical sensor102 can use spectrophotometry techniques to measure a variety of blood constituents, including for example, oxygen saturation, hemoglobin, methemoglobin, carboxyhemoglobin, other hemoglobin species, concentrations of the same, and the like. In addition, theoptical sensor102 can also be used to measure a variety of other physiological parameters, including pulse rate, perfusion, and the like. Theoptical sensor102 can include one or more emitters that shine one or more wavelengths of light through tissue of a living person, such as through a finger, toe, or foot. One or more detectors can receive the transmitted light after attenuation by the tissue and can generate one or more signals responsive to the attenuated light.
• In certain embodiments, theparameter calculator110 derives a photoplethysmograph from the optical sensor data The photoplethysmograph (sometimes referred to herein as a “plethysmograph,” “photopleth” or “pleth”) can be a waveform that can represent changes in blood volume as measured by one or more wavelengths of light irradiated at a tissue site of a patient. The photoplethysmograph can be caused by arterial pulsation, and as such, can be related to arterial blood pressure. Thus, in some embodiments, theparameter calculator110 uses the optical sensor data to derive an indication of blood pressure for a patient.
• In one embodiment, theparameter calculator110 can use the optical sensor data and theadditional sensor data106 to derive one or more indications of blood pressure. For example, a combination of the optical sensor data and data from electrical sensor(s), acoustic sensor(s)104, and/or bioimpedance sensors(s) can be used to determine an amount of time that it takes for a pulse to travel through an artery from a patient's heart to a measurement site. This time can be referred to as an arterial pulse-wave transit time (a-PWTT). Advantageously, in certain embodiments, theparameter calculator110 can more accurately determine the a-PWTT based at least in part on theadditional sensor data106 obtained from an acoustic (104) and/or bioimpedance sensor. In particular, using theadditional sensor data106, theparameter calculator110 can account for a patient's cardiac pre-ejection period (PEP) when calculating a-PWTT. In other embodiments, theparameter calculator110 uses the acoustic sensor data derived fromacoustic sensors104 alone to derive one or more indications of blood pressure. Using the estimated changes in blood pressure, theparameter calculator110 can trigger ablood pressure cuff108 to obtain an occlusive blood pressure measurement.
• Theparameter calculator110 canoutput parameter data113 indicative of the calculated parameters, including blood pressure, for presentation to a user. Theparameter data113 can be displayed on adisplay device120. In another embodiment, theparameter calculator110 provides parameter values as anoutput112 to another device, for example, a device providing an audible response, or over a network to a remote device. For example, a remote device might be a computer located at a nurses' station or a clinician's handheld device.
• Theparameter calculator110 can also calculate trend data reflecting trend information for theparameter data113. Theparameter calculator110 can also synthesize or scale waveform data. In addition to outputting theparameter data113, theparameter calculator110 can output trend data114, synthesized, scaled, oractual waveforms115,calibration data116, and alarms117. Theparameter calculator110 can provide theoutputs113,114,115,116 to thedisplay120, to a separate patient monitoring device, or to another device configured to receive physiological parameter information.
• In an embodiment, theparameter calculator110 is implemented in a single monitoring device. In an embodiment, the features of theparameter calculator110 are distributed among separate devices. In an embodiment, theparameter calculator110 includes a processor, processor board, or an Original Equipment Manufacture (OEM) board. In an embodiment, theparameter calculator110 is portable. Data communicated between the various components of theparameter calculation system100 can be communicated through cables or wirelessly. Other inputs and/or outputs can be included with thesystem100.
• Thedisplay120 of theparameter calculation system100 can be part of a patient monitor (not shown), which can also include other components, such as a speaker, power button, removable storage or memory (e.g., a flash card slot), an AC power port, and one or more network interfaces, such as a universal serial bus interface or an Ethernet port. Thedisplay120 can indicate a measurement for blood pressure, for example, a measurement of the systolic and diastolic blood pressure in mmHg. Other physiological parameter values, waveforms, trend data, calibration data, alarms, and the like can also be output on thedisplay120.
• Although theparameter calculator110 is described as calculating changes in blood pressure, in some embodiments, theparameter calculator110 calculates actual blood pressure values from the acoustic and/or additional sensor data. In addition, theparameter calculation system100 can also measure other physiological parameters besides blood pressure, such as pulse rate, oxygen saturation (SpO₂), hemoglobin, total hemoglobin, hemoglobin species (e.g., methemoglobin, carboxyhemoglobin, or the like), carbon monoxide or dioxide, perfusion, and glucose, among a variety of other parameters.
• Further, in some embodiments, theparameter calculator110 uses acoustic sensor data (from the acoustic sensor(s)104) to determine any of a variety of respiratory parameters of a patient, including respiratory rate, inspiratory time, expiratory time, inspiratory to expiratory (I:E) ratio, inspiratory flow, expiratory flow, tidal volume, minute volume, apnea duration, hypopnea duration, breath sounds (e.g., rales, rhonchi, and stridor), and changes in breath sounds such as decreased volume or change in airflow (either increase or decrease). In addition, in some cases theparameter calculator110 monitors other physiological sounds from theacoustic sensor104 data, such as heart rate (e.g., to help with probe off detection), heart sounds (S1, S2, S3, S4, and murmurs), and change in heart sounds such as normal to murmur or split heart sounds indicating fluid overload. Moreover, a secondacoustic sensor104 may be placed over the chest for better heart sound detection. Theparameter calculator110 may keep user inputs to a minimum (example, height) and use a Health Level 7 (HL7) interface to automatically input patient demography.
Example PWTT Calculations
• FIG. 2A depicts an embodiment of aplot200A that illustrates an example PWTT calculation. In theplot200A, aplethysmograph waveform210 and anECG waveform220 are shown. Theplethysmograph waveform210 can be obtained from anoptical sensor102 as described above. TheECG waveform220 can be obtained from an electrical sensor.
• Theplethysmograph waveform210 can reflect changes in pulsatile flow of blood in body tissue of a patient. TheECG waveform220 can reflect electrical activity of a patient's heart. TheECG waveform220 can have features including, for example, a Q-wave, an R-wave peak, and an S-wave, among others. A segment of theECG waveform220 from the Q point to the S point can be referred to as a QRS complex, which can represent ventricular activation.
• Velocity of a blood pressure wave in the arteries has a correlation with blood pressure. As the length of an artery is typically constant or approximately constant, the time that it takes for the blood pressure wave to travel from the heart to an extremity can be used to derive an indication of blood pressure. In currently available monitoring systems, a measure of such time, referred to as PWTT, has been used to infer changes in blood pressure. In some embodiments, the PWTT can represent a difference in time between a feature of theplethysmograph waveform210 and a feature of theECG waveform220. For example, in one embodiment, the PWTT can be obtained from the difference in time between the R-wave peak on theECG waveform220 and a foot point on theplethysmograph waveform210. The foot point of theplethysmograph waveform210 can correspond to the time of earliest onset of arrival of the pulse at a location away from the heart (e.g., at a patient's finger). In other embodiments, the PWTT can be obtained from the difference in time between either the Q-wave or the S-wave on theECG waveform220 and a feature on the plethysmograph waveform210 (e.g., a foot point, peak, or some other feature).
• Calculated in this manner from the plethysmograph andECG waveforms210,220, PWTT can be fairly accurate for some patients. However, in some cases, using PWTT to measure changes in blood pressure provides inaccurate or unexpected results. At least a partial explanation of these unexpected results can be seen inFIG. 2B.
• InFIG. 2B, anexample plot200 is shown that includes aplethysmograph waveform210 and anECG waveform220. Anoverall PWTT212 can be calculated from a point on theECG waveform220 to a point of theplethysmograph waveform210, as described above. However, theoverall PWTT212 can actually include at least two components—an arterial pulse wave transit time (sometimes referred to as “a-PWTT”)214 and a pre-ejection period (PEP)216.
• ThePEP216 can be defined in different ways. For instance, in certain embodiments, thePEP216 includes the difference in time between ventricular contraction and cardiac ejection of blood into the aorta. ThePEP216 can also be considered as a measured interval from the onset of ventricular depolarization, such as a Q-wave (or other feature) of an ECG, to the beginning of mechanical contraction of the heart muscle. For example, thePEP216 can represent the difference in time from the onset of the QRS complex of the ECG signal220 to when cardiac ejection actually occurs. Further, thePEP216 can also be considered as the time interval from the beginning of the electrical activation of the ventricles to the opening of the aortic valve.
• The value of thePEP216 can fluctuate based on patient condition, age, sex, and medications taken by the patient, among possibly other factors. In some patients, thePEP216 can account for a significant portion of thePWTT212, including even as much as about 50% of thePWTT212. BecausePEP216 can account for a significant portion ofoverall PWTT212, including thePEP216 in a PWTT measurement can result in inaccurate determinations of changes in a patient's blood pressure. In certain circumstances, this can even lead to not detecting a clinically significant change in blood pressure and not initiating an occlusive cuff measurement for confirmation.FIG. 2B also illustrates that it can prove difficult to derive thePEP216 from a feature of a plethysmograph signal and a feature of an ECG signal.
• Thus, in certain embodiments, thearterial PWTT214, which accounts for thePEP216, can more accurately correlate with changes in a patient's blood pressure than theoverall PWTT212. Thus, it can be advantageous to detect changes in the arterial PWTT214 and to use these changes to trigger occlusive cuff measurements. Thearterial PWTT214 can be determined in some embodiments by calculating thePEP216 and subtracting thePEP216 value from theoverall PWTT212.
• ThePEP216 can be derived, at least in part, from another physiological signal. Such a physiological signal can be indicative of cardiac ejection. Example physiological signals can include, but are not limited to, bioimpedance signals and acoustic signals. For example, in one embodiment, thePEP216 can be derived from a feature of theECG waveform220 and a feature of another physiological signal. As another example, in another embodiment, thePEP216 can be derived from a feature of theplethysmograph waveform210 and a feature of another physiological signal. As yet another example, thePEP216 can be accounted for by deriving arterial PWTT directly from a feature of an acoustic heart sound waveform and an acoustic waveform on an extremity, such as the hand, wrist, or limb.
• FIG. 2C depicts an embodiment of aplot200C that illustrates an example PWTT calculation that compensates for PEP using multiple acoustic sensors. In theplot200C, a firstacoustic waveform230 and a secondacoustic waveform240 are shown. The firstacoustic waveform220 can be obtained from a first acoustic sensor positioned proximate a heart of a patient to monitor heart sounds of the patient. The heart sounds can be indicative of the closing of heart valves: the atrioventricular valves (mitral valve and tricuspid valve) and the semilunar valves (aortic valve and pulmonary valve). The closing of the heart valves corresponds to ventricular systole and diastole. The secondacoustic waveform240 can be obtained from a second acoustic sensor positioned at an arterial location away from the heart and configured to monitor peripheral pulse pressure wave vibrations or sounds at the arterial location. For example, the second acoustic sensor can be positioned proximate a wrist artery (e.g., radial artery, ulnar artery), proximate a carotid artery on the neck of the patient, or proximate an artery of the leg.
• The firstacoustic waveform230 can include heart sounds of the patient corresponding to the closing of heart valves at the transition between ventricular systole and diastole. For example, the heart sounds can include first heart sounds (e.g., S1 heart sounds)232 corresponding to the closure of the atrioventricular valves at the time ventricular systole begins and ventricular diastole ends and second heart sounds (e.g., S2 heart sounds)234 corresponding to the closure of the aortic valve and the pulmonary valve at the time ventricular systole ends and ventricular diastole begins. The occurrence of the S1 heart sound can identify the start time of ejection of blood from the heart and the occurrence of the S2 heart sound can mark the end time or approximate end time of ejection of blood from the heart. Accordingly, the actual ejection of blood from the ventricles may occur for patients at the first heart sound (e.g., at the start, peak, or end of the first heart sound), between the start of the first heart sound (the S1 sound) and the start of the second heart sound (the S2 sound), or at the second heart sound (e.g., start, peak, or end of the second heart sound).
• The secondacoustic waveform240 can include an arterial pulse at a second location remote from the heart (e.g., at a patient's wrist or neck). The secondacoustic waveform240 shown inFIG. 2C includes audio information of the peripheral arterial pulse at a patient's wrist, hand, or arm (or foot, ankle, or leg). For convenience, although the peripheral arterial pulse can be detected at a variety of locations on the body, the wrist will be used as an illustrative example for the remainder of this specification. The secondacoustic waveform240 may display a pressure wave received at the patient's wrist some time prior to actual arrival of the blood at the periphery.
• In some embodiments, the arterial PWTT can represent a difference in time between a feature of the first acoustic waveform (e.g., heart sounds waveform) and a feature of the second acoustic waveform (e.g., wrist pulse waveform). For example, in one embodiment, the arterial PWTT can be obtained from the difference in time between a feature of the first heart sound (the S1 sound)232 on the acoustic heart sounds waveform240 (e.g., start, maximum peak, end, some other feature of the S1 sound242) and a feature of the acoustic wrist pulse waveform220 (e.g., bottom onset or upstroke point246 of thewaveform220 or start, maximum peak, end, some other feature of the waveform220). In other embodiments, the arterial PWTT can be obtained from the difference in time between a determined centroid location of theS1 sound232 or a location of the centroid of the energy from the start of theS1 sound232 until the start of theS2 sound234 and a feature on the acoustic wrist pulse waveform (e.g., a bottom onset, a foot point, a peak, or some other feature). In some embodiments, the envelope of thewaveform220 is obtained and used in the analysis described herein, e.g., by finding a feature of the envelope rather than thewaveform220 itself.
• In one embodiment, the arterial PWTT can be obtained from the difference in time from a location between the S1 and S2 heart sounds232,234 and/or between a centroid of the energy of the S1 and S2 sounds232,234 and the bottom onset of a corresponding pulse of an acoustic arterial pulse waveform (e.g., an acoustic wrist pulse waveform or an acoustic carotid pulse waveform). In some embodiments, an arterial PWTT determination based on identified features of two mechanical acoustic waveforms advantageously provides a more stable result than a PWTT determination based on an identified feature of an electrical waveform (e.g., ECG waveform) and an identified feature of a mechanical waveform (e.g., photoplethysmograph waveform).
• Turning toFIG. 2D, aplot200D is shown that includes a set of four example waveforms. The acoustic heart soundswaveform230 and the acousticwrist pulse waveform240 are illustrated again, along with a third acoustic waveform (an acoustic carotid pulse waveform)250 and anECG waveform260. Theplot200D helps to illustrate the two components that make up an overall PWTT measurement in certain embodiments: a pre-ejection period (PEP) component and an arterial transit time component. The use of two acoustic sensors results in PWTT or time-difference-of-arrival calculations that, in certain embodiments, more accurately reflect arterial pulse wave transit time, thereby allowing blood pressure measurements to be taken more efficiently.
• The acousticcarotid pulse waveform250 can help to provide another measurement indicator (e.g., a control or reference or another input component) that can be used to determine the arterial pulse wave transit time or other arterial properties because the distance is known between the two sensors. The acousticcarotid pulse waveform250 or any of the other waveforms can provide an indication of patient breathing, as shown by the presence ofnoise255 on the acousticcarotid pulse waveform250. In some implementations, the acousticcarotid pulse waveform230 can be used in place of thewrist pulse waveform240 to determine the arterial PWTT.
• In some embodiments, the PEP is derived from one or more features of one or more acoustic signals. In one embodiment, the PEP is determined from the time of the start of theS1 heart sound232 to the time of the end of theS1 heart sound232. In other embodiments, the PEP can be derived from a feature of anECG waveform260 and a feature of another physiological signal (e.g., a pleth signal or an acoustic signal). One way to measure PEP is to determine the difference between an R wave peak of theECG waveform260 and a feature of theS1 sound212 on the acoustic heart sounds waveform230 (e.g., a foot point, peak, centroid, or other feature or derived location). More generally, PEP can be determined as a difference in time between any feature of theECG waveform260 and any feature of the acoustic heart soundswaveform230, including in some embodiments, a feature of theS2 heart sound234. The PEP can vary depending on patient pathology, and a determination of PEP can provide information regarding heart conditions to a clinician. In some embodiments, PEP is not calculated or used at all by theparameter calculation system100 because the arterial PWTT is determined by the acoustic sensors.
• FIG. 3 illustrates an embodiment of a bloodpressure monitoring system300 that can determine PWTT measurements (including a-PWTT measurements) from signals received from various sensors. The bloodpressure monitoring system300 can implement certain features of theparameter calculation system100 described above. In particular, thesystem300 can periodically measure blood pressure using ablood pressure cuff320, which can be an automatic occlusive cuff or the like. In addition, thesystem300 can perform PWTT calculations to noninvasively detect changes in a patient's blood pressure. The illustrated bloodpressure monitoring system300 includes sensors and associated modules that can advantageously be used to monitor blood pressure. The depicted modules can be implemented in hardware and/or in software (e.g., as executed by one ormore processors330 or computing devices).
• In the depicted embodiment, thesystem300 includes aparameter calculator310, which can be implemented in hardware and/or software. Theparameter calculator310 is a more detailed implementation of theparameter calculator110 ofFIG. 1 and can include all the features thereof. Various sensors communicate with theparameter calculator310. These sensors include two or moreacoustic sensors302, one ormore ECG sensors304, one ormore bioimpedance sensors306, and one or moreoptical sensors308. Theacoustic sensors302 can include piezoelectric transducers or other acoustic transducers for measuring a patient's body sounds, such as breathing and heart sounds. The ECG sensor(s)304 can include ECG leads or the like for measuring the electrical activity of the heart. The bioimpedance sensor(s)306 can include electrodes placed on the neck and/or thorax for measuring the impedance of electrical signals in the body. More detailed embodiments of these sensors are described below with respect toFIGS. 5-8.
• Theexample parameter calculator310 shown includes ablood pressure analyzer312 and a noninvasiveblood pressure module314. Theblood pressure analyzer312 can calculate arterial PWTT using the outputs of some or all of thevarious sensors302,304,306, and308. Based at least in part on this calculated PWTT, theblood pressure analyzer312 can send a trigger signal to the noninvasiveblood pressure module314. In response to receiving this trigger signal, the non-invasiveblood pressure module314 can cause theblood pressure cuff320 to take a blood pressure measurement. In some embodiments, the non-invasiveblood pressure module314 is a separate component from theparameter calculator310, for example, on an Original Equipment Manufacture (OEM) board or the like.
• In one embodiment, theacoustic sensor302 is placed over the heart or near the heart of a patient so as to detect heart sounds of the patient. Theacoustic sensor302 can be positioned on the chest, back, neck, side, abdomen, or other area of the body so as to detect the heart sounds. Heart sounds can include, among others, first and second heart sounds. The first heart sound can correspond to systole, or the contraction of the ventricles and corresponding ejection of blood from the heart. PEP can therefore be measured as a time difference between a feature of the ECG waveform derived from the electrical sensor(s)304 and a first heart sound feature of an acoustic waveform derived from theacoustic sensor302. The second heart sound can correspond to the beginning of diastole.
• For example, referring toFIG. 4A, an exampleacoustic waveform410 andECG waveform412 are shown that illustrate one possible PEP calculation. Theacoustic waveform410 includespeaks422,424 that correspond to example first and second heart sounds, respectively. TheECG waveform412 includes a peak432 corresponding to the R wave of the QRS complex. One way to measure PEP is to determine the difference between theR wave peak432 and the firstheart sound peak422. In another embodiment, the PEP is measured as a difference between theR wave peak432 and afoot423aor423bof theacoustic waveform410. More generally, PEP can be determined as a difference in time between any feature of theECG waveform412 and any feature of theacoustic waveform410, including in some embodiments, a feature of the secondheart sound peak424.
• Referring again toFIG. 3, PWTT measurements (e.g., PEP or a-PWTT measurements) can also be calculated using the one ormore bioimpedance sensors306. Thebioimpedance sensors306 can implement principles of impedance cardiography, which can also be referred to as thoracic electrical bioimpedance. Thebioimpedance sensors306 can measure the impedance of a patient's chest cavity by injecting alternating (or direct) current through the patient's chest. The current tends to seek the path of least resistance, which is the patient's blood-filled aorta. The blood volume and velocity in the aorta can change with each heartbeat, resulting in corresponding changes in impedance measured by the bioimpedance electrodes. These changes in impedance can be used to derive PEP.
• For instance, referring toFIG. 4B, anexample bioimpedance waveform440 is shown together with theECG waveform412 ofFIG. 4A. The bioimpedance waveform includes a peak442 that corresponds to a peak change in impedance with respect to time. Thispeak442 can correspond to ejection of blood from the heart, corresponding to a current change in the aorta resulting from a heartbeat. Thus, PEP can be measured between a feature of theECG waveform412 and a feature of thebioimpedance waveform440. For instance, PEP can be measured between theR wave peak432 and afoot point444 of thebioimpedance waveform440. Thisfoot point444 is sometimes referred to as the “B” point of thebioimpedance waveform440 and corresponds to the maximum rate of change of thewaveform440. The PEP can also be measured from the R wave peak432 (or another feature of the ECG waveform412) and thepeak442 of the bioimpedance waveform.
• Thus, in certain embodiments, the acoustic and/orbioimpedance sensors302,306 ofFIG. 3 can be used in conjunction with the ECG sensor(s)304 to calculate PEP. In some implementations, only anacoustic sensor302 and ECG sensor(s)304 are used to calculate PEP. Other implementations employ only bioimpedance sensor(s)306 and ECG sensor(s)304 to calculate PEP. Still other embodiments of thesystem300 can calculate PEP using acoustic sensor(s)302 and separately calculate PEP using bioimpedance sensor(s)306. Theblood pressure analyzer312 can average or otherwise combine the PEP calculations from thesedifferent sensors302,306 in some embodiments.
• In other implementations, twoacoustic sensors302 ofFIG. 3 are used to calculate PWTT measurements (e.g., a-PWTT or PEP measurements). In some implementations, multiple techniques can be used to calculate PEP or a-PWTT. In some embodiments, theblood pressure analyzer312 can average or otherwise combine the PWTT calculations from these different sensors. In other embodiments, PWTT measurements obtained from the one or more of the sensors can be used to assess confidence in the PWTT measurements obtained from other of the sensors. Additional details will be further described below in connection withFIGS. 9A-9C.
• FIG. 5A illustrates a more detailed embodiment of a bloodpressure monitoring system500A including an acoustic sensor coupled to a patient. The bloodpressure monitoring system500A can implement certain features of the bloodpressure monitoring system300 andparameter calculation system100 described above. The illustrated bloodpressure monitoring system500A includes sensors, associated modules, and aprocessor530 that can advantageously be used to monitor blood pressure. The depicted modules can be implemented in hardware and/or in software (e.g., as executed by the processor530).
• The illustrated bloodpressure monitoring system500A is coupled to apatient510. Thepatient510 is shown with acuff522 attached to an upper arm. Thecuff522 can be implemented in combination with an automatic occlusivecuff control unit520. Thecuff522 can be in communication with the automatic occlusivecuff control unit520 via acable524. Thecontrol unit520 can control the inflation of thecuff522 and receive signals from thecuff522 regarding systolic and diastolic blood pressure.
• In addition to the automatic occlusive cuff,ECG sensors562A and562B can be coupled to thepatient510. TheECG sensors562A,562B can provide any of the ECG signals described above. TheECG sensors562A and562B can be implemented as dual electrodes or split electrodes. While the illustrated blood pressure monitoring system includes twoECG sensors562A and562B, in other embodiments, only one ECG sensor can be coupled topatient510. In yet other embodiments, more than two ECG sensors can be coupled to thepatient510, such as three or more sensors. In addition, ECG sensors can be placed at different measurement site(s) than illustrated inFIG. 5A. For example, one or more ECG sensors could be coupled to the back of thepatient510.
• The illustratedECG sensors562A,562B can be coupled to anECG unit560 viacables564A and564B, respectively. TheECG unit560 can interface with theECG sensors562A,562B and provide an ECG signal to theprocessor530. In some embodiments, the ECG unit can convert and output ofECG sensors562A,562B from an analog signal to a digital signal and/or perform other pre-processing. TheECG unit560 can be implemented separate from theprocessor530 or alternatively as part of theprocessor530.
• Anoptical sensor552 can also be coupled to thepatient510. Theoptical sensor552 can provide any of the plethysmograph waveforms illustrated inFIGS. 2A and 2B and/or theoptical sensor data104 described above in connection withFIG. 1. The illustratedoptical sensor552 can be coupled tooximeter unit550 via cable554. Theoximeter unit560 can interface with theoptical sensor552 and provide an optical signal to theprocessor530. In some embodiments, theoximeter unit550 can convert and output ofoptical sensor552 from an analog signal to a digital signal and/or perform other pre-processing. Theoximeter unit550 can be implemented separate from theprocessor530 or alternatively as part of theprocessor530. It should be noted that in certain embodiments, theoptical sensor552 can provide data to a monitor other than a pulse oximeter.
• Anacoustic sensor572 can also be coupled to thepatient510. Theacoustic sensor552 can provide any of the acoustic signals and waveforms described above. The illustratedacoustic sensor572 is coupled to thepatient510 at a measurement site near the patient's heart. In other embodiments, the acoustic sensor can be coupled to thepatient510 at different measurement sites, so long as the acoustic sensor can provide useful information indicative of cardiac ejection. In other embodiments, more than oneacoustic sensor572 can be used. More detail regarding theacoustic sensor572 will be provided below in connection withFIG. 6.
• Theacoustic module570 can be coupled to theacoustic sensor572 via cable574. Theacoustic module570 can interface with theacoustic sensor572 and provide an acoustic signal to theprocessor530. In some embodiments, the acoustic module can convert and output of theacoustic sensor572 from an analog signal to a digital signal and/or perform other pre-processing. Theacoustic module570 can be implemented separate from theprocessor530 or alternatively as part of theprocessor530.
• Theprocessor530 can advantageously be used to measure indicators of blood pressure, such as PWTT, PEP, and/or arterial PWTT, using information provided by theECG unit560, theoximeter unit550, and theacoustic module570. Theprocessor530 can include, for example, one or more microprocessors, microcontrollers, cores, digital signal processors (DSPs), or the like. Theprocessor530 can store instructions in a computer-readable medium. Theprocessor530 can also perform other operations for the bloodpressure monitoring system500A that are not explicitly described herein.
• Theprocessor530 can also be coupled to the automatic occlusivecuff control unit520 vialine532. Theprocessor530 can receive information regarding a blood pressure measurement of the patient510 from the automatic occlusivecuff control unit520, and can activate the automatic occlusivecuff control unit520 via theline532.
• In certain embodiments, theprocessor530 determines the arterial PWTT, compensated for PEP, using theacoustic sensor572, theECG sensors562A, and theoptical sensor552. Theprocessor530 can track changes in this arterial PWTT and trigger the automatic occlusivecuff control unit520 to take a blood pressure measurement with thecuff522. The PEP-compensated arterial PWTT calculations can be more accurate than currently-available PWTT calculations that do not take PEP into account. Thus, in certain embodiments, the bloodpressure monitoring system500A can more effectively monitor the patient's510 blood pressure with potentially greater comfort for thepatient510.
• FIG. 5B illustrates another embodiment of a blood pressure monitoring system500B including bioimpedance sensors582 coupled to thepatient510. The blood pressure monitoring system500B can implement certain features of the bloodpressure monitoring system300 andparameter calculation system100 described above. The illustrated blood pressure monitoring system500B can be substantially similar to the bloodpressure monitoring system500A ofFIG. 5A. However,bioimpedance sensors582A,582B,582C,582D and abioimpedance module580 are included in place of theacoustic sensor572 and theacoustic module570.
• Thebioimpedance sensors582A,582B,582C,582D can be coupled to thepatient510 at various measurement sites. For example, as illustrated, twobioimpedance sensors582A and582B can be coupled to the sides of thepatient510 and twobioimpedance sensors582C and582D can be coupled to the neck of thepatient510. Other suitable measurement sites can be used in other embodiments. Thebioimpedance sensors582A,582B,582C,582D can be used to provide the bioimpedance waveforms described above with respect toFIG. 4B and/or any of theadditional sensor data106 described above in connection withFIG. 1. Suitable bioimpedance sensors582 can include electrodes or the like.
• Thebioimpedance unit580 can be coupled to one or more of thebioimpedance sensors582A,582B,582C,582D via one ormore cables584A,584B,584C,584D, respectively. Thebioimpedance unit580 can interface with one or more of thebioimpedance sensors582A,582B,582C,582D and provide one or more bioimpedance signals to theprocessor530. In some embodiments, thebioimpedance unit580 can convert an output of one or more of thebioimpedance sensors582A,582B,582C,582D from an analog signal to a digital signal and/or perform other pre-processing. Thebioimpedance unit580 can be implemented separate from theprocessor530 or alternatively as part of theprocessor530.
• In certain embodiments, theprocessor530 determines the arterial PWTT, compensated for PEP, using the bioimpedance sensors582, theECG sensors562A,562B and theoptical sensor552. Theprocessor530 can track changes in this arterial PWTT and trigger the automatic occlusivecuff control unit520 to take a blood pressure measurement with thecuff522. The PEP-compensated arterial PWTT calculations can be more accurate than currently-available PWTT calculations that do not take PEP into account. Thus, in certain embodiments, the blood pressure monitoring system500B can more effectively monitor the patient's510 blood pressure with potentially greater comfort for thepatient510.
• Although shown separately for ease of illustration, in certain embodiments theECG sensors562A,562B and the bioimpedance sensors582 can be combined. For instance, the bioimpedance sensors582 can obtain ECG data in some implementations, and vice versa.
• FIG. 6 illustrates example positioning locations for the acoustic sensors that can be used in the various systems and methods described herein (such as theparameter calculation system100 and the blood pressure monitoring system300). As described above, a firstacoustic sensor602 can be placed over the heart or near the heart of a patient so as to detect heart sounds of the patient. In some embodiments, the acoustic heart soundssensor602 can be positioned over the heart in or near the second intercostal space; however, the acoustic heart soundssensor602 can be positioned at other locations (such as the chest, back, neck, side, abdomen, or other area of the body) so as to more accurately detect particular heart sounds (e.g., the S1 and S2 sounds). In some implementations, the acoustic heart soundssensor602 can be positioned at a location at which both the S1 and S2 heart sounds can be effectively measured. In other implementations, the acoustic heart soundssensor602 can be positioned at a location configured to obtain increased signal strength for a particular heart sound without regard to other heart sounds.
• With continued reference toFIG. 6, a secondacoustic sensor604 can be placed at or near an artery on or near the wrist of a patient to measure an arterial pulse at a distance from the heart. For example, the secondacoustic sensor604 can be placed over the ulnar artery or the radial artery to obtain information indicative of a wrist pulse. A thirdacoustic sensor606 can be placed on a neck of a patient at a location over or near a carotid artery to obtain information indicative of a carotid pulse. In some implementations, all three acoustic sensors are used. In other implementations, only the acoustic heart soundssensor602 and one of the other two sensors are used. The sensors ofFIG. 6 can be coupled to a parameter calculation system (e.g., theparameter calculation system100 or the blood pressure monitoring system300).FIG. 6 merely illustrates example locations for the types of sensors that can be used in various implementations; other suitable locations may also be used.
• FIG. 7 illustrates example positioning locations for acoustic, electrocardiograph (ECG), bioimpedance and optical sensors that can be used in the various systems and methods described herein (such as the blood pressure monitoring system300). Theacoustic sensors702A-702C are illustrated as circles, theECG sensors704A-704C are illustrated as squares, and thebioimpedance sensors706A-706D are illustrated as triangles.FIG. 7 also illustrates anoptical sensor708 and an occlusiveblood pressure cuff710. The sensors ofFIG. 7 can be coupled to a blood pressure monitoring system (e.g., the blood pressure monitoring system300). AlthoughFIG. 7 illustrates multiple sensors and multiple positioning locations, all of the sensors and/or positioning locations need not be used;FIG. 7 merely illustrates example locations for the types of sensors that can be used in various implementations. The number, locations, and type of sensors used can vary.
• Theblood pressure cuff710 can be attached to an upper arm. Thecuff710 can be implemented in combination with an automatic occlusive cuff control unit (not shown). Thecuff710 can be in communication with the automatic occlusive cuff control unit via a cable and/or hose. The control unit can control the inflation of thecuff710 and receive signals from thecuff710 regarding systolic and diastolic blood pressure.
• Theacoustic sensors702A-702C can be coupled to an acoustic module (not shown) via one or more cables or wirelessly. The acoustic module can interface with theacoustic sensors702A-702C and provide acoustic signals to theprocessor330. In some embodiments, the acoustic module can convert an output of theacoustic sensors702A-702C from an analog signal to a digital signal and/or perform other pre-processing. The acoustic module can be implemented separate from theprocessor330 or alternatively as part of theprocessor330 or even as part of one or more of thesensors702A-702C.
• TheECG sensors704A-704C can provide any of the ECG signals described herein. TheECG sensors704A-704C can be implemented as dual electrodes or split electrodes. While thepatient700 inFIG. 7 has threeECG sensors704A-704C, in other embodiments, only one ECG sensor or two ECG sensors can be coupled to thepatient700. In yet other embodiments, more than three ECG sensors can be coupled to thepatient700, such as four or more sensors. In addition, ECG sensors can be placed at different measurement site(s) than illustrated inFIG. 7. For example, one or more ECG sensors could be coupled to the back of thepatient700.
• The illustratedECG sensors704A-704C can be coupled to an ECG unit (not shown) via cables or wirelessly. The ECG unit can interface with theECG sensors704A-704C and provide an ECG signal to aprocessor330. In some embodiments, the ECG unit can convert an output ofECG sensors704A-704C from an analog signal to a digital signal and/or perform other pre-processing. The ECG unit can be implemented separate from theprocessor330 or alternatively as part of theprocessor330.
• Thebioimpedance sensors706A-706D can be coupled to thepatient700 at various measurement sites. For example, as illustrated, twobioimpedance sensors706A and706B can be coupled to the sides of thepatient700 and twobioimpedance sensors706C and706D can be coupled to the neck of thepatient700. Other suitable measurement sites can be used in other embodiments. Thebioimpedance sensors706A-706D can be used to provide bioimpedance waveforms that can be used in determining measurements indicative of blood pressure (e.g., overall PWTT, PEP, arterial PWTT).Suitable bioimpedance sensors706A-706D can include electrodes or the like.
• Thebioimpedance sensors706A-706D can be coupled to a bioimpedance unit (not shown). The bioimpedance unit can interface with one or more of thebioimpedance sensors706A-706D and provide one or more bioimpedance signals to theprocessor330. In some embodiments, the bioimpedance unit can convert an output of one or more of thebioimpedance sensors706A-706D from an analog signal to a digital signal and/or perform other pre-processing. The bioimpedance unit can be implemented separate from theprocessor330 or alternatively as part of theprocessor330.
• Although shown separately for ease of illustration, in certain embodiments theECG sensors704A-704C and thebioimpedance sensors706A-706D can be combined. For instance, thebioimpedance sensors706A-706D can obtain ECG data in some implementations, and vice versa.
• Anoptical sensor708 can also be coupled to the patient700 (e.g., to a patient's finger). The illustratedoptical sensor708 can be coupled to an oximeter unit (not shown) via a cable or wirelessly. The oximeter unit can interface with theoptical sensor708 and provide an optical signal to aprocessor330. In some embodiments, the oximeter unit can convert an output ofoptical sensor708 from an analog signal to a digital signal and/or perform other pre-processing. The oximeter unit can be implemented separate from theprocessor330 or alternatively as part of theprocessor330. It should be noted that in certain embodiments, theoptical sensor708 can provide data to a monitor other than a pulse oximeter.
• Theprocessor330 ofFIG. 3 can advantageously be used to measure indicators of blood pressure, such as PWTT, PEP, and/or arterial PWTT, using information provided by the acoustic module, the ECG unit, the oximeter unit, and/or the bioimpedance unit. Theprocessor330 can include, for example, a microprocessor, microcontroller, a core, a digital signal processor (DSP), or the like. Theprocessor330 can store instructions in a computer-readable medium. Theprocessor330 can also perform other operations for the bloodpressure monitoring system300 that are not explicitly described herein.
• Theprocessor330 can also be coupled to the automatic occlusive cuff control unit. Theprocessor330 can receive information regarding a blood pressure measurement of the patient700 from the automatic occlusive cuff control unit, and can activate the automatic occlusive cuff control unit.
• In certain embodiments, theprocessor330 determines the arterial PWTT, using two of theacoustic sensors702A-702C, independently or in conjunction with an additional acoustic sensor, one or more of theECG sensors704A-704C, one ormore bioimpedance sensors706A-706D, and theoptical sensor708. In some embodiments, theprocessor330 determines the arterial PWTT using a combination of one or more acoustic sensors and one or more non-acoustic sensors (for example, using two bioimpedance sensors and one acoustic sensor). Theprocessor330 can track changes in this arterial PWTT and trigger the automatic occlusive cuff control unit to take a blood pressure measurement with thecuff710. The arterial PWTT calculations obtained can be more accurate than currently-available PWTT calculations that do not take PEP into account or require subtraction of the PEP to determine the arterial PWTT.
• FIG. 8 illustrates an exampleacoustic sensor system800 that can be used in any of the blood pressure monitoring systems described herein, such as thesystems100 and300.FIG. 8 is a top perspective of asensor system800 including anacoustic sensor assembly801 suitable for use as any acoustic sensor described herein and amonitor cable811. Thesensor assembly801 can include anacoustic sensor815, acable assembly817, and aconnector805. Thesensor815, in one embodiment, can include asensor subassembly802 and anattachment subassembly804. Thecable assembly817 of one embodiment can include acable807 and apatient anchor803. The various components can be connected to one another via thesensor cable807. Thesensor connector subassembly805 can be removably attached to amonitor connector809, which can be connected to parameter calculator or other physiological monitor (not shown) via themonitor cable811. In one embodiment, thesensor assembly801 can communicate with the physiological monitor wirelessly.
• In an embodiment, thesensor assembly801 can include a sensing element, such as, for example, a piezoelectric device or other acoustic sensing device. The sensing element can generate a voltage that is responsive to vibrations generated by the patient, and the sensor can include circuitry to transmit the voltage generated by the sensing element to a processor for processing. In an embodiment, theacoustic sensor assembly801 can include circuitry for detecting and transmitting information related to biological sounds to a physiological monitor. These biological sounds may include heart, breathing, and/or digestive system sounds, in addition to many other physiological phenomena. Theacoustic sensor815 in certain embodiments can be a biological sound sensor, such as the sensors described herein. In some embodiments, the biological sound sensor is one of the sensors such as those described in the U.S. patent application Ser. No. 12/044,883, filed Mar. 7, 2008, entitled “Systems and Methods for Determining a Physiological Condition Using an Acoustic Monitor,” which is incorporated in its entirety by reference herein. In other embodiments, theacoustic sensor815 can include a biological sound sensor such as those described in U.S. Pat. No. 8,661,161, which is incorporated by reference herein. Other embodiments include other suitable acoustic sensors.
• Theattachment sub-assembly804 can include a firstelongate portion806 and a secondelongate portion808. The firstelongate portion806 and the secondelongate portion808 can include patient adhesive (e.g., in some embodiments, tape, glue, a suction device, etc.) attached to anelongate member810. The adhesive on theelongate portions806,508 can be used to secure thesensor subassembly802 to a patient's skin. Theelongate member810 can beneficially bias thesensor subassembly802 in tension against the patient's skin and reduce stress on the connection between the patient adhesive and the skin. A removable backing can be provided with the patient adhesive to protect the adhesive surface prior to affixing to a patient's skin.
• Thesensor cable807 can be electrically coupled to thesensor subassembly802 via a printed circuit board (“PCB”) (not shown) in thesensor subassembly802. Through this contact, signals can be communicated from the sensor subassembly to the physiological monitor via thesensor cable807 and thecable811.
• In various embodiments, not all of the components illustrated inFIG. 8 are included in thesensor system800. For example, in various embodiments, one or more of thepatient anchor803 and theattachment subassembly804 are not included. In one embodiment, for example, a bandage or tape is used instead of theattachment subassembly804 to attach thesensor subassembly602 to a measurement site. Moreover, such bandages or tapes may be a variety of different shapes including generally elongate, circular and oval, for example.
• FIGS. 9A, 9B, and 9C illustrate embodiments ofprocesses900A,900B, and900C for calculating PWTT values compensated for PEP. Theprocesses900A,900B,900C can be implemented as part of any of the blood pressure monitoring systems described herein, such as the bloodpressure monitoring systems100,300,500A, and500B. More particularly, theprocess900A illustrates an embodiment of obtaining compensated PWTT using information obtained from an acoustic sensor. Theprocess900B illustrates an embodiment of PWTT calculation using bioimpedance sensors. Theprocess900C illustrates an embodiment of calculating PWTT without using an ECG sensor.
• Referring specifically toFIG. 9A, plethysmograph, ECG, and acoustic waveforms are received by theprocess900A. Atblock902, an initial PWTT is calculated from the plethysmograph and ECG waveforms. This initial PWTT is determined in one embodiment by determining a time difference between features of the ECG and plethysmograph waveforms. Features on these waveforms can be determined using signal processing techniques including, but not limited to, taking derivatives of the waveforms, detecting peaks and troughs of the waveform, comparing the waveforms to models and/or thresholds, and the like. For example, an R-wave peak on an ECG waveform can be determined by detecting a point at which the ECG signal is above a (potentially adaptive) threshold and the derivative of the ECG signal is zero. As another example, a foot point on a plethysmograph waveform can be identified by the point at which the derivative of the plethysmograph waveform is zero in a time window defined after an R-wave peak. Such signal processing can be performed dynamically in real time. Alternatively or additionally, such signal processing can be performed during post processing of data.
• As described above, this initial PWTT can include the arterial PWTT as well as the PEP. Thus, to obtain the arterial PWTT, further blocks of theprocess900A can determine PEP so as to compensate the initial PWTT for the PEP. Atblock904, a first heart sound is detected from the acoustic waveform. The PEP is then calculated atblock906 by determining a time difference between a feature of the ECG waveform and a feature of the first heart sound. The feature used on the ECG waveform to calculate thePEP906 can be the same feature used to calculate the initial PWTT atblock902. Alternatively, different ECG features can be used to calculate PEP and the initial PWTT. Atblock908, the PEP is subtracted from the initial PWTT to produce an arterial PWTT value. This arterial PWTT value can be used in determining if an automatic occlusive cuff should take a blood pressure measurement from a patient (seeFIG. 10).
• Advantageously, in certain embodiments, theprocess900A can be performed continuously or substantially continuously so as to monitor a patient's arterial PWTT over time. Theprocess900A can therefore dynamically determine a patient's changing PEP over time and can compensate initial PWTT values according to this changing PEP. Theprocess900A can therefore, in certain embodiments, calculate more accurate PWTT than currently-available devices.
• It should further be noted that theprocess900A shown can also be more accurate than processes that calculate the arterial PWTT without using an ECG waveform. Because of its prominent R-wave peak, the ECG waveform can be a more reliable triggering point than cardiac ejection signals such as heart sounds or bioimpedance signals. Accordingly, the ECG signal can be used to identify relevant cardiac ejection signal. For example, the subsequent cardiac ejection signal after an ECG signal may be the next cardiac ejection signal of interest.
• Another advantage of using the ECG as a triggering signal for calculating PWTT is related to the occasional ambiguity of the cardiac ejection signal. In some cases, it may be difficult or impossible to distinguish a cardiac ejection signal from noise. In such cases, theparameter calculator110 or310 can use the previous PEP value as the current PEP value and calculate arterial PWTT as the difference between the ECG-to-arterial pulse signal and the previous PEP value. As PEP values may change infrequently, this substitution may be accurate.
• Further, because the occurrence and/or location of the cardiac ejection signal can be ambiguous, it can be beneficial to more aggressively average the PEP measurements over time to compensate for this ambiguity. For example, a relatively slower averaging filter, which may include a longer time window of data points, may be applied to the PEP values. In contrast, the overall PWTT values may change more rapidly (see, e.g.,FIG. 11), and it may be beneficial to apply a relatively faster averaging filter having a relatively shorter time window to these values.
• FIG. 9B illustrates another embodiment of aprocess900B that uses bioimpedance information to calculate arterial PWTT. Theprocess900B is substantially similar to theprocess900A, except that theprocess900B analyzes a bioimpedance waveform instead of an acoustic waveform.
• Like theprocess900A, theprocess900B receives a plethysmograph waveform and an ECG waveform. Atblock912, an initial PWTT is calculated from these waveforms in the manner described above with respect toFIG. 9A. Atblock914, the bioimpedance waveform (or impedance cardiogram) is analyzed to detect a feature, such as a foot point, a peak change in impedance, an inflection point, or the like. Atblock916, the PEP is calculated as a time difference between a feature of the ECG waveform and the feature on the bioimpedance waveform. For example, the point of maximum impedance change can be compared with a feature of the ECG waveform to determine the PEP. Atblock918, the calculated PEP value is subtracted form the initial PWTT to determine the arterial PWTT. This arterial PWTT value can be used in determining if an automatic occlusive cuff should take a blood pressure measurement from a patient (seeFIG. 10).
• Like theprocess900A, theprocess900B can be performed continuously or substantially continuously so as to monitor a patient's arterial PWTT over time. Theprocess900B can therefore dynamically determine a patient's changing PEP over time and can compensate initial PWTT values according to this changing PEP. Theprocess900B can therefore, in certain embodiments, calculate more accurate PWTT than currently-available devices.
• FIG. 9C illustrates another embodiment of aprocess900C for calculating PWTT. However, theprocess900C can advantageously calculate PWTT using an optical and acoustic signal or an optical and bioimpedance signal, without using an ECG signal. Thus, theprocess900C uses fewer sensors to calculate arterial PWTT.
• Theprocess900C receives a plethysmograph and acoustic waveform, or a plethysmograph and bioimpedance waveform. Atblock922, a feature of the plethysmograph waveform is identified. Atblock924, a feature of the acoustic or bioimpedance waveform is identified. Cardiac ejection can precede the arrival of the resulting pulse at an extremity. Thus, a feature from the plethysmograph waveform obtained at the extremity can be identified when it occurs at a later time than the identified feature from the acoustic or bioimpedance waveform.
• Atblock926, a time difference between the identified waveform features is determined. This time is the arterial PWTT. This arterial PWTT value can be used in determining if an automatic occlusive cuff should take a blood pressure measurement from a patient (seeFIG. 10).
• Like theprocesses900A and900B, theprocess900C can be performed continuously or substantially continuously so as to monitor a patient's arterial PWTT over time. Theprocess900C can therefore dynamically determine a patient's changing PEP over time and can compensate initial PWTT values according to this changing PEP. Theprocess900C can therefore, in certain embodiments, calculate more accurate PWTT than currently-available devices.
• Theprocess900C can be modified in one embodiment to receive an ECG signal input to assist with identifying the feature of the plethysmograph waveform atblock922. It can be difficult to properly identify features from the plethysmograph waveform due to the double-peak nature of the waveform, as well as due to noise. An ECG signal, on the other hand, can have clearly-identifiable landmarks, including the R wave peak, among others. An ECG signal can therefore be used as a gating function to determine which feature of the plethysmograph waveform should be considered and/or which feature of the acoustic or bioimpedance waveform should be considered. In one embodiment, the first peak, foot point, or other feature in the plethysmograph occurring after a feature identified from the ECG signal can be identified as the relevant plethysmograph feature. Similarly, the first foot peak, or other feature of the acoustic or bioimpedance waveform occurring after a feature identified from the ECG signal can be identified as the relevant cardiac ejection signal feature. The ECG signal can therefore resolve ambiguities in the plethysmograph waveform, thereby improving noise immunity and/or noise reduction. Similarly, the ECG signal can also be used as a reliable identifier of a cardiac ejection signal, thereby improving the accuracy of the arterial PWTT calculation.
• FIGS. 9D-9F illustrate additional embodiments ofprocesses900D,900E,900F for calculating arterial PWTT values, refining calculated arterial PWTT values, and/or determining confidence of arterial PWTT values. In some implementations, arterial PWTT values are determined from acoustic waveforms of signals from only two acoustic sensors, as described in connection withFIG. 9D. In other implementations, arterial PWTT values are determined using waveforms of signals from three or more acoustic sensors (as described in connection withFIG. 9E) or from one or more non-acoustic sensors, in addition to the two acoustic sensors (as described in connection withFIG. 9F). Each of the depicted blocks or modules of the processes900 can be implemented by hardware and/or software modules.
• With reference toFIG. 9D, theprocess900D can advantageously calculate arterial PWTT using two acoustic signals, without using an ECG signal. Thus, theprocess900D uses two mechanical signals rather than one electrical signal and one mechanical signal, thereby improving stability and accuracy of the arterial PWTT calculations. The use of the two acoustic signals also removes the necessity to subtract out the PEP component of the overall PWTT and instead reflects the arterial PWTT alone.
• Theprocess900D receives an acoustic heart sounds waveform and an acoustic pulse waveform. The acoustic pulse waveform can be, for example, an acoustic wrist pulse waveform or an acoustic carotid pulse waveform, as described above; however, the acoustic pulse waveform can be received from an acoustic sensor positioned at other locations as well. Atblock927, a feature of the acoustic heart sounds waveform is identified. Atblock928, a feature of the acoustic pulse waveform is identified. The feature of the acoustic heart sounds waveform can correspond to the time of cardiac ejection, which precedes the time of arrival of the resulting pulse at a location of the acoustic sensor from which the acoustic pulse waveform is generated.
• Features on these waveforms can be determined using signal processing techniques including, but not limited to, taking derivatives of the waveforms, detecting peaks and troughs of the waveform, comparing the waveforms to models and/or thresholds, determining the centroid of a feature of the waveform, taking an envelope of a portion of the waveform and identifying a feature of the envelope, combinations of the same, and the like. For example, a foot point on an acoustic waveform can be identified by the point at which the derivative of the acoustic waveform is zero in a time window defined after an electrical trigger (e.g., as identified by an ECG sensor). As another example, a centroid or “middle” of a pulse, burst or wave of an acoustic waveform can be determined. In some embodiments, the centroid can be determined by constructing an amplitude envelope of the pulse, sound burst, or pressure wave and then performing a normalized weighting of each time point according to the envelope amplitude, with the highest energy point of the envelope being the centroid time of the pulse or sound (e.g., heart sound). The centroid determination can be performed in the time domain or frequency domain. In some embodiments, the centroid determination includes sampling and/or filtering of the pulse, sound burst, or pressure wave. Such signal processing can be performed dynamically, in real time. Alternatively or additionally, such signal processing can be performed during post processing of data.
• Atblock929, a time difference between the identified waveform features is determined. This time can be considered to be the arterial PWTT in one embodiment. This arterial PWTT value can be used in determining if an automatic occlusive cuff should take a blood pressure measurement from a patient (seeFIG. 8). Theprocess900D can be performed continuously or substantially continuously so as to monitor a patient's arterial PWTT over time. Alternatively, theprocess900D can be performed as a spot-check upon occurrence of predetermined conditions or at predetermined time intervals.
• Theprocess900D can be modified in one embodiment to receive an ECG signal input to assist with identifying the feature of the acoustic heart sounds waveform atblock927. It can be difficult to properly identify features from acoustic waveforms due to noise or other characteristics. An ECG signal, on the other hand, can have clearly-identifiable landmarks, including the R wave peak, among others. An ECG signal can therefore be used as a gating function to determine which heart sound should be considered. In one embodiment, the first heart sound of the acoustic heart sounds waveform occurring after a feature identified from the ECG signal can be identified as the relevant acoustic heart sounds feature. The ECG signal can therefore resolve ambiguities in the acoustic waveforms, thereby improving noise immunity and/or noise reduction. In some embodiments, the ECG signal can be used to identify the transition from electrical to mechanical behavior. The time delay between the ECG signal and the first heart sound (e.g., S1 sound) detected by the acoustic heart sounds sensor can provide an indication of the health of the heart or other potential pathological conditions that may warrant medical attention. However, the ECG signal input is not used or needed in other embodiments.
• FIGS. 9E and 9F illustrate embodiments ofprocesses900E,900C for determining PWTT values, refining calculated PWTT values, and/or determining confidence of PWTT values. More particularly, theprocess900E illustrates an embodiment of a process for calculating, refining, or assessing PWTT values using information obtained from multiple acoustic sensors, one of which is an acoustic heart sounds sensor. Theprocess900C illustrates an embodiment of a system for arterial PWTT calculation, refinement, or assessment using acoustic sensors in conjunction with additional sensors, such as ECG sensors, optical sensors, and/or bioimpedance sensors. Theprocesses900E,900C can be implemented as part of theparameter calculation system100 or the bloodpressure monitoring system300. Each of the depicted blocks or modules of the processes900 can be implemented by hardware and/or software modules.
• Referring specifically toFIG. 9E, theprocess900E receives three acoustic waveforms. These three acoustic waveforms can include an acoustic heart sounds waveform and two acoustic pulse waveforms (a first pulse waveform and a second pulse waveform). For example, the acoustic pulse waveforms can include a wrist pulse waveform and a carotid pulse waveform. The acoustic heart sounds waveform and a first pulse waveform (e.g., wrist pulse waveform) can be received byPWTT determination block932 and the acoustic heart sounds waveform and a second pulse waveform can be received byPWTT determination block934. The PWTT determination block932 can calculate a first arterial PWTT (a-PWTT A) value from the acoustic heart sounds waveform and the first acoustic pulse waveform. The PWTT determination block934 can calculate a second arterial PWTT (a-PWTT B) value from the acoustic heart sounds waveform and the second acoustic pulse waveform. These initial arterial PWTT measurements are determined in one embodiment by determining a time difference between features of the acoustic heart sounds waveform and the acoustic pulse waveform (as described in connection withFIG. 9A).
• In some implementations, the initial arterial PWTT values (a-PWTT A and a-PWTT B) can be provided to acomparison module936. Thecomparison module906 can analyze the two initial arterial PWTT values to generate a final arterial PWTT output value. In some implementations, thecomparison module936 can compare the two initial arterial PWTT values to determine a difference between the two values. Thecomparison module936 can derive a confidence value from the calculated difference. In other implementations, thecomparison module936 can average or otherwise combine the two initial arterial PWTT values to output a refined final arterial PWTT output value. In some embodiments, the two initial arterial PWTT values are weighted. In yet other implementations, thecomparison module936 can select one of the two initial arterial PWTT values to output as the final arterial PWTT output value. Thecomparison module936 can make this selection based on confidence values, based on comparison to historical data or thresholds, based on patient-specific factors, and/or the like.
• In some embodiments, the second pulse acoustic waveform can simply be used to provide another reference or gating point in identifying a feature of the first pulse acoustic waveform. In other embodiments, the second pulse acoustic waveform can be used to identify other patient characteristics (e.g., patient breaths, respiratory rate, respiratory pause, or other respiratory conditions, as discussed above).
• FIG. 9F illustrates an embodiment of aprocess900C that uses information obtained from one or more non-acoustic sensors in addition to one or more acoustic sensors to calculate, refine, or assess arterial PWTT measurements.
• Like theprocess900E, theprocess900E can calculate two arterial PWTT values (a-PWTT A and a-PWTT B) at PWTT determination blocks932 and934, respectively. Theprocess900C can also receive an ECG signal input, a plethysmograph signal input, and/or a bioimpedance signal input in addition to the acoustic signal inputs. An overall PWTT value (including arterial PWTT and PEP components) can be calculated at PWTT determination block937 by determining a time difference between a feature of a plethysmograph waveform and a feature of an ECG waveform. Further blocks of theprocess900C can determine PEP so as to compensate the overall PWTT for the PEP. The PEP is calculated atblock938 by determining a time difference between a feature of the ECG waveform and a feature of the acoustic heart sounds waveform or a bioimpedance waveform in the manner described above. The feature used on the ECG waveform to calculate thePEP938 can be the same feature used to calculate the overall PWTT atblock936. Alternatively, different ECG features can be used to calculate PEP and the overall PWTT. Atblock940, the PEP is subtracted from the initial PWTT to produce a third arterial PWTT value (a-PWTT C).
• In other embodiments, the third arterial PWTT value can be derived from a time difference between a feature of an acoustic waveform or a bioimpedance waveform and a pleth waveform such that PEP need not be calculated, as described above.
• Similar to theprocess900E, the three arterial PWTT values can be received by thecomparison module936 and analyzed to output a final arterial PWTT value. In some implementations, thecomparison module936 can compare the three initial arterial PWTT values to determine differences between any two of the three values. Thecomparison module936 can derive a confidence value from the calculated differences. In other implementations, thecomparison module936 can average or otherwise combine the three initial arterial PWTT values to output a refined final arterial PWTT output value. In some embodiments, the three initial arterial PWTT values are weighted. In yet other implementations, thecomparison module936 can select one of the three initial arterial PWTT values to output as the final arterial PWTT output value. Thecomparison module936 can make this selection based on confidence values, based on comparison to historical data or thresholds, based on patient-specific factors, and/or the like.
• In other embodiments, arterial PWTT values can be calculated using one or more non-acoustic sensors in combination with one or more acoustic sensors. For example, arterial PWTT values can be calculated from time differences between a feature of a bioimpedance waveform (e.g., generated from signals received from two bioimpedance sensors) and a feature of an acoustic waveform (e.g., received from an acoustic pulse sensor positioned over or proximate a wrist, leg or carotid artery).
• FIG. 10 illustrates an embodiment of aprocess1000 for determining whether to trigger an alternative blood pressure measurement. Thisprocess1000 can be implemented by any of thesystems100,300, described above. Advantageously, in certain embodiments, theprocess1000 can determine, based at least partly on non-invasive PWTT measurements, whether to trigger an automatic occlusive cuff. As a result, continuous or substantially continuous monitoring of a user's blood pressure can occur, allowing the frequency of occlusive cuff measurements to potentially be reduced.
• Atblock1002, a first arterial PWTT measurement is determined at a first point in time. The arterial PWTT can be determined using any of the techniques above, such as by calculating PEP and by compensating an overall PWTT value with the PEP. Similarly, a second arterial PWTT measurement is taken at a second point in time atblock1004. These arterial PWTT measurements can be taken from successive heart beats in one embodiment. In another embodiment, the first and second arterial PWTT values each represent arterial PWTT values averaged over multiple heartbeats.
• Atblock1006, a difference between the two arterial PWTT measurements is determined. It is then determined atdecision block1008 whether a new blood pressure measurement is required. In certain embodiments, this decision can be made by determining whether the difference between the two measurements is greater than a threshold. A difference greater than a threshold can be indicative of a change in a patient's blood pressure. Therefore, if the difference is greater than the threshold, an occlusive cuff is triggered to take a new blood pressure measurement atblock1010. If the difference is not greater than the threshold, then theprocess1000 loops back toblock1002. Effectively, theprocess1000 therefore can trigger occlusive cuff measurements when the threshold is exceeded and can continue monitoring arterial PWTT measurements otherwise.
• In certain embodiments, theprocess1000 analyzes changes in arterial PWTT measurements using an absolute difference technique or a moving difference technique. With the absolute difference technique, theprocess1000 measures the PWTT at a first fixed time. Subsequent arterial PWTT measurements (e.g., the second measurement at block1004) are compared to the initial arterial PWTT at the first fixed time to determine whether the difference between these measurements exceeds a threshold. With the moving difference technique, the first and second arterial PWTT measurements are compared for successive points in time. The first arterial PWTT measurement is therefore not taken at a fixed time but instead changes over time. Thus, the moving difference technique can approximate a derivative of the arterial PWTT measurements. The moving difference can be compared to a threshold atblock1008. An advantage of using the moving difference technique is that it can potentially ignore drifts in arterial PWTT measurements due to calibration changes or other errors.
• Thus, in certain embodiments, theprocess1000 can refrain from triggering an occlusive cuff until the non-invasive measurement differs enough to trigger such a measurement. Advantageously, in certain embodiments, theprocess1000 can therefore allow a user to postpone the discomfort and potential physiological damage associated with occlusive blood pressure measurements, while the non-invasive measurement (e.g., arterial PWTT) is within a certain tolerance.
• Although the PWTT measurements have been described herein as being used to trigger an occlusive cuff, in certain embodiments the PWTT measurements can also or instead be used to derive an estimate of blood pressure. A calibration function or curve can be determined that maps PWTT measurements to blood pressure values. The slope and intercept of the calibration curve can be determined experimentally.
PWTT Noise Compensation
• As described above, in addition to compensating PWTT for PEP, theparameter calculation system100 can also compensate PWTT data for noise.FIGS. 11 and 12 illustrate embodiments for compensating PWTT data for noise. By compensating PWTT data for noise, more accurate PWTT measurements can be obtained. The features described with respect toFIGS. 11 and 12 can be used in combination with any of the features described herein.
• FIG. 11 depicts anexample plot1100 that illustrates noise in a set of arterial PWTT data. In theexample plot1100, PWTT values taken over time are represented as aPWTT waveform1110. Variability in thePWTT waveform1110 reflects noise in thePWTT waveform1110. Noisy PWTT data can lead to lower quality PWTT measurements, resulting in too frequent or too infrequent triggering of a blood pressure cuff. In certain circumstances, noisy PWTT can even lead to missing a clinically significant change in blood pressure.
• One technique to reduce the effects of noise in the PWTT data is to average the PWTT data over time. A time-averagedPWTT waveform1120 is shown inFIG. 11. The average can be calculated from PWTT data points taken over a certain time period. PWTT data points can be averaged by applying a smoothing filter, such as a low pass filter. Determining a number of PWTT data points and/or an averaging time to include in an average can involve tradeoffs. A longer averaging time can reduce the presence of noise in the PWTT data. However, a longer averaging time can also reduce the benefits of using PWTT as an indicator of blood pressure to trigger an occlusive cuff. For example, a long averaging time can cause rapidly-changing features on a waveform to be missed. In contrast, a shorter averaging time using fewer data PWTT data points may not sufficiently filter out noise.
• In certain embodiments, averaging can be improved by dynamically adjusting the averaging time according to the amount of noise present in the PWTT data. If a higher level of noise is present, a longer averaging time can be selected. Conversely, if less noise is present, a shorter averaging time can be used. However, it can be difficult to differentiate noise from signal in the PWTT data. Thus, determining the level of the noise in the PWTT data and therefore the amount of averaging to perform can be nontrivial.
• To select an appropriate averaging time, in one embodiment a signal correlated with the PWTT data can be analyzed. Heart rate is such a correlated signal, and respiratory rate is another. An example heart rate signal is shown plotted over time as awaveform1130 inFIG. 11. The variability of theheart rate waveform1130 corresponds to the variability of thePWTT data1110, indicating that the heart rate and PWTT can be correlated. The heart rate waveform1130 (or respiratory rate) can therefore be analyzed to determine the amount of noise present in the PWTT data to thereby select an averaging time for the PWTT data. Likewise, in some applications, such as where accurate heart rate measurements are desired, it can be useful to analyze the PWTT data to determine an averaging time for the heart rate data. Dynamic signal averaging is described in greater detail below with respect toFIG. 10.
• Another possible use of the heart rate (or respiratory rate) data is to use the heart rate (or respiratory rate) as a noise reference in an adaptive filter. For example, the system100 (or any other system described herein) can adaptively filter or adaptively cancel noise in the PWTT data using the heart rate as a noise reference. Any of a variety of adaptive algorithms can be used by theparameter calculation system100. For instance, the adaptive algorithm can implement one or more of the following: a least mean squares algorithm (LMS), a least squares algorithm, a recursive least squares (RLS) algorithm, a Kalman filter, a joint process estimator, an adaptive joint process estimator, a least-squares lattice joint process estimator, a least-squares lattice predictor, a correlation canceller, optimized or frequency domain implementations of any of the above, any other linear predictor, combinations of the same, and the like. In still other embodiments, the PWTT data can be both adaptively filtered and averaged using a dynamically-adjusted averaging filter.
• Further, as can be seen inFIG. 11, the PWTT data in thewaveforms1110 or1120 can be correlated with a systolicblood pressure data1125 over time. As shown, the relationship between the PWTT data and the systolicblood pressure data1125 can be inverse or approximately inverse.
• FIG. 12 illustrates an embodiment of asignal averaging process1200. Thesignal averaging process1200 can be implemented by any of thesystems120,300,500A, or500B described above. Advantageously, in certain embodiments, theprocess1200 uses heart rate data to dynamically adjust the averaging time of an averaging filter applied to the PWTT data.
• Atblock1202, statistics of the heart rate data are analyzed to determine a measure of the noise in the heart rate data. A variety of statistics of the heart rate data can be calculated atblock1202, including variance, standard deviation, and entropy, combinations of the same, and the like. For example, the standard deviation can be calculated from the heart rate data to measure the degree of dispersion or variability in the heart rate data. A higher standard deviation can reflect greater variability in the heart rate data, which can in turn reflect more noise in the heart rate data. Conversely, a lower standard deviation can reflect lower noise in the heart rate data. More generally, any measure of variation or change in the heart rate data can be calculated atblock1202.
• Atblock1204, one or more averaging parameters of an averaging filter can be adjusted based on the statistical calculation. In one embodiment, an averaging parameter such as averaging time is adjusted. For example, if the standard deviation of the heart rate data is relatively high, reflecting more noise, a relatively long averaging time can be selected. On the other hand, if the standard deviation is relatively lower, reflecting less noise, a relatively shorter averaging time can be selected. More generally, the one or more averaging parameters can include any parameter of the averaging filter that can be tuned. The averaging parameters can therefore include the order of the averaging filter, the number of points averaged (e.g., for a simple moving average filter), the time constant of the filter (e.g., for a low pass averaging filter), and an array of coefficients for more complex averaging filters (such as any of the adaptive filters described above with respect toFIG. 9).
• Atblock1210, the averaging filter with the selected averaging time is applied to the PWTT data. Averaging can include computing a mean, median, and/or mode. In some embodiments, a weighted average of PWTT measurements can be applied. For example, the averaging filter can place a greater weight on more recent PWTT data and/or data with a corresponding statistic that indicates lower variability. For signals with relatively lower variability in the heart rate data, the averaging filter may not be applied to the PWTT input at all. Further, in other embodiments, the statistics of the PWTT signal can be evaluated to determine the amount of noise and therefore averaging time, rather than evaluating the statistics of the heart rate signal.
Phase Impact Reduction
• In some embodiments, the PWTT measurements and calculations described herein are determined in terms of time. Accordingly, phase delay or phase noise can be introduced between successive filters of the parameter calculation system or blood pressure monitoring system.FIG. 13 illustrates an embodiment offront end circuitry1300 that can be configured to condition analog electrical signals output by the acoustic sensors for processing. In some embodiments, thefront end circuitry1300 can reduce the phase impact introduced into the system. Thefront end circuitry1300 can include apre-amplification module1305, one ormore filters1310, and an analog-to-digital converter (ADC)1315.
• Thepre-amplification module1305 can receive and amplify the voltage signal from the acoustic sensors, such as by a predetermined gain. In some implementations, thepre-amplification module1305 introduces a high input impedance to drop the high pass filtering effect of the piezoelectric acoustic sensors. In some implementations, thepre-amplification module1305 includes circuitry to convert an AC signal output from the acoustic sensor to a DC signal. In some implementations, the one ormore filters1310 include one or more low pass filters to allow the low frequency heart and pulse acoustic signals detected by the acoustic sensors while attenuating high frequency signals, such as high frequency noise. The one ormore filters1310 may include other types of filters, such as high pass or bandpass filters, instead of or in addition to, the one or more low pass filters. In some implementations, the one ormore filters1310 can be adjusted to reduce the phase distortion by changing the location of the zeros and/or poles of the one ormore filters1310.
• In some implementations, the cutoff frequency of the acoustic sensors (which can be represented as capacitors in terms of frequency response analysis) can effectively be changed to a much lower frequency than an acoustic sensor used for respiration detection (e.g., to 0.01 Hz or 0.1 Hz). According to some embodiments, the acoustic sensors can have a cutoff frequency that is inversely dependent on the capacitance of the sensor and the input resistance of the amplifier or other circuit that receives the signal. The acoustic sensors may be modified to achieve a much lower cutoff frequency. For example, the low frequency response can be extended by changing the piezoelectric characteristics of the acoustic sensor (e.g., making the capacitance of the piezoelectric material itself larger, such as by the composition, size, thickness, etc. of the piezoelectric material), by making the input resistance higher, and/or by placing a large (as compared to the piezoelectric capacitance) capacitor in shunt with the acoustic sensor. In some embodiments, rather than lowering the cutoff frequency of the acoustic sensor itself, a compensatory filter can be inserted in the signal path. In one embodiment, the cutoff frequency is allowed to be significantly higher than the frequency band of interest, such that the acoustic sensor is effectively responsive to the derivative of the signal. Thus, the compensatory filter can be an integrator, which recreates the original acoustic signal. In some embodiments, it may be neither necessary nor desirable to extend the low frequency response of the acoustic sensors. The output from thefilters1310 can be amplified into a higher voltage signal and then converted to a digital signal by theADC converter1315. In some embodiments, using asingle ADC converter1315 to convert each signal, or multiple of the signals, including acoustic, electrical, and/or optical, can also reduce phase distortion between the signals. Further, oversampling in theADC1315 can also reduce phase distortion between the signals. The digital signal can then pass to a digital signal processor for processing in preparation for display on a monitor or screen.
Patient Calibration
• FIGS. 14A and 14B illustrate an embodiment of aprocess1400 for calibrating pulse wave transit time measurements based on a personalized patient calibration factor. Pulse wave transit time measurements can vary depending on changes in blood pressure. The variance can be defined by a slope measurement, which can be the change in pulse wave transit time per the change in pressure (e.g., change in number of milliseconds per change in number of millimeters of mercury (mmHg)). The variance, or slope, of the calibration curve can differ greatly from patient to patient. For example, the slope can vary by a factor of between 2:1 to 4:1 or greater between patients. This variance can be due to a variety of physiological factors, such as hemodynamic factors of capillary flow or perfusion. The slope may also vary based on the introduction of drugs (including even blood pressure regulation drugs), patient comorbidity, gender, age, or other factors. Accordingly, it can be advantageous to determine the slope of the PWTT/blood pressure calibration curve for the individual patient and use the determined slope to interpret subsequent PWTT measurements.
• The individualizedpatient calibration process1400 can reduce the effect of slope variability between patients by uniquely and individually calibrating the PWTT measurements for the individual patient, thereby reducing the inaccuracy of blood pressure measurement frequency or false alarms based on PWTT measurements. The individualizedpatient calibration process1400 therefore advantageously provides a mechanism for more efficient interpretation of PWTT measurements for individual patients by compensating for the variability of slope. Alternatively, in one embodiment, a patient monitor (such as any of those described herein) can calculate a baseline PWTT variability offset that is applied to the calculation of PWTT for multiple patients. This baseline adjustment may, but need not, be calibrated to individual patients or segments of patients (e.g., segments for drug users, segments based on gender, age, or comorbidity).
• The term PWTT measurements as used herein with respect toFIGS. 14A and 14B can refer to overall PWTT (including PEP and a-PWTT components) or arterial PWTT. The individualizedpatient calibration process1400 can be implemented by the blood pressure monitoring systems and/or parameter calculation systems described herein (e.g., systems140 and300).
• Thecalibration process1400 begins atblock1402, wherein at least two blood pressure measurements of a patient are obtained at different times. The blood pressure measurements can include just systolic blood pressure measurements, just diastolic blood pressure measurements, and/or combinations of the systolic and diastolic blood pressure measurements (for example, mean or median blood pressure measurements).
• In some implementations, blood pressure measurements can be taken at two different elevations of known hydrostatic pressure at two different times. For example, a first blood pressure measurement can be taken when the patient is lying down on a bed or table with the arm at the patient's side and a second blood pressure measurement can be taken with the arm raised up at least approximately perpendicularly to the patient's body. The blood pressure measurements can be normalized to account for changes in pressure due to the change in elevation. The changes in pressure due to the change in elevation can be determined based on a measured or known distance between the locations of the measurements and calculated or known hydrostatic pressure and/or based on actual hydrostatic pressure measurements. The distance and/or pressure measurements can be obtained by any of a variety of sensors or transducers. In some embodiments, a wrist cuff is used to obtain the blood pressure measurements (which, in some embodiments, reduces the effects of the arterial-to-capillary time delays addressed above). Other methods can be used to obtain blood pressure measurements over time in accordance with yet other implementations.
• Atblock1404, pulse wave transit times are measured to correspond with the blood pressure measurements. Althoughblock1404 is illustrated as occurring afterblock1402, the pulse wave transit times can be measured simultaneously or substantially simultaneously (e.g., within a few seconds or within a few milliseconds) with the blood pressure measurements. Accordingly, the pulse wave transit times can be mapped to their corresponding blood pressure measurements. The blood pressure measurements and pulse wave transit time measurements can be stored in memory (at least temporarily). In some embodiments, the memory can include a predetermined capacity of measurements and the oldest measurements can be replaced by new measurements when capacity has been reached. The stored measurements can be used to generate a calibration curve defined by a plot of the pulse wave transit times on one axis (e.g., vertical axis) and the corresponding blood pressure measurements on the other axis (e.g., horizontal axis).
• Atblock1406, an individual patient's slope corresponding to the change in pulse wave transit time per change in blood pressure (e.g., slope of the calibration curve) is determined from the blood pressure and PWTT measurements obtained atblock1402 andblock1404. A patient's PWTT ms/mmHg slope can be determined from two or more blood pressure measurements and two or more corresponding PWTT measurements. In accordance with some embodiments, the slope is inferred over time from a plurality of frequent blood pressure and corresponding PWTT measurements. The slope determination can then be refined over time through less frequent natural blood pressure and corresponding PWTT measurements.
• Based on the determined slope of the patient, a personalized, or individualized, patient calibration factor can be identified atblock1408 and used to interpret subsequent pulse wave transit time measurements of the patient with reduced concern for patient variability.
• Atblock1410, a pulse wave transit time is measured (for example, an arterial pulse wave transit time can be measured as described herein). Atblock1412, theprocess1400 estimates a change in the patient's blood pressure based on the measured pulse wave transit time. The measured pulse wave transit time has been calibrated using the individualized patient calibration factor, thereby reducing patient variability.
• Atdecision block1414, theprocess1400 determines whether the estimated blood pressure change satisfies a condition that warrants performance of an action (e.g., whether the estimated blood pressure change is significant enough to perform an action). The condition can be, for example, a blood pressure change that exceeds a threshold blood pressure change value, cuff triggering value, a patient alarm triggering value, and/or the like. The condition can be patient-specific based on patient-specific factors or generic to all patients. If the condition is satisfied, theprocess1400 proceeds to block1416, wherein an appropriate action is performed based on the satisfied condition. For example, if an estimated blood pressure change exceeds a threshold change, a blood pressure cuff measurement can be triggered to determine and/or confirm the accuracy of the estimated blood pressure change. In some embodiments, the new blood pressure measurement can be added to a running tally to refine the individualized patient calibration factor.
• As another example, if an estimated blood pressure change exceeds an alarm condition, a patient alarm can be triggered so that the patient can receive immediate medical attention. In some embodiments, the alarm is triggered only if confidence in the estimated blood pressure change is above a predetermined level. In some embodiments, a blood pressure measurement is obtained to confirm the accuracy of the estimated blood pressure change prior to triggering an alarm.
• In some implementations, the determination of whether the blood pressure change is significant enough to warrant an action atblock1408 includes a multivariate determination based on other factors in addition to pulse wave transit time (e.g., heart rate). For example, if heart rate and pulse wave transit time are changing, it is more likely that the blood pressure is changing in a manner such that an action is recommended to be performed.
• In some situations, the patient's calibration curve slope can change due to changes in patient conditions (e.g., changes in hemodynamic state, introduction of drugs, etc.). Accordingly, in some implementations, the slope is tracked over time to determine whether a recalibration process may be recommended. Other inputs or physiological parameters, such as heart rate, can be analyzed in determining whether recalibration of the personalized patient calibration factor should be performed as well.
• In other implementations, a slope of a patient's blood pressure over time can be determined instead of, or in addition to, the PWTT/blood pressure calibration curve slope and used to identify a personalized patient calibration factor, which can in turn be used to interpret pulse wave transit time measurements.
Terminology
• Many other variations than those described herein will be apparent from this disclosure. For example, depending on the embodiment, certain acts, events, or functions of any of the algorithms described herein can be performed in a different sequence, can be added, merged, or left out all together (e.g., not all described acts or events are necessary for the practice of the algorithms). Moreover, in certain embodiments, acts or events can be performed concurrently, e.g., through multi-threaded processing, interrupt processing, or multiple processors or processor cores or on other parallel architectures, rather than sequentially. In addition, different tasks or processes can be performed by different machines and/or computing systems that can function together.
• The various illustrative logical blocks, modules, and algorithm steps described in connection with the embodiments disclosed herein can be implemented as electronic hardware, computer software, or combinations of both. To clearly illustrate this interchangeability of hardware and software, various illustrative components, blocks, modules, and steps have been described above generally in terms of their functionality. Whether such functionality is implemented as hardware or software depends upon the particular application and design constraints imposed on the overall system. For example, thevehicle management system110 or210 can be implemented by one or more computer systems or by a computer system including one or more processors. The described functionality can be implemented in varying ways for each particular application, but such implementation decisions should not be interpreted as causing a departure from the scope of the disclosure.
• The various illustrative logical blocks and modules described in connection with the embodiments disclosed herein can be implemented or performed by a machine, such as a general purpose processor, a digital signal processor (DSP), an application specific integrated circuit (ASIC), a field programmable gate array (FPGA) or other programmable logic device, discrete gate or transistor logic, discrete hardware components, or any combination thereof designed to perform the functions described herein. A general purpose processor can be a microprocessor, but in the alternative, the processor can be a controller, microcontroller, or state machine, combinations of the same, or the like. A processor can also be implemented as a combination of computing devices, e.g., a combination of a DSP and a microprocessor, a plurality of microprocessors, one or more microprocessors in conjunction with a DSP core, or any other such configuration. A computing environment can include any type of computer system, including, but not limited to, a computer system based on a microprocessor, a mainframe computer, a digital signal processor, a portable computing device, a personal organizer, a device controller, and a computational engine within an appliance, to name a few.
• The steps of a method, process, or algorithm described in connection with the embodiments disclosed herein can be embodied directly in hardware, in a software module executed by a processor, or in a combination of the two. A software module can reside in RAM memory, flash memory, ROM memory, EPROM memory, EEPROM memory, registers, hard disk, a removable disk, a CD-ROM, or any other form of non-transitory computer-readable storage medium, media, or physical computer storage known in the art. An exemplary storage medium can be coupled to the processor such that the processor can read information from, and/or write information to, the storage medium. The storage medium can be external to or remote from the processor. For example, the storage medium may be a storage device accessible by the processor over a network. In the alternative, the storage medium can be integral to the processor. The processor and the storage medium can reside in an ASIC. The ASIC can reside in a user terminal. In the alternative, the processor and the storage medium can reside as discrete components in a user terminal.
• Conditional language used herein, such as, among others, “can,” “might,” “may,” “e.g.,” and the like, unless specifically stated otherwise, or otherwise understood within the context as used, is generally intended to convey that certain embodiments include, while other embodiments do not include, certain features, elements and/or states. Thus, such conditional language is not generally intended to imply that features, elements and/or states are in any way required for one or more embodiments or that one or more embodiments necessarily include logic for deciding, with or without author input or prompting, whether these features, elements and/or states are included or are to be performed in any particular embodiment. The terms “comprising,” “including,” “having,” and the like are synonymous and are used inclusively, in an open-ended fashion, and do not exclude additional elements, features, acts, operations, and so forth. Also, the term “or” is used in its inclusive sense (and not in its exclusive sense) so that when used, for example, to connect a list of elements, the term “or” means one, some, or all of the elements in the list.
• While the above detailed description has shown, described, and pointed out novel features as applied to various embodiments, it will be understood that various omissions, substitutions, and changes in the form and details of the devices or algorithms illustrated can be made without departing from the spirit of the disclosure. As will be recognized, certain embodiments of the inventions described herein can be embodied within a form that does not provide all of the features and benefits set forth herein, as some features can be used or practiced separately from others.

Claims (15)

1-33. (canceled)

34. A method of reducing blood pressure cuff measurements, the method comprising:

estimating a blood pressure measurement from a plurality of non-invasive sensors;

determining that the estimated blood pressure change satisfies a condition; and

triggering a blood pressure cuff measurement based on the determination.

35. The method ofclaim 34, wherein the plurality of non-invasive sensors comprise a pulse oximeter and an electrocardiogram (ECG) sensors.

36. The method ofclaim 35, wherein the plurality of non-invasive sensors further comprise an acoustic sensor.

37. The method ofclaim 34, wherein the condition comprises exceeding a threshold blood pressure change value.

38. The method ofclaim 34, wherein the condition is based on patient-specific factors.

39. The method ofclaim 34, wherein the determination is further based on heart rate measurements.

40. The method ofclaim 34, further comprising using an adaptive filter for the estimations of the blood pressure measurement.

41. A system of reducing blood pressure cuff measurements, the system comprising one or more hardware processors configured to:

estimate a blood pressure measurement from a plurality of non-invasive sensors;

determine that the estimated blood pressure change satisfies a condition; and

trigger a blood pressure cuff measurement based on the determination.

42. The system ofclaim 41, wherein the plurality of non-invasive sensors comprise a pulse oximeter and an electrocardiogram (ECG) sensors.

43. The system ofclaim 42, wherein the plurality of non-invasive sensors further comprise an acoustic sensor.

44. The system ofclaim 41, wherein the condition comprises exceeding a threshold blood pressure change value.

45. The system ofclaim 41, wherein the condition is based on patient-specific factors.

46. The system ofclaim 34, wherein the determination is further based on heart rate measurements.

47. The system ofclaim 34, further comprising using an adaptive filter for the estimations of the blood pressure measurement.

Images