Non-invasive system and method for detecting and accurately quantifying subclinical and clinical systolic and diastolic heart failure - Patent Application 20070122999

Abstract

A system and method are described that provides non-invasive detection and quantification of systolic and diastolic heart failure, including subclinical, asymptomatic systolic and diastolic heart failure. The invention is a novel method for processing physiological data and deriving clinical implications and prognosis. The invention is based on extracting data from simultaneous data streams of electrocardiogram, phonocardiogram, and left ventricular pressure to be graphically displayed in a Wiggers diagram that depicts the entire cardiac cycle.

Description

[Technical Field]

[0001] CROSS-REFERENCE TO RELATED APPLICATIONS This application claims the benefit of U.S. Provisional Patent Application No. 63 / 358,416, filed July 5, 2022.

[0002] The objective of the present invention is to provide non-invasive detection and quantification of systolic and diastolic heart failure, including latent, asymptomatic systolic and diastolic heart failure. In this regard, the present invention is a novel method for processing physiological data and deriving clinical implications and prognosis. The present invention is based on extracting data that can be graphically displayed on the well-known Wiggers diagram (hereinafter "Wiggers diagram"), which shows the entire cardiac cycle with simultaneous data streams of electrocardiogram, phonocardiogram, and left ventricular pressure. See https: / / commons.wikimedia.org / wiki / File:Wiggers_Diagram.svg. [Background technology]

[0003] Current state-of-the-art techniques use two-dimensional ultrasound imaging to measure parameters such as ejection fraction (which is not invariant to afterload or preload), myocardial strain, and myocardial strain rate. These prior art measurements are entirely mechanical and completely ignore the complex, interactive electrical and pathoelectrophysiological characteristics of the transduction dysfunction in systolic and diastolic heart failure. The pathoelectrophysiological hallmark of such dysfunction is the rapid and massive movement of calcium ions from the sarcoplasmic reticulum (SRP) into the cytosol, followed by a brief pause with each heartbeat and then a very rapid return to the SRP. Current methods only quantify the mechanical characteristics of the disease, ignoring the ion permeability-membrane flux characteristics of the disease. Ion membrane flux is expressed in units of sec^-1 as the reciprocal of (E-M), i.e., 1 / (E-M). The quantity 1 / (E-M) is essentially the transduction rate or transduction rate of the membrane flux of calcium ions during both systole and diastole: "transductions per second" during systole and "detransduction rate" during diastole. Thus, current measurements, both systolic and diastolic, essentially fail to capture half of the physiological characteristics of heart failure, a significant deficiency.

[0004] Thus, current diagnostic and quantitative methods are only useful for characterizing and quantifying symptoms in people who already have the disease or are living with symptoms. These methods lack sufficient sensitivity and reliability to serve as screening tests for subclinical or preclinical disease, but are not currently screened. A test for "pre-HF" (pre-heart failure), similar to the use of HbA1c to detect "pre-diabetes," could identify opportunities for interventions—such as afterload-reducing medications, reduced salt intake, and fluid management—early in the natural history of HF, potentially extending survival, improving quality of life, and modifying the natural course of the disease. Such a test, if feasible, could save healthcare systems significant costs otherwise spent on frequent rehospitalizations for acute decompensation of heart failure, requiring costly ICU stays with endotracheal intubation and mechanical ventilation. Metrics that combine electrical and mechanical properties across the full range of possible contractile states, even in patients with clinically known HF, would be more accurate, reproducible, and physiologically meaningful. Summary of the Invention [Problem to be solved by the invention]

[0005] As stated above, the objective of the present invention is the non-invasive detection and quantification of systolic and diastolic heart failure, including latent, asymptomatic systolic and diastolic heart failure. The present invention is a novel method for processing physiological data and deriving clinical implications and prognosis. The present invention is premised on extracting data from simultaneous data streams of electrocardiogram, phonocardiogram, and left ventricular pressure to be graphically displayed in a Wiggers diagram, which depicts the entire cardiac cycle.

[0006] The Wiggers diagram shows the relationship between the Q wave, the maximum second derivative of the systolic ECG, and the slightly delayed S1 heart sound. This time interval is the (EM)ino, or intropic electrical-mechanical interval. By taking the time point at which the T wave first exhibits maximum upward acceleration and then noting the time delay to the S2 heart sound, the (EM)lusi, or lusitropic electrical-mechanical interval, is identified. When a vibratory cardiogram obtained with a precordial accelerometer is available, the vibration signals corresponding to S1 and S2, or their peak derivatives, can serve as the "M" of the systolic or diastolic electromechanical interval (EM), respectively. The ratio 1 / (EM) can be understood as the rate of inotropic or lusitropic electromechanical transduction. The rate of electromechanical transduction during systole (or electromechanical detransduction during diastole) has been shown to be linearly proportional to the natural logarithm of the magnitude of the systolic or diastolic strain rate obtained from a 2D transthoracic echocardiogram, respectively, during systole or diastole, as described in the inventor's prior patent applications identified herein. [Means for solving the problem]

[0007] Experimental results obtained by the inventors from dobutamine stress testing using new metrics of inotropic and lusitropic function, electromechanical intervals (EM)ino and (EM)lusi, yielded two sets of linear, individual calibration curves relating ln(Strain Rate) to 1 / (EM). Five evaluable subjects were studied. One set of five calibration curves describes systolic function for each of the five subjects, and the other set of five calibration curves describes diastolic function, both for the same five subjects. The calibration curves are expressed as ln(Strain Rate) = a + b / (EM), where "a" and "b" are the y-intercept and slope, constants for a given individual. (EM)ino is used for systolic function, and (EM)lusi is used for diastolic function. Plotting the y-intercept "a" as a function of the slope "b" for all five subjects in both the systolic and diastolic cases yields a line with a very high correlation coefficient, called the "intercept-slope trade-off function," with a downward sloping intercept "p" and slope "q." There is one "intercept-slope trade-off function" for the systolic case and another "intercept-slope trade-off function" for the diastolic case, the only difference between them being the values ​​of the intercept "p" and slope "q." [Brief explanation of the drawings]

[0008] [Figure 1] FIG. 1 is a Cartesian coordinate mapping showing the relationship between the inotropic intercept and slope trade-off functions for the five healthy subjects described herein. [Figure 2] FIG. 2 is a Cartesian coordinate mapping showing the relationship between lusitropy intercept and slope trade-off functions for the five healthy subjects described herein. [Figure 3] FIG. 3 plots the slope and intercept for individual patient "m" and individual patient "n" against a Cartesian mapping of the relationship between the inotropy intercept and slope trade-off function of FIG. [Figure 4]FIG. 4 plots the slope and intercept for individual patient "r" and individual patient "s" against a Cartesian mapping of the relationship between the lusitropy intercept and slope trade-off function of FIG. [Figure 5] FIG. 5 is a Cartesian mapping of the first derivative of a typical oscillating cardiogram and the second derivative of a typical electrocardiogram versus time. DETAILED DESCRIPTION OF THE INVENTION

[0009] As stated above, the objective of the present invention is the non-invasive detection and quantification of systolic and diastolic heart failure, including latent, asymptomatic systolic and diastolic heart failure. The present invention is a novel method for processing physiological data and deriving clinical meaning and prognosis therefrom. The present invention is premised on extracting data from simultaneous data streams of electrocardiogram, phonocardiogram, and left ventricular pressure to be graphically displayed in a Wiggers diagram showing the entire cardiac cycle.

[0010] The Wiggers diagram shows the relationship between the Q wave, which is the maximum second derivative of the systolic ECG, and the slightly delayed S1 heart sound. This time interval is the (EM)ino, or intropic electrical-mechanical interval. By taking the time when the T wave first exhibits maximum upward acceleration and then noting the time delay to the S2 heart sound, the (EM)lusi, or lusitropic electrical-mechanical interval, is identified. When a vibratory cardiogram obtained with a precordial accelerometer is available, the oscillatory signals corresponding to S1 and S2, or their peak derivatives, can serve as the "M" of the systolic or diastolic electromechanical interval (EM), respectively. The ratio 1 / (EM) can be understood as the rate of inotropic or lusitropic electromechanical transduction. The rate of electromechanical transduction during systole (or electromechanical detransduction during diastole) has been shown to be linearly proportional to the natural logarithm of the magnitude of the systolic or diastolic strain rate obtained from a 2D transthoracic echocardiogram, respectively, during systole or diastole, as described in the inventor's prior patent applications identified herein.

[0011] Experimental results obtained by the inventors from dobutamine stress testing using new metrics of inotropic and lusitropic function, electromechanical intervals (EM)ino and (EM)lusi, yielded two sets of linear, individual calibration curves relating ln(Strain Rate) to 1 / (EM). Five evaluable subjects were studied. One set of five calibration curves describes systolic function for each of the five subjects, and the other set of five calibration curves describes diastolic function, both for the same five subjects. The calibration curves are expressed as ln(Strain Rate) = a + b / (EM), where "a" and "b" are the y-intercept and slope, constants for a given individual. (EM)ino is used for systolic function, and (EM)lusi is used for diastolic function. Plotting the y-intercept "a" as a function of the slope "b" for all five subjects in both the systolic and diastolic cases yields a line with a very high correlation coefficient, called the "intercept-slope trade-off function," with a downward sloping intercept "p" and slope "q." There is one "intercept-slope trade-off function" for the systolic case and another "intercept-slope trade-off function" for the diastolic case, the only difference between them being the values ​​of the intercept "p" and slope "q."

[0012] Figure 1 shows the intercept-slope tradeoff function for inotropy, and Figure 2 shows the intercept-slope tradeoff function for lusitropy, both expressed as intercept = PQ(Slope). Consequently, the lines plotted in Figures 1 and 2 should be clearly understood to represent the upper parametric boundaries of systolic and diastolic function, respectively, of a healthy, normal heart. This applies across a wide range of subjects, regardless of their weight, height, age, or gender, because strain rate represents the function of healthy myocardial tissue regardless of size, shape, or the patient's height, weight, sex, or age. In this application, these calibration curves are referred to as the "universal tradeoff function for inotropic function" and the "universal tradeoff function for lusitropic function," respectively. The specifics of constructing the universal tradeoff function for inotropic function and the universal tradeoff function for lusitropic function are described in paragraphs

[0079] to

[0111] later in this specification.

[0013] The upper limits of systolic and diastolic function are obtained across the physiological range of catecholamine receptor activation, resulting in concomitant levels of inotropy and lusitropy. These global metrics of systolic and diastolic function, expressed across the physiological range of inotropy and lusitropy, are novel and valuable compared to current state-of-the-art techniques such as the ejection fraction at rest. These metrics incorporate clinical information across the full range of an individual patient's ability to compensate for any stress encountered in daily life. The fitting parameters of the trade-off function (p,q) may prove true for individual cardiomyocytes or for a random population of cardiomyocytes. The five subjects varied widely in age, weight, and height, and represented both males and females. The primary commonality among the subjects was that they all possessed healthy, native myocardial tissue. These linear universal or collective parameters, p and q—one for systole and one for diastole—indicate healthy myocardial tissue when functionally tested in a dobutamine stress test. These parameters are accurate representations of overall myocardial performance, as well as of small longitudinal muscle segments, which are averaged using proprietary algorithms by 2D echocardiography manufacturers to calculate strain rates. The fitting parameters (p, q) of the intercept-slope tradeoff function are expected to accurately represent individual myocytes, or any large random statistical sample of individual myocytes. For a given slope to an individual calibration curve, if a person has heart failure (HF), either systolic or diastolic, each point on the calibration curve will have a lower ln(Strain Rate) value than a person with healthy, normal myocardial tissue for all values ​​of 1 / (EM). Expressed algebraically, this means that in HF, the y-intercept of the individual calibration curve for a HF patient is lower than the y-intercept of a known normal heart. In two-dimensional {intercept, slope} space, each individual calibration curve can be uniquely represented as a point in the two-dimensional vector space representing the intercept vs. slope tradeoff function.The precise and reproducible degree of HF pathology is given by the vertical distance between point (b, a), which represents the standard curve for affected patients, and a line representing the normal standard curve as much as possible. This line is a trade-off function. If a person is asymptomatic and the vertical distance of an individual point from this universal line is still measurable, no matter how small, significantly exceeding the measurement error limit, it can be used to detect and quantify preclinical disease and alter clinical management accordingly. This technique has the potential to identify individuals with systolic and / or diastolic heart failure very early in the natural history of the disease, similar to the ability of blood HbA1c analysis to predict those who will develop or are at risk for developing diabetes, a state currently understood as "pre-diabetes." Essentially, this technology represents a new and highly sensitive analysis of inotropic and lusitropic function and, therefore, the first method to detect and measure "pre-HF" (pre-HF).

[0014] Current techniques use two-dimensional ultrasound imaging to measure parameters such as ejection fraction (which is not invariant to afterload or preload), myocardial strain, and myocardial strain rate. These prior art measurements are entirely mechanical and completely ignore the complex, interactive electrical and patho-electrophysiological characteristics of transduction dysfunction in systolic and diastolic heart failure. The patho-electrophysiological hallmark of such dysfunction is the rapid and massive movement of calcium ions from one cellular compartment, the sarcoplasmic reticulum, into the cytosol, followed by a brief pause and then very rapid movement back to the sarcoplasmic reticulum with each heartbeat. Current methods only quantify the mechanical hallmark of this disease, ignoring the transmembrane flux of ions. Ion transmembrane flux is expressed as the reciprocal of (EM), i.e., 1 / (EM), with units of sec^-1. The quantity 1 / (EM) is essentially the transduction rate or transduction rate of calcium ion membrane flux during systole, with units of "transductions per second," and the "detransduction rate" during diastole. Therefore, current measurement methods, both systolic and diastolic, are seriously flawed, effectively ignoring half of the physiological characteristics of heart failure. Thus, current diagnostic and quantification methods are only useful for characterizing and quantifying symptoms in individuals already suffering from the disease or living with the condition. These methods lack sufficient sensitivity and reliability to serve as screening tests for asymptomatic or preclinical individuals who are not currently screened. A test for "pre-heart failure," similar to the use of HbA1c to detect "pre-diabetes," could reveal opportunities to intervene early in the natural history of heart failure in ways that could extend life, improve quality of life, or modify the natural course of the disease, using drugs such as afterload-reducing agents, reduced salt intake, or fluid management. The availability of such testing could also save significant medical costs associated with frequent hospital readmissions due to acute decompensation of heart failure, which requires costly ICU stays with endotracheal intubation and mechanical ventilation.Even in patients with clinically known heart failure, metrics that combine electrical with mechanical properties across the full range of possible contractile states are more accurate, reproducible, and physiologically meaningful.

[0015] The method and system of the present invention geometrically and simply defines the relationship between the electrical and mechanical properties of transduction and tension development in normal, healthy myocardial tissue. Current approaches only address the mechanical properties of cardiac wall motion, contraction, or stretching with tension release or filling. Prior art approaches essentially ignore the pathophysiology of diseases that clearly have both electrical and mechanical characteristics. The graphical simplicity of the present invention makes the electrical and / or mechanical dysfunctions that cause heart failure, if any, easily and clearly visible, thus providing a simple and reproducible method for quantifying such analysis. The simplicity of the present approach is similar to the simplicity with which pediatricians plot a child's height and weight on a growth chart. Growth charts are universal and stratified by percentiles. A patient's height or weight is plotted as a single point at a time or as multiple points over time. Deviations from a universal norm are clearly indicated. Prior art approaches to quantifying systolic or diastolic heart failure are entirely mechanical in nature. This approach ignores important features of the disease and the relationships between disease features. The present invention allows clinicians to detect "pre-diabetes" in a manner similar to how "pre-diabetes" is currently diagnosed, providing opportunities for pharmacological intervention as well as fluids, electrolytes, dietary restriction, and exercise before subclinical symptoms progress to clinical symptoms and patients require intubation, mechanical ventilation, ventricular assist devices, or heart transplantation.

[0016] The progress of this management can be effectively monitored and adjusted in real time using wearable electronic and mechanical technologies that measure, transmit, and process (EM)ino and (EM)lusi as described in U.S. Pat. No. 10,085,665 ("the '665 patent") and U.S. Pat. No. 10,918,300 ("the '300 patent") and related divisional applications described in published U.S. application US 2021 / 0128047 ("the '047 application"). The preventative approach described herein is much less expensive than traditional approaches that require multiple hospitalizations for congestive heart failure decompensation. This preventative approach in accordance with the present invention can utilize wearable technology, such as the electromechanical systems shown in the '665 patent, the '300 patent, and the '047 application. Such electromechanical systems can operate under the programmatic control of the methods and algorithms described in the '665 patent, the '300 patent, and the '047 application, and the methods combined with such systems put low-cost early detection and prevention in the clinician's toolkit.

[0017] Heart failure is generally characterized by systolic (heart failure without preserved ejection fraction) and diastolic (heart failure with preserved ejection fraction) phases. Assume a patient undergoes a dobutamine stress test in which the dobutamine infusion rate is increased several times at steady state. Assume further that the ln(Strain Rate) during systole and diastole, along with simultaneous values ​​of 1 / (EM)ino and 1 / (EM)lusi, are measured and graphed, and a linear calibration curve for ln(Systolic Strain Rate) vs. 1 / (EM)ino and another for ln(Diastolic Strain Rate) vs. 1 / (EM)lusi are constructed, with the slope b and y-intercept a calculated for each. By plotting the systolic values ​​of a and b as points (b,a) on the graph in Figure 1, with the slope on the x-axis and the intercept on the y-axis, and the diastolic values ​​of a and b as points (b,a) on the graph in Figure 2, it is quite possible that both points will lie below the universal or intercept-slope trade-off function line on each graph. Measuring the perpendicular distance between these points and their respective lines parallel to the y-axis—in units of ln(sec^-1), or ln(Strain Rate)—provides an accurate and reproducible metric of systolic heart failure (Figure 1) and diastolic heart failure (Figure 2). In other words, a patient may have both diastolic and systolic heart failure, but to different degrees, which may measurably change relative to each other over time during the five-year natural history of the disease. Improvements in treatment outcomes can also be accurately tracked in this way. This alone is a significant advantage over the current state of metrology and monitoring.

[0018] The vertical distance between the universal inotropy intercept-slope tradeoff function (Figure 1) and a point on the intercept and slope of an individual patient's ln(Systolic Strain Rate) vs. 1 / (EM)ino calibration curve is referred to herein as VDino. The vertical distance between the universal lusitropy intercept-slope tradeoff function and a point on the intercept and slope of the same patient's lusitropy ln(Diastolic Strain Rate) vs. 1 / (EM)lusi calibration curve is referred to herein as VDlusi. Note that the units of both VDino and VDlusi are ln(Sec^-1), i.e., ln(Strain Rate).

[0019] Thus, where F is the ratio of the log of systolic dysfunction to the log of diastolic dysfunction, we can write F = VDino / VDlusi, and where T is the sum of systolic and diastolic dysfunction in a patient with heart failure, we can write T = VDino + VDlusi. Note that alternatively, we can plot (VDino, VDlusi) on the (x, y) axes and add them vectorially, noting the angle between the x-axis and the vectors.

[0020] This angle theta θ is given by the formula tan(θ)=VDlusi / VDino.

[0021] The magnitude of this vector is given by the formula VDtotal=(Dino^2+VDlusi^2)^1 / 2.

[0022] The angle θ provides a graphical and quantitative description of the contribution of diastolic dysfunction (y-axis) and systolic dysfunction (x-axis) to the clinical picture of heart failure, and whether systolic or diastolic forms of myocardial damage dominate the clinical picture. If θ<45°, systolic dysfunction predominates. If θ<45°, diastolic dysfunction predominates. The magnitude of the vector is a measure of the overall extent of the heart failure problem being diagnosed and treated. Over time, changes in this vector will reveal objective changes in the severity and characteristics of the condition, much like the pediatric growth curves described above.

[0023] This approach allows medical professionals to assess the risk of heart failure in a more robust, quantitative, accurate and reproducible manner than is currently possible with the current state of the art.

[0024] A second observation regarding Figures 1 and 2 is noteworthy. When the lines are plotted together on the same graph, the lusitropy intercept-slope tradeoff function is a downward-sloping line drawn 0.92863-(-0.03091) = 0.96773ln(sec^-1) above the inotropy intercept-slope tradeoff function, on the order of one logarithmic unit. The slopes in Figures 1 and 2, i.e., -4.61361 and -4.95698, are approximately equal within experimental error.

[0025] This graphical representation can be seen to demonstrate that strain rates in healthy patients are significantly faster during diastole than during systole. Normal hearts relax faster than they contract (squeeze). This human data is consistent with data published by the present inventors in 2011, in which pressure catheters were used in the left ventricles of septic pigs to measure the magnitude of the absolute value of the maximum first derivative of left ventricular pressure during systole (LVP'(t)max) compared to that during diastole over a period of hours, before and after placement of an E. coli septic clot in the pig's peritoneum. See the present inventors' 2011 abstract, published by the American Society of Anesthesiologists, entitled "Lusitropic / Inotropic Relation in Porcine Septic Shock: An Early Biomarker of Cardiodynamic Decompensation?"

[0026] In healthy animals, LVP'(t)max during diastole is greater than that during systole, even though the directions of motion are opposite during diastole and systole. As sepsis progresses, the situation reverses, and the ratio LVP'Distole / LVP'Systole, initially greater than 1, subsequently exceeds 1 as the pig's heart becomes hypercapnic and unable to compensate for the decline in SVR and blood pressure caused by progressive sepsis. Further scaled-down data from the same pigs referenced above, unpublished data, show a linear relationship between ln(LVP'(t)max) and 1 / (EM) for both inotropy and lusitropy. Because ln(LVP'max) and ln(Strain Rate) are both linearly proportional to 1 / (EM), it follows that ln(LVP'max) and ln(Strain Rate) are linearly proportional to each other. In other words, ln(LVP'max) and ln(Strain Rate), which are the maximum rates of change of pressure inside the left ventricle, can be understood as surrogate metrics for inotropy and lusitropy. If ln(LVP'(t)max) is used instead of ln(Strain Rate), only the magnitude of the fitting parameters, the slope and intercept, change. Of course, strain rate has the great advantage of being able to be measured noninvasively.

[0027] With respect to the complement of the drawings shown in the present application, each is described in more detail below:

[0028] Figure 1 summarizes the inotropic calibration curves for five healthy subjects of varying height, weight, and age who underwent dobutamine stress testing while simultaneously measuring 1 / (EM)ino and ln(abs(Systolic Strain Rate)). This figure illustrates the universal trade-off function of inotropic function. Each of the five points shown represents an individual subject. Each point is represented by (b, a), where b is the slope and a is the y-intercept of the individual subject's linear inotropic calibration curve. This calibration curve has the form ln(abs(Systolic Strain Rate)) = a + b / (EM)ino. The downward-sloping function shown in this figure is referred to as the inotropic intercept-slope trade-off function. This trade-off function is expected to hold for an arbitrarily large sample of individuals with good cardiovascular health and fitness. Furthermore, the fitting parameters (p, q) = (intercept, slope) of this inotropic intercept-slope trade-off function are considered to be a reasonable approximation of the universal parameters of inotropic function in healthy myocardial tissue.

[0029] In patients with heart failure, the linear inotropic calibration curve is expected to lie below the line depicted in the trade-off function above at a point (b, a). The vertical distance between this point and the trade-off function is VDino, a metric of systolic dysfunction in heart failure. Furthermore, the fitting parameters (p, q) = (intercept, slope) of this inotropic intercept-slope trade-off function are believed to be reasonable approximations of the universal parameters of inotropic function in healthy myocardial tissue.

[0030] Figure 2 summarizes the lusitropic calibration curves for the same five healthy subjects shown in Figure 1. This figure illustrates a universal tradeoff function for lusitropic function. These five healthy subjects, varying in height, weight, and age, underwent dobutamine stress testing while simultaneously measuring 1 / (EM)lusi and ln(Diastolic Strain Rate). Each of the five points shown represents an individual subject. Each point is represented by (b, a), where b is the slope and a is the y-intercept of the individual subject's linear lusitropic calibration curve. This curve has the form ln(Diastolic Strain Rate) = a + b / (EM)lusi. The downward-sloping function shown in this figure is referred to as the lusitropic intercept-slope tradeoff function. This tradeoff function is expected to hold for an arbitrarily large sample of individuals with good cardiovascular health and fitness. Furthermore, the fitting parameters (p, q) = (intercept, slope) of this inotropic intercept-slope trade-off function are considered to be a reasonable approximation of the universal parameters of lusitropic function in healthy myocardial tissue.

[0031] Heart failure patients are similarly expected to have their linear lusitropy calibration curve at some point (b, a) below the line shown in the trade-off function shown above. The vertical distance between this point and the trade-off function is VDlusi, which is a metric of diastolic dysfunction in heart failure.

[0032] The solution, implemented geometrically, uses easily understood linear functions to very simply define the relationship between the electrical and mechanical properties of transduction and tension development in normal, healthy myocardial tissue. The angle of a vector, or "arrow," with the x-axis can represent the systolic or diastolic characteristics of a person's disease, and the length of the arrow represents the severity of that disease compared to all healthy individuals. Alternatively, the magnitude of a deficit in inotropic or lusitropic function can be expressed as a quantity that depends only on the perpendicular distance between the point defining a given individual's calibration function in {slope, intercept} space and a downward-sloping linear function defining normal cardiac health in the same space. Inotropic and lusitropic health have similar but distinct normal population linear functions that differ only in their fitting parameters, i.e., slope and intercept, which are universal and graphically very simple and easy to understand. Current approaches only address the mechanical properties of cardiac wall motion, contraction, or stretch accompanied by loss of tension or filling.

[0033] This current state-of-the-art essentially ignores the pathophysiology, with its obvious electrical and mechanical characteristics, of a disease that is life-threatening and costly to care for. This graphical simplicity makes the electrical and / or mechanical dysfunction that leads to heart failure, when present, easily and clearly visible and easily and reproducibly quantifiable. This simplicity is similar to the ease with which a pediatrician plots a child's height and weight on a growth curve. This growth curve is universal and stratified by percentiles. The patient's height or weight is plotted as a single time point or at multiple time points over time. What becomes apparent is deviation from a universal norm, here given by a slope-intercept tradeoff function.

[0034] Current methods for quantifying systolic or diastolic heart failure are entirely mechanical in nature, ignoring important features of the disease and the relationships between them. This invention allows clinicians to detect "pre-heart failure" in a manner similar to how "pre-diabetes" is currently diagnosed, providing the opportunity for pharmacological intervention, as well as fluid, electrolyte, dietary, and exercise therapy, before a subclinical condition becomes clinical and patients require intubation, mechanical ventilation, ventricular assist devices, or cardiac transplantation. This management process can be effectively monitored and adjusted in real time using wearable technologies that measure, transmit, and process (EM)ino and (EM)lusi as described in the previous disclosure.

[0035] This preventative approach is much cheaper than traditional multiple hospitalizations for congestive heart failure decompensation, and it puts early detection and prevention squarely in the clinician's diagnostic and management toolkit.

[0036] Heart failure is generally characterized as systolic heart failure (heart failure with reduced ejection fraction, HFrEF) or diastolic heart failure (heart failure with preserved ejection fraction, HFpEF). Assume a patient undergoes a dobutamine stress test using several increasing dobutamine infusion rates at steady state. Assume further that the systolic and diastolic ln(Strain Rate) along with simultaneous values ​​of 1 / (EM)ino and 1 / (EM)lusi are measured and graphed to generate a calibration curve for ln(Systolic Strain Rate) versus 1 / (EM)ino and a calibration curve for ln(Diastolic Strain Rate) versus 1 / (EM)lusi, with slopes B and y-intercepts A calculated for each. If the systolic value (B,A) is plotted as a point on the graph in Figure 3 and the diastolic value (B,A) is plotted as a point on the graph in Figure 4, there will often be a distance between the point (B,A) and the trade-off function line on either or both the systolic and diastolic graphs. By measuring the perpendicular distance between these individual patient points and the lines parallel to the respective y-axes—the units of which are ln(sec^-1), i.e., ln(Strain Rate)—an accurate and reproducible metric of systolic heart failure from Figure 3 and diastolic heart failure from Figure 4 is obtained. This is clearly shown in Figure 3 for the systolic case and in Figure 4 for the diastolic case. Paragraphs

[0112] through

[0126] below provide additional specificity regarding the creation of individual trade-off functions for a patient's inotropic function and for a patient's lusitropic function.

[0037] In other words, if a patient has both diastolic and systolic heart failure, varying degrees of systolic heart failure, we might find that these measurably change relative to each other over time over the five-year natural course of the disease. This is an advantage over the current state of measurement and monitoring. Figures 3 and 4 show the same universal intercept vs. slope tradeoff functions for systolic and diastolic heart failure as shown in Figures 1 and 2. Here, for the first time, in addition to the universal tradeoff functions, both Figures 3 and 4 also show two distinct patients with very different levels of myocardial pathology, represented as points m and n for the inotropic case and points r and s for the lusitropic case.

[0038] Upon review, each individual's point simply represents the slope and intercept of a linear calibration curve where ln(Strain Rate) corresponds to 1 / (EM). Note that in both Figures 3 and 4, points m and r are fairly close to the universal trade-off function. While these patients may not have symptoms, this technique reveals that they have a deficit, no matter how small, that separates them from the norm.

[0039] In both Figures 3 and 4, patients represented by point n (Figure 3) and point s (Figure 4) are both suffering from symptoms of congestive heart failure. Patient n has inotropic heart failure or HFrEF (heart failure with reduced ejection fraction), and patient s has lusitropic heart failure or HFpEF (heart failure with preserved ejection fraction).

[0040] By using these graphs, representing each individual patient as a point and comparing the location of that point to a standard defined by a line, much can be learned about an individual's myocardial health, or lack thereof. In this way, individuals with subclinical inotropic or lusitropic disease, i.e., "pro-MI," can be identified in a way that is not currently possible. More importantly, the defect can be classified as inotropic or lusitropic, and the defect can be quantified accurately, reliably, and reproducibly. This represents a significant improvement over the state of the art.

[0041] Referring to Figure 3, we can illustrate how to quantify patient m's inotropy, shown at point Yino2m. We can then clearly see what normal, healthy inotropy looks like at point Yino1m, given the slope of the calibration curve. Yino1m is simply a point on the line that defines the inotropic intercept-slope tradeoff function. Note that both of these quantities, Yino1m and Yino2m, are, in the abstract, the natural logarithm of what the systolic strain rate would be at the limit where the (EM)ino interval becomes very long and approaches infinity. While this is physiologically impossible, the thought experiment of it happening is a useful metric of how well inotropic patient m actually is, given the slope of his or her calibration curve.

[0042] A well-known property of logarithms is that ln(b) - ln(a) = ln(b / a). Therefore, Yino2m - Yino1m = ln(Strain Rate2m / Strain Rate1m). However, the right-hand side of this equation is simply the natural logarithm of the ratio of intercept A of the calibration curve Y for patient m to intercept A of the calibration curve Y for a patient selected from a population of perfectly healthy individuals. The difference between Yino2m and Yino1m can thus be thought of as a kind of inotropic fraction (IF), which is equal to 1 if the patient being compared to the population is perfectly healthy and less than 1 if there is some deficit in inotropic function. IF is easily intuitively understood by clinicians, just as ejection fraction (EF) is traditionally understood. Traditionally, EF is expressed as a percentage, but that is a figure, not a substance.

[0043] Therefore, we can write IF = exp(VDino) = exp(Yino2m-Yino1m). Note that VDino is a signed quantity, and here it is less than 0. From Figure 3, it can be clearly seen that IF is slightly less than 1 for patient m, simply because the vertical distance VDino = (Yino2m-Yino1m) is slightly less than 0, and exp(0) = 1.

[0044] Now consider patient n in Figure 3. If the calibration curve for patient n has a steeper slope than that for patient m, it means that a given change in 1 / (EM)ino will produce a larger change in ln(Systolic Strain Rate) for patient n than for patient m. Testing reveals that the difference between the intercept Yino1n for a population of healthy people and the actual Y-intercept Yino2n of the calibration curve for patient n is much larger than for patient m. Patient n is in dire straits. He cannot lie in bed without feeling short of breath. With a stethoscope, we can hear rasps at the base of both lungs. He has congestive heart failure and requires diuretic therapy.

[0045] Again, the slope and intercept of the calibration curve for patient m, shown as points in Figure 3, summarize and include all variables such as age, sex, height, and weight that are typically thought to be used in determining the value of a quantity such as the inotropic fraction, IF.

[0046] Switching to Figure 4, we can clearly see how measuring lusitropy proceeds by analogy with measuring inotropy. Figure 4 shows the lusitropy intercept-slope tradeoff function for all possible calibration curves measured in a population of subjects known to have good myocardial health. For two individuals, r and s, the slopes and intercepts of their calibration curves are plotted in lusitropy {slope, intercept} space as shown. Patient r has a slope that deviates from the lusitropy intercept-slope tradeoff function that empirically defines myocardial health. The vertical difference (i.e., vertical distance) VDlusi = (Ylusi2r - Ylusi1r) is a number less than zero. Patient r has no symptoms of CHF. However, if patient r is considered to be "Pre-HFpEF" and requires monitoring, he or she should be observed by a healthcare professional. And it's simply because, with hands-on management, lifestyle changes, exercise, hydration, and afterload reduction agents, it's possible to bend the patient's morbidity curve, adding years to the patient's life and lives to their age. All the patient and their clinician have to do is absorb the information from the monitoring system over time, over the years, and "close the loop," just as they do with diabetes management. This can benefit the patient and also save a lot of money from health care by reducing hospitalizations, which increase in number over time like a bouncing stone.

[0047] Therefore, the lusitropic fraction LF can be written as LF = exp(VDlusi) = exp(Ylusi2r - Ylusi1r). Note that, graphically, VDlusi is a signed quantity whose magnitude is small and less than 0. Therefore, exp(VDlusi) approaches 1, but is less than 1, which makes physiological sense since patient r is asymptomatic and in a preclinical state.

[0048] Now consider patient s, shown below patient r in Figure 4. Patient s is critically ill. He or she can barely breathe, a problem that worsens when he or she attempts to walk. The patient's lungs are congested, making it difficult to oxygenate them with room air. However, this patient is ill through a pathophysiologically different pathway and an entirely different mechanism than patient n in Figure 3. This patient suffers from HFpEF, a condition in which the myocardium loses tension and fails to relax when it should. This is a process that requires ATP, which is used to actively transport Ca++ ions from the cytosol, away from tropinin receptors, and back into the sarcoplasmic reticulum compartment against their concentration gradient in anticipation of the next systole. This process requires oxygen. The medical management of patient s may differ from that of patient n over time as new approaches to this problem emerge.

[0049] But the key point is that by using these (EM)ino and (EM)lusi measurements of patients who are not necessarily hospitalized, who are ambulatory and living freely in the community, and by using these simple graphical and algebraic notations, we can see the difference, we can quantify that difference, and we can track that difference. And by "closing the loop," we can make care easily accessible to everyone on their mobile phone, while achieving better outcomes at a lower cost.

[0050] To complete this analogy, the lusitropic fraction LF can be written as LF = exp(VDlusi) = exp(Ylusi2s - Ylusi1s). For this patient s, the vertical difference (or delta) VDlusi is significant, on the order of (-1.5 - (-0.5)) = -1.5 + 0.5) = -1 natural logarithmic units. Thus, for patient s, LF = exp(-1) = 1 / 2.718 = 0.368, which intuitively "feels" like a low ejection fraction, except that it represents a clinical case of low lusitropic state, not inotropic state. Note: The "VD" in VDino and VDlusi stands for "vertical distance," or more appropriately, "vertical displacement." This is because starting with the smaller Y-intercept from the subject and then subtracting the larger Y-intercept from the population is important, resulting in a number that is less than zero if the subject has some disease and zero if the subject does not have any disease. IF and LF, or inotropic and lusitropic fractions, are the ratios of the strain rate observed from an individual's Y-intercept to the strain rate from the Y-intercept of a healthy population, at a given slope of the individual's calibration curve. Thus, these fractions are measures of ejection fraction, but for all possible values ​​of 1 / (EM).

[0051] In summary, when we refer to the lusitropic fraction as LF = exp(VDlusi) or the inotropic fraction as IF = exp(VDino), all we are saying is that both IF and LF are fractions of the expected strain rate at the y-intercept from a healthy population, expressed in terms of the observed strain rate at the y-intercept derived from a calibration curve of individuals whose cardiac function has been individually assessed. VDino and VDlusi are signed quantities less than or equal to zero, and their measurements are clearly illustrated and shown in Figures 3 and 4.

[0052] It is expected that some patients will exhibit a "mixed" picture of inotropic and lusitropic dysfunction. By combining IF and LF in some physiologically meaningful way, the overall myocardial functional fraction, TMFF, can be quantified. Because inotropy and lusitropy are distinct processes that are mutually inverse but physiologically and pathophysiologically coupled, it can be useful to think of LF and IF as mutually perpendicular vectors acting on the same organ. Thus, IF and LF can be vectorially added, such that TMFF = [(IF)^2 + (LF)^2]^1 / 2.

[0053] The only drawback to this approach is that in perfect inotropic health, IF = 1, LF = 1, and TMFF = 2^(1 / 2), approximately 1.414. Therefore, it makes intuitive sense to normalize the magnitude of this vector by dividing by the square root of 2, so that a patient in perfect inotropic and lusitropic health would have a TMFF = 1.

[0054] Thus, we can write TMFF = {[(IF)^2 + (LF)^2]^1 / 2} / (2^(1 / 2)). The magnitude of this vector indicates how large the heart failure problem is being treated. Over time, it reveals objective changes in the severity and characteristics of the condition, like a pediatrician's growth curve. The component metrics of IF and LF can be tracked separately over time.

[0055] Alternatively, (IF,LF) can be plotted on the (x,y) axes and added vectorially, noting the angle between the x-axis and the vector. This angle quantitatively describes, graphically, the contribution of diastolic function (y-axis) and systolic function (x-axis) to the clinical picture of heart failure, and which form of myocardial dysfunction, systolic or diastolic, predominates in the clinical picture, helping to provide a quantitatively more robust, more accurate, and more reproducible basis for the assessment of heart failure than is available with conventional techniques.

[0056] Here's another way to look at Figures 1 and 2: If you plot them on the same graph, the lusitropic intercept-slope tradeoff function is a downward-sloping line drawn 0.92863-(-0.03091)=0.96773ln(sec^-1) above the inotropic intercept-slope tradeoff function. This is on the order of 1 logarithmic unit. This is highly significant.

[0057] The slopes in Figures 1 and 2, i.e., -4.61361 and -4.95698, are approximately equal within experimental error. Note that in the inotropic case, the (EM)ino interval ends at the "M" event, the first derivative peak. This is not the case in the lusitropic case, where the "M" event is a standard oscillatory cardiogram. This is arbitrary and heuristic. If the same notation were used in the inotropic case, the "M" event would occur several milliseconds later, the (EM)ino interval would be longer, and 1 / (EM)ino would be smaller than with a standard oscillatory cardiogram signal. This would translate the x-axis of Figure 1 to a smaller dimension, increasing the magnitude of the upstroke (or "downstroke" in this case, since the direction is negative) in the stroke, thereby increasing the magnitude of the negative slope. This, in turn, would move the slope of Figure 1 closer to matching the slope of Figure 2.

[0058] The significant difference between the intercepts in Figures 1 and 2 suggests that in healthy conditions, the strain rate is always significantly faster during diastole than during systole. Normal hearts relax faster than they contract. This is consistent with unpublished data obtained in 2011, in which we used a pressure catheter in the left ventricle of septic pigs to compare the first derivative of LVP during systole with that during diastole over time, before and after placement of an E. coli septic clot in the pig's peritoneum. In healthy conditions, the magnitude of LVP'max during diastole is greater than during systole, but the direction of movement is reversed. As sepsis progresses, the situation reverses. That is, the ratio of LVP'max diastole / LVP'max systole starts out greater than 1 and then exceeds 1 as the pig's heart attempts to compensate for the decline in SVR and blood pressure associated with the progression of sepsis, resulting in a high cardiac output. (See ASA Abstracts Hirsh, Torjman, Goldfarb, 2011) (See http: / / www.asaabstracts.com / strands / asaabstracts / abstract.htm?year=2011&index=4&absnum=5333.)

[0059] Furthermore, it is clear that LVP'max, the maximum value of the first derivative of left ventricular pressure, and both systolic and diastolic strain rates are related a priori because their natural logarithms are both linearly proportional to 1 / (EM) (unpublished data). Two quantities that are both linearly proportional to 1 / (EM) must also be linearly proportional to each other. Because ln(LVP'(t)max) and ln(Strain Rate) have different units, only the fitting parameters, the slope and intercept, will change.

[0060] Furthermore, to specifically clarify this process, since the cardiology community recognized that loss of lusitropic function can cause congestive heart failure, it has become necessary to rename or "rebrand" the more classic form of congestive heart failure (CHF) due to loss of systolic or inotropic function.

[0061] Historically, congestive heart failure due to loss of systolic function—"pump failure"—was called "dropsy." It was treated with drugs like digitalis, extracted from the leaves of the foxglove plant, which increased the tension of the left ventricle as it contracted. Nowadays, CHF is more often treated with afterload-reducing drugs like calcium channel blockers, which lower blood pressure and slow the heart's work. This prevents fluid from backing up into the lungs and causing congestion. And patients no longer feel short of breath, especially when lying on their backs in bed.

[0062] At this point, heart failure due to loss of systolic function is called heart failure with reduced ejection fraction (HFrEF). A reduced ejection fraction (which increases the diffusion barrier to oxygen transport across the dilated capillaries of the lung alveoli, resulting in loss of oxygen saturation, congestive symptoms, shortness of breath, and fluid-filled alveoli that make the patient feel as if they are drowning) is clearly visible to any clinician equipped with a 2D echocardiogram. A 2D echocardiogram also provides end-diastolic and end-systolic views. The echocardiographer then expresses the ratio of the left ventricular cross-sectional area in these two views as a fraction, end-systolic / end-diastolic. This is the "ejection fraction (EF)." In systolic heart failure, EF is reduced.

[0063] In diastolic heart failure, the left ventricle "remodels" (remodels). It can become thicker, losing its compliance (dV / dP). It becomes stiffer as it fills with blood. It does not release tension or pressure from end-systole (dP / dt) as quickly as a healthy heart. In this situation, the patient is said to have heart failure with preserved ejection fraction (HFpEF). Moreover, ejection fraction is a simple, intuitive measurement, and its preservation or decline helps distinguish between the two types of heart failure: systolic (inotropic) and diastolic (lusotropic). Before 2D ultrasound machines became almost as common as laptop computers, clinicians did not worry about this distinction.

[0064] The ejection fraction is an imperfect proxy for contractility. Under high afterload, such as with phenylephrine, the ejection fraction is low even when contractility is normal. Conversely, under low afterload, such as with shock, the ejection fraction is high or even normal, even when contractility is deficient due to circulating bacterial toxins. A normal ejection fraction is 55% or greater.

[0065] The present disclosure helps provide a highly accurate and reproducible method for quantifying whether heart failure is lusitropic, inotropic, or both, and the precise extent to which the condition is due to inotropic dysfunction. The following procedures are used to achieve this highly accurate categorical diagnostic determination: (1) A dobutamine stress test is performed on patients with suspected or subclinical heart failure who have not yet deteriorated enough to cause symptoms but who will worsen over time if left untreated. In patients where this is possible, exercise can be used instead of dobutamine. Detecting subclinical heart failure in this manner serves as a screening tool, which is not possible with the current state-of-the-art technology. (2) Simultaneous 2D echocardiographic measurements of systolic and diastolic strain rates and corresponding (EM)ino and (EM)lusi measurements are performed at rest over several consecutive heartbeats, and this is repeated at (at least) two increasing levels of dobutamine infusion rate. Alternatively, patients can be asked to exercise as much as possible on an inclined treadmill at various speeds in a timed manner, such as the Bruce protocol. Simultaneous measurements of strain rate and (EM) are taken after the patient stops maximal exercise and quickly lies down. This generates a systolic function calibration curve with ln(abs(Systolic Strain Rate)) on the Y-axis and 1 / (EM)ino on the X-axis. This is a monotonically increasing line with a Y-intercept A1 and a slope B1. (3) This also generates a diastolic function calibration curve with ln(Diastolic Strain Rate) on the Y-axis and 1 / (EM)lusi on the X-axis. This is a monotonically increasing line with a Y-intercept A2 and a slope B2. (4) Next, (B1, A1) is plotted on the universal inotropic intercept-slope tradeoff curve described earlier in this disclosure. This will likely be below the downward sloping line relating intercept to slope for all healthy myocardial tissue. Measure the vertical distance VDino between (B1,A1) [(slope, intercept)] and the inotropic intercept-slope tradeoff function.The signed quantity VDino, whose units are ln(sec^-1) and are < or = 0, is a metric of the absolute magnitude, if any, of the decline in inotropic function from normal, healthy conditions in this particular patient. VDino represents the natural logarithm of the ratio of an individual patient's inotropic function to that of healthy, normal people with the same slope as that patient's calibration function. The signed quantity VDino is calculated by subtracting the intercept of the calibration curve for healthy, normal people given by the trade-off function for a given slope from the intercept A1 of the individual patient's calibration curve. Due to the properties of logarithms, ln(a / b) = ln(a) - ln(b), the signed difference VDino, exponent of the transcendental number e (approximately 2.178...), yields the ratio of the individual patient's inotropic function to that of a normal, healthy heart with the same slope as that patient's calibration curve. Since VDino is negative, i.e., 0, and the inotropic fraction is IF = exp(VDino), it follows that IF must be <1 or =1. VDino may exist empirically and be measurable, but it is still subclinical. If so, detecting and quantifying it represents an opportunity to intervene on a preventive basis and thereby maintain inotropic function over the long term. (5) Analogously, the same procedure is performed for lusitropy. On the universal lusitropy intercept-slope trade-off function, the (slope, intercept) point (B2,A2) from the calibration curve is plotted. Analogously, it lies below the line. Measure the vertical distance VDlusi (signed < or =0, as explained above) from the universal lusitropy intercept-slope trade-off function curve to point (B2,A2), with the slope given by the slope of the individual patient's lusitropy calibration curve.

[0066] This signed quantity, VDlusi, is measured in units of ln(sec^-1) and is an accurate and reproducible metric of lusitropic decline, if this particular patient suffers from it. Again, it may exist empirically, but it is still subclinical. As before, we define this lusitropic ratio as LR = exp(VDlusi). Since VDlusi is <0 or =0, it follows that exp(0) = 1, and therefore exp(VDlusi) <1 or =1. LR cannot be greater than 1, because physiologically, VDlusi is 0 only in perfect health, and is less than 0 in other states.

[0067] If a patient's heart failure is purely systolic (inotropic), then VDlusi = 0. If a patient's heart failure is purely diastolic (lusotropic), then VDino = 0. Essentially, one can assume that pure systolic or pure diastolic heart failure does not exist. Rather, all cases of heart failure are likely to be "mixed" systolic and diastolic at some level. However, one or the other may be very small or negligible.

[0068] As mentioned elsewhere in this specification, the overall picture of HF can be expressed using TMFF = {(VDino^2 + VDlusi^2)^1 / 2} / (2^(1 / 2)). To calculate VDino or VDlusi, the higher intercept from a universal intercept-slope trade-off function at a given slope must be subtracted from the lower intercept obtained from the calibration curve of the individual patient whose condition is being characterized and quantified. That is why VDino and VDLusi are negative numbers and equal 0 only if the patient being evaluated has healthy myocardium.

[0069] In another example, let {VDino, VDlusi} be a two-dimensional (x, y) vector space, and let (VDino, VDlusi) be a point in that space that describes a particular patient with heart failure according to the procedure described above.

[0070] Then (VDino,VDlusi) describes the heart failure vector HF in the {VDino,VDlusi} space, which unambiguously quantifies in a clear and precise way the characteristics and magnitude of heart failure in a particular patient at a certain point in time.

[0071] The vector HF can be normalized as HF=(VDino^2+VDino^2)^1 / 2. HF provides a metric for the patient's overall heart failure disease magnitude. The vector HF makes an angle THETAhf with the x-axis, and VDino is such that tan(THETAhf)=VDlusi / VDino. (Recall that VDino is represented on the x-axis in {VDino,VDlusi}, and VDlusi is represented on the y-axis.)

[0072] Furthermore, if THETAhf is <45 degrees, the patient's HF is said to be systolic dominant, and if THETAhf is >45 degrees, the patient's HF is said to be diastolic dominant.

[0073] If THETAhf=45 degrees, the patient's HF is characteristically half systolic and half diastolic at the time the measurement is taken.

[0074] It is clear that the magnitude and angle of the HF vector changes over the normal 5-year natural history of heart failure, and clinically useful lessons can be learned from following its trajectory over time in {VDino,VDlusi} space. By truly following it, we can determine whether our interventions and treatments have "moved the needle" to the patient's benefit. This innovation will also be of interest to pharmaceutical companies seeking to develop new, money-saving solutions within this costly clinical space.

[0075] The above will help to place the classification and quantification of systolic and diastolic HF on a firmer and more definitive quantitative basis than is currently the case in the state of the art.

[0076] While the preferred embodiments of the invention disclosed herein utilize the electromechanical systems and specific methods disclosed in the '665 patent, the '300 patent, and the '047 application, it is important to understand that other system embodiments are encompassed herein and that the present application is not limited to the preferred embodiments. Similarly, the specific methods disclosed in the '665 patent, the '300 patent, or the '047 application can be utilized to provide an electronic signal that can be used to measure the vertical distance from a downward sloping line of a universal Y-intercept versus slope plot (a trade-off function, as in FIGS. 1 and 2) based on the results of a large number of individual calibration curves from healthy, normal subjects, to the point (b, a) = (slope, intercept) that represents the individual calibration curve of a patient with an inotropic or lusitropic disorder being evaluated.

[0077] To reiterate, the calibration curve takes the form ln(Strain Rate) = a + b / (EM), where (EM) is (EM)ino for the inotropic function and (EM)lusi for the lusitropic function. Strain rate is similarly inotropic, with 1 / (EM)ino, or lusitropic, with 1 / (EM)lusi. If the argument of the function is <0, as is conventionally the inotropic case, then ln(Strain Rate) is undefined, and only the magnitude of the strain rate is used.

[0078] It should be noted that this disclosure is not intended to limit the scope of the inventions described herein to any single embodiment or application. For example, but not by way of limitation, in addition to the more directly specific applications described herein, other applications are expected to be encompassed for solving other cardiac-related problems, such as the detection of phonocardiograms and vibrocardiograms in morbidly obese individuals. Solutions to these problems currently require amplification and advanced digital filtering and signal processing. Another solution is to implant an accelerometer subcutaneously, closer to the rib cage, thereby improving the signal-to-noise ratio. Yet another solution is to perform transesophageal echocardiography and dobutamine stress testing under anesthesia. Another approximation of the vibrocardiogram, derived from millimeter-wavelength radiofrequency currents that readily penetrate the chest wall of morbidly obese patients and reflect from the cardiac surface, may serve as a basis for the timing and amplitude of "M" events in the (EM) interval. Yet another approach is to use a precordial Doppler ultrasound transducer, the output of which is fed to a frequency-to-voltage converter. The resulting waveforms can also be used to obtain the timing and amplitude of "M" events in the systolic and diastolic (EM) intervals. Both of these radiofrequency and Doppler ultrasound techniques could help provide a solution to the problem of monitoring patients suffering from morbid obesity. Both are still "wearable." The trade-off here is that both radiofrequency and Doppler ultrasound techniques require a continuous source of external energy directed at the patient's heart, necessitating larger batteries or alternative power sources. However, good engineering and clever signal processing and amplification techniques may eliminate the need for these alternative techniques, allowing us to rely entirely on the natural signals emitted by the human heart.

[0079] It is important to understand that the universal intercept-slope tradeoff function describes a useful property of a particular set of linear fitting parameters, (slope, intercept), for an inotropic or lusitropic calibration curve. The calibration curve is a linear function relating exclusively mechanical cardiac performance metrics (such as the maximum (systolic) and minimum (diastolic) values ​​of the first derivative of the left ventricular pressure curve obtained by invasive catheterization, or myocardial strain rate obtained from 2D transthoracic echocardiography) to noninvasively obtained electrical-mechanical metrics such as 1 / (EM)ino and 1 / (EM)lusi.

[0080] If the exclusively mechanical performance data from systole or diastole are logarithmically transformed using natural logarithms and placed on the y-axis, and 1 / (EM)ino from systole or 1 / (EM)lusi from diastole on the x-axis, the relationship between these two variables, one exclusively mechanical (y-axis) and the other exclusively electro-mechanical (x-axis), is linear. Specifically, systolic events and metrics are exclusively related to 1 / (EM)ino. Diastolic events are exclusively related to 1 / (EM)lusi. These linear relationships have two fitting parameters (slope and intercept). The teachings of the present invention regarding the construction of inotropic and lusitropic calibration curves are explained in detail in paragraphs

[0116] to

[0126] of this specification.

[0081] There are two universal intercept-slope tradeoff functions: one for systolic inotropy and one for diastolic lusitropy. The universal inotropic intercept-slope tradeoff function consists solely of (slope, intercept) data obtained during systole. And the universal lusitropic intercept-tradeoff function consists solely of (slope, intercept) data obtained during diastole. There is no "temporal mixing" of systolic and diastolic data. These are called "tradeoff functions" because, according to a simple linear law, as the slope rises, the intercept falls, and vice versa. A change in one of these quantities is effectively "tradeoffed" with an opposite change in the other.

[0082] These universal intercept-slope tradeoff functions describe only the behavior of a perfectly healthy heart, unburdened by any disease, such as myocardial hypertrophy, coronary artery disease, heart failure or cardiomyopathies, or valvular disease. These universal tradeoff functions have useful properties and can serve as benchmarks for classifying, determining, and measuring myocardial pathology. In particular, they can serve as benchmarks for classifying, determining, and measuring heart failure, regardless of whether one is classifying and measuring inotropic heart failure with reduced ejection fraction (HFrEF) or lusitropic heart failure with preserved ejection fraction (HFpEF). Specifically, the universal inotropic intercept-slope tradeoff function consists only of the fitting parameters (slope, intercept) of a linear inotropic calibration curve for many individuals who enjoy excellent cardiac health. Similarly, the universal lusitropic intercept-slope tradeoff function consists of the fitting parameters (slope, intercept) of a lusitropic calibration curve for many individuals who enjoy excellent cardiac health. What is extraordinary about the universal intercept-slope tradeoff function is that it is obtained regardless of the subject's age, sex, height, or weight, as long as the subject enjoys excellent cardiac health. To put this another way, within practical limits, given cardiac health, the slope and intercept of the universal intercept-slope tradeoff function itself are invariant with respect to the subject's age, sex, height, and weight, for both inotropic and lusitropic cases.

[0083] The process by which a universal inotropic or lusitropic intercept-slope trade-off function is constructed follows quite naturally from the above definition:

[0084] Find N willing subjects who have good exercise tolerance and enjoy excellent cardiac health by all reasonable clinical standards and metrics. N is high enough to produce meaningful statistics and allow for the calculation of standard errors of the mean and 95% confidence intervals. N subjects should also be diverse in a sense, including age, height, weight, sex, and race. The goal is to produce approximate results that are universally usable for all of humanity.

[0085] For each subject, the inotropic strain rate during systole and the lusitropic strain rate during diastole are measured.

[0086] Simultaneously, the (EM)ino and (EM)lusi intervals are measured during the systolic and diastolic portions of each cardiac cycle, respectively. For details on how these time intervals are measured, see paragraphs

[0116] to

[0126] herein.

[0087] Each subject then performs an exercise protocol according to a clinically acceptable protocol, such as the Bruce protocol on a treadmill or stair climbing at a metronome-controlled rate. Alternatively, a dobutamine stress test can be performed using a drug infusion rate of several mg / kg / min to create a series of graded hemodynamic steady states according to the clinical protocol for such a test. If heart rate does not increase appreciably, atropine can be administered at its highest dose.

[0088] At each steady state of increased exercise volume or increased drug infusion rate, m consecutive heartbeats are measured simultaneously, where "m" is a number on the order of 10 consecutive heartbeats, intended to allow averaging of (EM)ino and (EM)lusi over respiratory variability over at least two consecutive breaths.

[0089] With each increment in exercise volume or drug infusion rate, the inotropic and lusitropic strain rates are measured simultaneously, as well as both the (EM)ino and (EM)lusi time intervals characteristic of the increments in inotropic and lusitropic function that occur with each increment in exercise volume or drug infusion rate.

[0090] Here, we separate the inotropic strain rate from the lusitropic strain rate, and the (EM)ino interval from the (EM)lusi interval. We then pair the inotropic strain rate with its concurrent (EM)ino interval, and the lusitropic strain rate with its concurrent (EM)lusi interval.

[0091] For each subject, this creates a set of m ordered pairs ((EM)ino, (inotropic strain rate)) and ((EM), lusitropic strain rate)) at each hemodynamic steady state, regardless of whether the hemodynamic steady state was created by exercise or drug infusion.

[0092] Let there be i steady-state levels of inotropic and lusitropic function, including one baseline level at rest and four different exercise or drug infusion rate levels, where i is a number greater than 1 and small on the order of 5.

[0093] For the ith exercise or drug infusion level, average all m consecutive values ​​of (EM)ino and separately average all m consecutive values ​​of (EM)lusi to obtain AVG(EM)ino = SUM[(EM)ino] / m, and AVG(EM)lusi = SUM[(EM)lusi] / m.

[0094] Next, for each i-th value of AVG(EM)ino and AVG(EM)lusi, calculate the reciprocals of 1 / (AVG(EM)ino)i and 1(AVG(EM)lusi)i.

[0095] This gives, for each healthy subject, i pairs of averaged 1 / (EM)ino and 1 / (EM)lusi, where each i pair represents a different, stepwise increasing hemodynamic steady state.

[0096] Here, for each ith 1 / (AVG(EM)ino), we pair it with m simultaneous and consecutive inotropic strain rates measured by 2D echocardiography that correspond to the same heartbeats used to measure 1 / (EM)ino.

[0097] Let AVG(Strain Rate)ino = SUM(Strain Rate) / m.

[0098] Now, taking the natural logarithm of the i-th average strain rate, for the i-th exercise level, we obtain ln(AVG(Strain Rate)ino)i = ln(SUM(Strain Rate)ino) / m)i.

[0099] Next, the same is done for m simultaneous and consecutive lusi-relaxation strain rates at the i-th exercise level: (ln(AVG(Strain Rate)lusi)i = ln(SUM(Strain Rate)lusi / m)i

[0100] Here, there are two ordered pairs of i data, one for inotropy and one for lusitropy, {1 / (AVG(EM)ino)i,(ln(AVG(Strain Rate)ino)i} and {(1(AVG(EM)lusi)i),ln(AVG(Strain Rate)lusi)i}.

[0101] For every i-th change in hemodynamic steady state, plot (1 / (AVG(EM)ino)i on the x-axis and (ln(AVG(Strain Rate)ino)i on the y-axis. This will generate an inotropic calibration curve for this particular healthy individual who is a member of the set of N healthy individuals.

[0102] Finally, for each i-th change in hemodynamic steady state, plot (1 / (AVG(EM)lusi)i on the x-axis and (ln(AVG(Strain Rate)lusi)i on the y-axis. This will generate a lusi-relaxation calibration curve for this particular healthy individual who is a member of the set of N healthy individuals.

[0103] Construct inotropy and lusitropy calibration curves for all N healthy subjects in a diverse sample. Initially, restrict membership in set {N} to healthy adults only. Infants and children can be studied subsequently. It can be inferred that as people grow from prematurity to infancy to adulthood, there are developmental changes in inotropy and lusitropy, and their relationship to changes in (EM) that can be assessed using these methods, which may need to be adjusted for developmental age by percentile stratification as a function of chronological age since birth, rather like a pediatrician's height-weight growth curves.

[0104] Each linear calibration curve has two fitting parameters, which can be written as an ordered pair (slope, intercept). There are N pairs of inotropic functions and N pairs of lusitropic functions.

[0105] Now plot the ordered pair (slope, intercept) of N inotropic functions in (x,y) space. This is the universal inotropic intercept-slope tradeoff function.

[0106] Next, the ordered pairs (slope, intercept) of the remaining N lusitropy functions are plotted in (x,y) space, which is the universal lusitropy intercept-slope tradeoff function.

[0107] On the x-axis of both these intercept-slope trade-off functions, the units of slope are ln(Strain rate) / (1 / sec)=sec*ln(sec^-1).

[0108] The y-axis of both these intercept-slope trade-off functions has units ln(Strain Rate)=ln(sec^-1) because strain is itself a dimensionless quantity.

[0109] Both of these functions, described in paragraphs

[0105] and

[0106] , are downward-sloping linear functions of the form intercept = PQ(slope). The constants (q,p)ino are the slope and intercept of the universal inotropic intercept-slope tradeoff function. The constants (q,p)lusi are the slope and intercept of the universal lusitropic intercept-slope tradeoff function. We assume that (q,p)ino and (q,p)lusi are discoverable and universal constants for all healthy humans, and that, with N large enough, we can calculate meaningful and useful standard errors of the mean for them, along with 95% confidence intervals. In a sense, (q,p)ino and (q,p)lusi are the Vitruvian constants of robust cardiac health, the so-called "Cor sanum in corpore sano" (Latin), or "A healthy heart in a healthy body." (See Figures 1 and 2.)

[0110] We further hypothesize that these universal intercept-slope tradeoff functions can be used as benchmarks for myocardial health, and that if the slopes and intercepts of an individual patient's calibration curves do not line up when mapped with these universal functions, and if there is a clear gap between the individual's point and the universal line, this indicates functional myocardial pathology, either inotropic or lusitropic, or both. In premature infants, infants, and children, this same deviation, or "clear gap," can help accurately quantify the age-related extent of cardiac development toward normal adult function. Such deviations can be normalized as a function of birth age and used diagnostically by pediatric cardiologists to individually assess patients in a manner similar to the way general pediatricians use growth curves.

[0111] Using this approach, myocardial pathology can be detected, quantified, and monitored early in the natural history of heart failure. Patients whose intercepts are low by more than a 95% confidence interval near the value given by the universal intercept-slope tradeoff function but who are asymptomatic can be considered to have pre-heart failure. Reference to the inotropic or lusitropic intercept-slope tradeoff function will reveal whether the pre-heart failure is of the inotropic or lusitropic type.

[0112] As mentioned above, the present invention teaches the importance of measuring the electrical event "E" and the mechanical event "M" in the electrical-mechanical intervals of the (EM)ino and (EM)lusi of an individual study subject, i.e., an individual patient, during systole and diastole, respectively.

[0113] Paragraphs

[0079] to

[0111] above disclose how the time intervals for (EM)ino and (EM)lusi can be used to generate calibration curves for individual study subjects, i.e., individual patients. They also disclose how, starting from a set of N individuals, the (slope, intercept) fitting parameters of those individual calibration curves can be pooled to generate universal inotropy and lusitropy intercept-slope trade-off functions, which can then be used by clinicians to evaluate individual patients.

[0114] An individual's inotropic calibration curve allows one to determine the natural logarithm of myocardial systolic strain rate, given by 1 / (EM)ino. A separate lusitropic calibration curve allows one to determine the natural logarithm of myocardial diastolic strain rate, given by 1 / (EM)lusi. To obtain the actual strain rate from its natural logarithm, all one has to do is raise the base of the natural logarithm, e (approximately 2.718...), to the power equal to the natural logarithm of the strain rate.

[0115] The reciprocals of the time between the electrical event of a signal and the accompanying mechanical event, 1 / (EM)ino and 1 / (EM)lusi, can be intuitively understood as the "rate of electrical-mechanical transduction" and "rate of electrical-mechanical detransduction" for an average cardiomyocyte or portion of myocardial tissue. Each calibration curve describes the range over which myocardial strain rate is an exponential function of the "rate of electrical-mechanical transduction" in the inotropic case and the "rate of electrical-mechanical detransduction" in the lusitropic case.

[0116] We now clarify how (EM)ino and (EM)lusi are calculated. These calculations follow intuitively from consideration of the Wiggers diagram, mentioned herein, which depicts the precise physiological choreography between critical electrical events on the electrocardiogram (ECG) and subsequent mechanical events. During systole, the electrical QRS complex is seen immediately preceding the rapid closure of the mitral valve and the report of the S1 heart sound on the phonocardiogram. During diastole, the electrical T wave is seen to quickly follow the QRS complex, and the onset of the T wave is followed by rapid closure of the aortic valve, followed by the S2 heart sound. Both the S1 and S2 heart sounds have clear, detectable event analogs on the vibratory cardiogram.

[0117] The relationship between "E" and "M" in the (EM) interval is that of a preceding electrical event relative to a subsequent causally related mechanical event, in the same sense that lightning precedes thunder. Mechanical "M" in the (EM)ino or (EM)lusi interval is not limited to events in a phonocardiogram or vibratory cardiogram. "M" can be an event in a ventricular pressure wave or its first (or higher) time derivative. "M" can be an event in a peripheral arterial pressure wave or its time derivative at a predetermined distance from the aortic valve. "M" can be an event in a Doppler signal from a 1 MHz ultrasound transducer placed in an anatomically standardized location above the left ventricle, its output applied to a frequency-to-voltage converter or its time derivative. "M" can be a millimeter-wavelength radio signal reflected from the surface of the heart through the chest wall, its Doppler shift also passed through a frequency-to-voltage converter or its time derivative. Similar Doppler measurements can be made using near-infrared light-emitting diodes. The important thing is that "M" is a signal event in any metric of mechanical heart wall motion or blood motion. These last three examples may be advantageously used to obtain useful signals from obese patients. The trade-off is that these last three examples require an external power source. Regardless of how precisely "M" is measured, the linear equation relating (EM) to the natural logarithm of the mechanical activity in the calibration curve will still work. Therefore, the linear equation for the intercept and slope trade-off function will also continue to work. The only difference is that the values ​​of the constants—slope and intercept—will change to accommodate the different units of "M."

[0118] For (EM), the "E" event can be the Q wave of a single lead II ECG (electrocardiogram). In another example, the ECG can be differentiated twice with respect to time, thereby amplifying and inverting the Q wave, allowing the signal to usefully rise above noise, yielding the time of the Q'max event (see Figure 1). This Q wave or Q"max event is the turning point where the myocardium fully and irreversibly "commits" to depolarization and subsequent contraction. Using appropriate software, the time of this event is extracted from the real-time ECG data stream for each heartbeat. The "Q wave" or Q"max is understood to be the time of the prominent electrical event that precedes ventricular contraction.

[0119] The "M" event, whose time is indicated by Mino, can be extracted from electrocardiogram data substantially near the time of the S1 sound, which coincides with the rapid closure of the mitral valve as the left ventricle begins to contract rapidly and forcefully. This is evident from an examination of the Wiggers diagram referred to herein. The S1 heart sound is a complex process of finite duration. To pinpoint the exact time indicating M, it is useful to low-pass filter the audio signal obtained from the stethoscope to frequencies below the order of 100 Hz. A large amplitude peak will appear, which can serve as the "Mino." In another embodiment, the low-pass filtered phonocardiogram can be differentiated with respect to time, and a very large peak, whose time can serve as the Mino, will appear near the time of mitral valve closure. The time of the peak of the first derivative will necessarily be immediately before the peak of the undifferentiated phonocardiogram signal.

[0120] Alternatively, or simultaneously, it is possible to identify oscillatory cardiogram events occurring near the time of S1 in the phonocardiogram. When the ECG is filtered below frequencies on the order of 100 Hz, a distinct amplitude peak appears. The time of this peak can be used to represent Min. The low-pass filtered ECG can be advantageously differentiated with respect to time, resulting in several very sharply defined acceleration peaks. The rate of acceleration change per unit time, formally defined as "jerk" or dA / dt, is usefully located near the time of mitral valve closure and can also serve as Min. The time of the peak in the first derivative necessarily immediately precedes the peak of the undifferentiated oscillatory cardiogram signal.

[0121] Considering the above, since Mino always occurs after Q"max, and the timestamp Mino is always greater than the timestamp Q"max, we can simply define (EM)ino = (Mino - Q"max). By definition, (EM)ino > 0.

[0122] For (EM) lusi, the "E" event requires differentiating the ECG twice with respect to time, which amplifies and inverts the T wave, resulting in two distinct peaks on either side of the inverted T wave dip. See Figure 5. The first (leftmost) second derivative peak, just before the inverted T wave dip, usefully raises the signal above noise and indicates the time of the T" event, which is the turning point at which the myocardium fully and irreversibly "commits" to repolarization and the subsequent relaxation of systolic tension. Using appropriate software, the time of this T" event is extracted from the real-time ECG data stream for each heartbeat. T" is understood to be the time of a significant electrical event preceding ventricular relaxation.

[0123] The "M" event of (EM)lusi, whose time is designated Mlusi, can be extracted from electrocardiogram data substantially near the time of the S2 sound, which coincides with the rapid closure of the aortic valve. Note that the left ventricle relaxes from its peak and pressure begins to drop just before the aortic valve closes. This is evident from an examination of the Wiggers diagrams referred to herein. The S2 heart sound is a complex process of finite duration. To accurately identify the time indicating Mlusi, it is useful to low-pass filter the audio signal obtained from the stethoscope below a frequency on the order of 100 Hz. A large amplitude peak that can serve as Mlusi will appear. Alternatively, the low-pass filtered phonocardiogram can be differentiated with respect to time, and a very large peak, whose time can serve as Mlusi, will appear near the time of aortic valve closure. The time of the peak of the first derivative necessarily lies just before the peak of the undifferentiated phonocardiogram signal. In our pilot studies, we heuristically used undifferentiated phonocardiograms in cases of lusi-relaxation.

[0124] Alternatively, or simultaneously, it is possible to identify vibratory cardiogram events occurring near the time of S2 in the phonocardiogram, as shown in Figure 5. When the vibratory cardiogram is low-pass filtered below frequencies on the order of 15 Hz, a distinct amplitude peak appears. See Figure 5. The time of this peak can be used to represent Mlusi. The low-pass filtered electrocardiogram can be differentiated with respect to time. The result is several very sharply defined acceleration peaks. The rate of acceleration change per unit time, formally defined as "jerk" or dA / dt, is usefully located near the time of aortic valve closure and can serve as Mlusi. The units of "jerk" are m / sec^3. In our pilot studies, we heuristically used undifferentiated vibratory cardiograms in lusi-relaxation cases.

[0125] Considering the above, since Mlusi always occurs after T"max, and the timestamp Mlusi is always greater than the timestamp T"max, we can simply define (EM)lusi = (Mlusi - T"max). By definition, (EM)lusi > 0.

[0126] Figure 5 shows an example of a processed ECG"(t) with two QRS complexes overlaid on scaled SEISMO'(t) data. ECG"(t) is in black, and SEISMO'(t) is in red. The upper left shows Q"max and Mino during systole. The x-axis shows time in seconds: (Mino - Q"max) = (EM)ino. The diastolic T"max and Mlusi are shown in chronological order: (Mlusi - T"max) = (EM)lusi. In our pilot study, we used the first derivative SEISMO'(t) to determine Mino, and heuristically determined Mlusi using the undifferentiated SEISMO(t) [not shown].

[0127] As noted above, while the purpose of this application is to describe preferred embodiments of the present invention, this application should not be construed as denying application to similar embodiments of the system of the present invention that may be used to achieve desired monitoring and patient care results using the methods and algorithms described herein.

Claims

1. 1. A system for non-invasively detecting and quantifying systolic and diastolic heart failure in a patient, comprising: a non-invasive electronic cardiac function measuring device that provides electronic outputs related to lusitropic and inotropic electrical cardiac activity; a non-invasive mechanical cardiac function measuring device that provides mechanical outputs related to lusitropic and inotropic mechanical cardiac activity; a conversion unit connected to the mechanical cardiology measurement device for converting the mechanical output of the mechanical cardiology measurement device into an electronic output related to the mechanical cardiac activity; a computer platform comprising a processing unit, an application program, a storage means, and an output means, wherein the storage means stores a universal intercept-slope trade-off function of inotropic function based on cardiac function of a healthy patient, and a universal intercept-slope trade-off function of lusitropic function based on cardiac function of a healthy patient; a connection from an output of said electronic cardiometry device to an input of said processing device; A connection from the output of the transformation unit to the input of the processing unit Equipped with The application program (a) an inotropic electromechanical time interval of the patient's cardiac function and a lusitropic electromechanical time interval of the patient's cardiac function; and (b) an inotropic calibration curve having a slope and intercept defined for the patient's cardiac function, and a lusitropic calibration curve having a slope and intercept defined for the patient's cardiac function; digitizing and processing the input to the processing unit to determine The application program compares the inotropic calibration curve with the stored universal intercept-slope trade-off function of inotropic function and compares the lusitropic calibration curve with the stored universal intercept-slope trade-off function of lusitropic function to evaluate the patient's myocardial health or myocardial pathology.

2. The system of claim 1 , wherein the assessment is the detection and quantification of systolic heart failure.

3. The system of claim 2 , wherein the patient is not symptomatic.

4. 10. The system of claim 1, wherein the assessment is the detection and quantification of diastolic heart failure.

5. The system of claim 4 , wherein the patient is not symptomatic.

6. 1. A method for non-invasive detection and quantification of systolic and diastolic heart failure in a patient, comprising: positioning a non-invasive electronic cardiometry device on the patient's chest to provide a first electronic signal related to the patient's electrical cardiac activity including a QRS complex; connecting said first electronic signal of such electronic cardiac measuring device to one input of a processing system having a memory; storing in the memory of the processing system a universal intercept-slope tradeoff function of inotropic function based on cardiac function of healthy patients and a universal intercept-slope tradeoff function of lusitropic function based on cardiac function of healthy patients; digitizing the first electronic signal within the processing system; placing a non-invasive mechanical cardiac measurement device on the patient's chest, the non-invasive mechanical cardiac measurement device providing an output related to mechanical cardiac activity; converting the output of the mechanical cardiac function measuring device into a second electronic signal; connecting the converted second electronic signal to a second input of the processing system; digitizing the second electronic signal within the processing system; processing the digitized input to the processing system to determine an inotropic electromechanical time interval of the patient's heart; processing the digitized input to the processing system to determine lusitropic electromechanical time intervals of the patient's heart; determining an inotropic calibration curve of defined intercept and slope for the patient's heart at rest and over various degrees of exercise using the inotropic time intervals and concurrent inotropic myocardial strain rate data obtained from a noninvasive electronic cardiometry device; determining a lusitropic calibration curve for the patient's heart at rest and over various degrees of exercise, with intercepts and slopes defined using the lusitropic time intervals and concurrent lusitropic myocardial strain rate data obtained from a non-invasive electronic cardiometry device; comparing the intercept and the slope of the inotropic calibration curve to the stored universal intercept-slope trade-off function of inotropic function; comparing the intercept and slope of the lusitropic calibration curve to the stored universal intercept-slope trade-off function of lusitropic function; assessing myocardial health or myocardial pathology in the patient using the results of the comparison of inotropic function and the results of the comparison of lusitropic function. A method consisting of:

7. The method of claim 6 , wherein the assessment is the detection and quantification of systolic heart failure.

8. The method of claim 7, wherein the patient is not symptomatic.

9. The method of claim 6 , wherein the assessment is the detection and quantification of diastolic heart failure.

10. 10. The method of claim 9, wherein the patient is not symptomatic.

11. 1. A method for determining a universal intercept-slope trade-off function of inotropic function, comprising: selecting a large number of cardiac healthy patients of various heights, weights, ages, and sexes; measuring inotropic strain rate and inotropic electromechanical intervals for each of said healthy patients using a non-invasive electronic cardiac function measuring device at rest and over a range of exercise with an exercise protocol or a catecholamine drug infusion protocol to achieve a series of inotropic steady states with different degrees of inotropy; plotting the natural logarithm of the absolute value of the inotropic strain rate against the reciprocal of said electromechanical interval at rest and at each inotropic steady state; As a result, the steady state appears as a point of a linear function in {1 / (E-M)ino,ln(abs(Inotropic Strain Rate))} space, with a well-defined slope and intercept for slopes > 0, and the function is defined as the inotropic calibration curve for the patient; writing the fitting parameters of the linear function as (slope, intercept) in {slope, intercept} space as (x, y) points; graphing the (slope, intercept) for each patient in {slope, intercept} space as a linear function of downward slope, and the resulting slope and intercept of the universal inotropy fitting parameters are graphed as a universal inotropy intercept-slope trade-off function; whereby deviations from said function in an individual being evaluated for cardiac disease are indicative of inotropic heart failure.

12. 1. A method for determining a universal intercept-slope trade-off function of lusitropic function, comprising: selecting a large number of cardiac healthy patients of various heights, weights, ages, and sexes; measuring, for each said healthy patient, lusitropic strain rate and lusitropic electromechanical intervals at rest and over a range of exercise with an exercise protocol or a catecholamine drug infusion protocol using a non-invasive electronic cardiac function measuring device to achieve a series of lusitropic steady states with different degrees of lusitropy; plotting the natural logarithm of the absolute value of the luciferase strain rate against the reciprocal of the electromechanical spacing at rest and at each luciferase steady state; As a result, the steady state appears as a point of a linear function in {1 / (E-M)ino,ln(abs(lusitropic Strain Rate))} space, with a well-defined slope and intercept for slopes > 0, and the function is defined as the patient's lusitropic calibration curve; writing the fitting parameters of the linear function as (slope, intercept) in {slope, intercept} space as (x, y) points; graphing the (slope, intercept) for each patient in {slope, intercept} space as a linear function of downward slope, and the resulting slope and intercept of the universal lusitropy fitting parameters are graphed as a universal lusitropy intercept-slope trade-off function; whereby deviations from said function in an individual being evaluated for cardiac disease are indicative of lusitropic heart failure.

13. A non-transitory computer readable medium having stored thereon a universal trade-off function for inotropic functions.

14. A non-transitory computer-readable medium storing a universal trade-off function for luciferase-relaxation functions.

Images